Professional Documents
Culture Documents
1. Introduction
1.1 Equation (a) of the problem statement is used to solve for h as & Q h= (a) A(T T ) The Principle of Dimensional Homogeneity is used to determine the dimensions of the heat transfer coefficient. Using the F-L-T system dimensions of the quantities in Equation (a) are & F L Q = (b) T [A] = L2 (b) [T T ] = [] (c) Thus from Equations (a)-(d) the dimensions of the heat transfer coefficient are [h] = F L 2 T L F (d) = T L N while Possible units for the heat transfer coefficient using the SI system are msK
[]
[ ]
possible units using the English system are 1.2 The Reynolds number is defined as
lb . ft s R
Re =
VD
(a)
The dimensions of the quantities on the left-hand side of Equation (a) are obtained using Table 1.2 as (b) [ ] = M3 L [V] = L (c) T [D] = [L] (d)
M L T Substituting Equations (b)-(e) in Equation (a) leads to
[ ] =
(e)
Chapter 1 M L 3 T L [Re] = L M LT M L3 T = 3 M L T = [1] Equation (f) shows that the Reynolds number is dimensionless. 1.3 The capacitance of a capacitor is defined by
(f)
i (a) dv dt The dimension of i is that of electric current, which is a basic dimension. The dimensions of electric potential are obtained from Table 1.2 as [v] = F L (b) iT Thus the dimensions of the time rate of change of electric potential are dv F L (c) dt = i T 2 Use of Equation (c) in Equation (a) leads to i [C ] = F L i T2 C= i2 T2 = FL (d)
1.4 (a) The natural frequency of a mass-spring system is k n = (a) m where m is mass with dimension [M] and k is stiffness with dimensions in the M-L-T M system of 2 . Thus the dimensions of natural frequency are T
Chapter 1
1 M 2 2 [ n ] = T M 1 (b) = T (b) The natural frequency of the system is 100 Hz, which for calculations must be converted to r/s, cycles n = 20 s r cycles = 20 2 cycles s
r s
(c)
(d)
2
r k = (0.1 kg )125.7 s = 1.58 x10 3 1.5 (a) The mass of the carbon nanotube is calculated as m = AL = r 2 L
m - 23 3.78x10 kg (b) Conversion between TPa and psi leads to N E = 1.1 TPa = 1.1x1012 2 m
3
N m
(e)
m 80 x10 9 m
)(
2
N lb 1 m = 1.1x1012 2 0.225 N 3.28 ft m lb = 1.60 x10 8 2 in (c) Calculation of the natural frequency leads to
1 ft 12 in
Chapter 1
EI AL4
= 22.37
4 12 N 0.34 x10 9 m 1.1x10 2 m 4 = 22.37 kg 9 2 80 x10 9 m 1300 3 0.34 x10 m r = 1.73x1010 s Converting to Hz gives r 1 cycle = 1.73 x1010 s 2 r = 2.75 x10 9 Hz
)(
900 kW hr 24 hr = 37.5 kW The power is converted to English units using the conversions of Table 1.1 P = 37.5 x10 3 W Nm = 37.5x10 3 s 0.225 lb 3.28 ft N m N m 3 = 37.5 x10 s ft lb = 2.77 x10 4 s Conversion to horsepower leads to 1 hp ft lb P = 2.77 x10 4 s 550 ft lb s = 50.3 hp P=
1.7 The conversion of density from English units to SI units is
(a)
(b)
(c)
Chapter 1
= 1.94
= 1.94
= 9.99 x10 3
(a)
m s2
(a)
The velocity is obtained using Equation (a) as v (t ) = 6t + C (b) The constant of integration is evaluated by requiring km v(t = 0) = 180 hr km 1000 m 1 hr = 180 hr km 3600 s m = 50 (c) s Using Equation (c) in Equation (b) leads to m v (t ) = 6t + 50 (d) s The train stops when its velocity is zero, 0 = 6t + 50 t = 8.33 s (e) The distance traveled is obtained by integrating Equation (d) and assuming x(0)=0, leading to (f) x(t ) = 3t 2 + 50t The distance traveled before the train stops is x(8.33) = 3(8.33) 2 + 50(8.33) (g) = 208.3 m 1.9 The differential equation for the angular velocity of a shaft is d + ct = T (a) J dt Each term in Equation (a) has the same dimensions, those of torque or [F L] . The 1 dimensions of angular velocity are . Thus the dimensions of ct are T
Chapter 1
F L [ct ] = 1 T
= [F L T ]
(b)
1.10 The equation for the torque applied to the armature is T = K a ia i f Equation (a) is rearranged as T Ka = ia i f
(a)
(b)
The dimensions of torque are [F L] thus the dimensions of the constant are [K a ] = F 2L i The equation for the back emf is v = K vi f Equation (d) is rearranged as v Kv = if
(c)
(d)
(e)
F L 1 The dimensions of voltage are and the dimensions of angular velocity are T . iT The dimensions of the constant K v are
FL T [K v ] = i 1 i T F L (f) = 2 i It is clear from Equations (c) and (f) that the dimensions of [K a ] and [K v ] are the same. These dimensions are the same as those of inductance (Table 1.2). & 1.11 (a) The dimensions of Q are determined from Equation (a) & Q = A (T 4 Tb4 )
F L 2 4 F L L2 T 4 L = T (b) The differential equations governing the temperature in the body is dT 4 c + (T 4 Tb ) = 0 dt The perturbation in temperature in the radiating body is defined by
(a)
[ ][ ]
(b)
(c)
Chapter 1 Tb = Tbs + Tb1 This leads to a perturbation in the temperature of the receiving body defined as T = Ts + T1 Substitution of equations (d) and (e) in Equation (c) leads to d c (Ts + T1 ) + (Ts + T1 )4 (Tbs + Tb1 )4 = 0 dt Simplifying Equation (f) gives 4 T 4 Tb1 dT1 4 4 1 c + Ts 1 + Tbs 1 + T =0 dt Ts bs Expanding the nonlinear terms, keeping only through the linear terms and noting Ts = Tbs (d) (e)
(f)
(g)
that
c c
1.12 The differential equation is linearized by using the small angle assumption which implies sin and cos 1 . Using these approximations in the differential equation leads to the linearized approximation as 1 2& 1 2 & 2 mL & cL + kL = 0 + (a) 3 4 1.13 The differential equation is linearized by using the small angle assumption which implies sin and cos 1 . Using these approximations in the differential equation leads to the linearized approximation as L 1 2& & & + mL & mg + & = L& y x (a) 3 2 1.14 The nonlinear differential equations governing the concentration of the reactant and temperature are dC A V + (q + Ve E /( RT ) )C A = qC Ai (a) dt dT & qc p Ti qc p T Q + Ve E /( RT ) C A = Vc p (b) dt The reactor is operating at a steady-state when a perturbation in flow rate occurs according to q = q s + q p (t ) (c) The flow rate perturbation induces perturbations in concentration and temperature according to
Chapter 1
C A = C As + C Ap (t )
(d)
T = Ts + T p (t ) (e) The steady-state conditions are defined by setting time derivatives to zero in Equation (a) leading to
(q
(f)
As
=0
(g)
Substitution of Equations (d) and (e) into Equations (a) and (b) leads to dC Ap E / [R (Ts +T p ) ] (C As + C Ap ) = (q s + q p )C Ai V + q s + q p + Ve dt
(h )
& (q s + q p )c p Ti (q s + q p )c p (Ts + T p ) Q + Ve
E / R (Ts +T p )
(C
As
+ C Ap = Vc p
dT p dt
(i)
It is noted from Equation (f) of Example (1.6) that a linearization of the exponential terms in Equations (h) and (i) is
E RTs e =e + e Tp RTs2 Use of Equation (j) in Equations (h) and (i) and rearrangement leads to E E RT dC Ap E RTs s + q s + q p + V e + V e T p (C As + C Ap ) = (q s + q p )C Ai dt RTs2 E E &+ V e RTs + E e RTs T C + C (q s + q p )c p Ti (q s + q p )c p (Ts + T p ) Q p As Ap RTs2
E R (Ts +T p ) E RTs E
(j)
(k)
)
(l)
= Vc p
dT p dt
Equations (g) and (h) are used to simplify Equations (k) and (l) to E E E RT dC Ap RTs V C Ap + V e s T p (C As + C Ap ) + q s C Ap + q p C As + q p C Ap + Ve RTs2 dt = q p C Ai
(m)
q p c p Ti c p (q p Ts + q s T p + q p T p ) + Ve
= Vc p dT p dt
E RTs
C Ap
E E RT e s T p C As + C Ap + V RTs2
)
(n)
Chapter 1
dC Ap dt
E RTs E E RT e s T p C As + V RTs2
+ q s C Ap + q p C Ap + Ve
C Ap
= q p C Ai q p C As
(o)
Vc p
dT p dt
q p c p Ti + c p
(q T
p
+ q sT p
) + Ve
E RTs
C Ap
E E RT + V e s T p C As = 0 (p) RTs2
1.15 The specific heat is related to temperature by c p = A1 + A2T 1.5 + A3T 2.6 The transient temperature is the steady-state temperature plus a perturbation, T = Ts + T p Substituting Equation (b) into Equation (a) leads to 1.5 2.6 c P = A1 + A2 (Ts + T p ) + (Ts + T p )
1.5 2.6 1.5
(a)
(b)
Tp Tp 2.6 (c) = A1 + A2Ts 1 + + A3Ts 1 + T T s s Using the binominal expansion to linearize Equation (c) leads to Tp Tp 1.5 2.6 c p = A1 + A2Ts 1 + 1.5 + A3Ts 1 + 2.6 (d) Ts Ts The differential equation for the time-dependent temperature is 1 dT 1 + T = T (e) cp dt R R Substituting Equations (b) and (d) into Equation (e) along with T = Ts + Tp leads to
T d 1 1 + A3Ts 2.6 1 + 2.6 p (Ts + T p ) + (Ts + T p ) = (Ts + Tp ) (f) dt Ts R R dTs = 0 and Ts = Ts reduces Equation (f) to Noting that the steady-state is defined by dt Tp T p dT p 1 1 1 .5 2 .6 (g) A1 + A2Ts 1 + 1.5 + A3Ts 1 + 2.6 dt + R T p = R Tp Ts Ts Tp 1 .5 A1 + A2Ts 1 + 1.5 Ts
Terms such as T p
dT p dt
Tp Ts
<< 1
(h)
(A + A T
1
1.5 2 s
+ A3Ts2.6 )
dT p dt
1 1 T p = Tp R R
1.16 The force acting on the piston at any instant is F = pA where A is the area of the piston head. The pressure is related to the density by p = C
(a)
(b)
Chapter 1 The mass of air in the cylinder is constant and is calculated when the piston is in equilibrium as m = 0 Ah (c) where 0 is the density of the air in equilibrium. Using Equation (b) in Equation (c) leads to
p (d) m = 0 Ah C where p 0 is the pressure in the cylinder when the piston is in equilibrium. At any instant the mass is calculated as m = A(h x)
1
p = A(h x) C Since the mass is constant, Equations (d) and (e) are equated leading to
h p = p0 h x Substitution of Equation (f) into Equation (a) leads to h F = p 0 A h x
(e)
(f)
(g)
x < 1 a binomial expansion can be used on the right-hand side of Equation (h). h Using the binomial expansion keeping only through the linear term leads to p A F = p0 A + 0 x (e) h The linear stiffness is obtained from Equation (e) as p A (f) k= 0 h
Since
10
Chapter 1 1.17 The appropriate superposition of the voltage in Figure P1.17 is illustrated below
The mathematical representation of the voltage source is v(t ) = 12[u (t ) u (t 2)] + (24 6t )[u (t 2) u (t 4)] 1.18 The superposition of the force of Figure P1.18 is illustrated below.
The mathematical representation of the force is F (t ) = 300[u (t ) u (t 15)] 300[u (t 15) u (t 30)]
(a)
11
Chapter 1 1.19 The superposition of the cam displacement over one period is shown below
(a) The mathematical representation of the displacement over one period is x (t ) = 0.04[u (t ) u (t 0.05)] + 0.002[u (t 0.05) u (t 0.25) + (0.012 0.04t )[u (t 0.25) u (t 0.3)] (a) (b) The period of the cycle is 0.5 s. Thus the displacement over the second period is obtained by replacing t by t+0.5 in Equation (a). The displacement over the kth period is obtained by replacing t by t+(k-1)(0.5) in Equation (a). The total displacement is obtained by summing over all periods
x(t ) = {0.04[u (t .5k + .5) u (t .5k + .45)] + .002[u (t 0.5k + .45) u (t .5k + .25)]
k =1 K
(b)
where K is the smallest integer greater than t/(0.05). 1.20 Integration of Newtons second law with respect to time leads to the principle of impulse and momentum I = Fdt = m(v 2 v1 )
t1 t2 t2
Fdt .
t1
applied instantaneously to the 4-kg particle when it is at rest. Application of the principle of impulse and momentum leads to
12
Chapter 1 12 N s = (4 kg )v 2 v2 = 12 N s 4 kg m =3 s
(b)
1.21 The equation for the voltage drop across an inductor is di v=L dt Integration of Equation (a) with respect to time leads to
(a)
vdt = L(i
0
i1 )
(b)
vdt
0
L 20 V s = 0.4 H = 50 A
(c)
1.22 The mathematical representation of the force is F (t ) = 100 (t ) + 150 (t 2.5) + 50 (t 3.8)
(a)
1.23 The MATLAB file Problem1_23 which determines the steady-state response of a series LRC circuit is listed below % Problem1_23.m % Steady-state response of seties LRC circuit clear disp('Steady-state response of series LRC circuit') % Input parameters disp('Input resistance in ohms') R=input('>> ') disp('Input capacitance in farads') C=input('>>') disp('Input inductance in henrys') L=input('>> ') disp('Input source frequency in r/s') om=input('>> ') disp('Input source amplitude in V') V0=input('>> ') % Calculates parameters disp('Natural freqeuncy in r/s =') 13
Chapter 1 omn=1/(L*C)^0.5 disp('Dimensionless damping ratio =') zeta=R/2*(C/L)^0.5 disp('Phase angle in rad=') C1=om^2-omn^2; C2=2*zeta*om*omn; phi=atan2(C1,C2) disp('Steady-state amplitude in A =') C3=V0*om/L; C4=1/(C1^2+C2^2)^0.5; I=C3*C4 tf=10*pi/om; dt=tf/200; for k=1:201 t(k)=(k-1)*dt; i(k)=I*sin(om*t(k)+phi); end plot(t,i) xlabel('t (s)') ylabel('i (A)') title('Steady-state response of series LRC circuit') str1=['R=',num2str(R),' \Omega']; str2=['C=',num2str(C),' F']; str3=['L=',num2str(L),' H']; str4=['\omega=',num2str(om),' r/s']; str5=['V_0=',num2str(V0),' V']; text(0.9*tf,I,str1) text(0.9*tf,0.8*I,str2) text(0.9*tf,0.6*I,str3) text(0.9*tf,0.4*I,str4) text(0.9*tf,0.2*I,str5) The MATLAB workspace from a sample execution of Problem1_23.m is >> Problem1_23 Steady-state response of series LRC circuit Input resistance in ohms >> 100 R= 100 Input capacitance in farads >>0.2e-6
14
Chapter 1 C= 2.0000e-007 Input inductance in henrys >> 0.5 L= 0.5000 Input source frequency in r/s >> 2000 om = 2000 Input source amplitude in V >> 120 V0 = 120 Natural freqeuncy in r/s = omn = 3.1623e+003 Dimensionless damping ratio = zeta = 0.0316 Phase angle in rad= phi = -1.5042 Steady-state amplitude in A = I=
15
Chapter 1
1.24 The MATALB file Prolbem1_24.m is listed below % Problem1_24.m %(a) Input two five by five matrices disp('Please input matrix A by row') for i=1:5 for j=1:5 str={['Enter A(',num2str(i),num2str(j),')']}; disp(str) A(i,j)=input('>> '); end end disp('Please input matrix B by row') for i=1:5 for j=1:5 str={['Enter B(',num2str(i),num2str(j),')']}; disp(str) B(i,j)=input('>> '); end end A
16
Chapter 1 B % (b) =A+B C=A+B % (c) D=A*B D=A*B % (d) det(A) detA=det(A) % eigenvalues and eigenvectors of A [x,Y]=eigs(A); disp('Eigenvalues of A') Y disp('Matrix of eigenvalues of A') x A sample output from execution of the file is shown below >> clear >> Problem1_24 Please input matrix A by row 'Enter A(11)' >> 1 'Enter A(12)' >> 0 'Enter A(13)' >> 12 'Enter A(14)' >> -1 'Enter A(15)' >> 21 'Enter A(21)' >> 14 'Enter A(22)' >> -3 'Enter A(23)' >> 2 'Enter A(24)' >> 0
17
Chapter 1 'Enter A(25)' >> -22 'Enter A(31)' >> 11 'Enter A(32)' >> 12 'Enter A(33)' >> 10 'Enter A(34)' >> -4 'Enter A(35)' >> 12 'Enter A(41)' >> 10 'Enter A(42)' >> 11 'Enter A(43)' >> 18 'Enter A(44)' >> 12 'Enter A(45)' >> 21 'Enter A(51)' >> 10 'Enter A(52)' >> 11 'Enter A(53)' >> 31 'Enter A(54)' >> 21 'Enter A(55)'
18
Chapter 1
>> 11 Please input matrix B by row 'Enter B(11)' >> 21 'Enter B(12)' >> -21 'Enter B(13)' >> 21 'Enter B(14)' >> 10 'Enter B(15)' >> 9 'Enter B(21)' >> 8 'Enter B(22)' >> 2 'Enter B(23)' >> 2 'Enter B(24)' >> 4 'Enter B(25)' >> -5 'Enter B(31)' >> 16 'Enter B(32)' >> 12 'Enter B(33)' >> 11 'Enter B(34)' >> 18 'Enter B(35)'
19
Chapter 1
>> 11 'Enter B(41)' >> 21 'Enter B(42)' >> 32 'Enter B(43)' >> 14 'Enter B(44)' >> 19 'Enter B(45)' >> 12 'Enter B(51)' >> 12 'Enter B(52)' >> 9 'Enter B(53)' >> -5 'Enter B(54)' >> 13 'Enter B(55)' >> 21 A= 1 14 11 10 10 0 -3 12 11 11 12 2 10 18 31 -1 21 0 -22 -4 12 12 21 21 11
20
Chapter 1 B= 21 8 16 21 12 -21 2 12 32 9 21 2 11 14 -5 10 9 4 -5 18 11 19 12 13 21
C= 22 22 27 31 22 -21 -1 24 43 20 33 4 21 32 26 9 30 4 -27 14 23 31 33 34 32
D= 444 38 547 1090 1367 280 -474 -107 601 955 34 420 249 493 812 480 -122 418 969 1244 570 -299 353 818 859
detA = -1171825 Eigenvalues of A Y= 43.3949 0 0 0 0 0 -18.9247 0 0 0 0 0 0 0 0 0 -1.8896 -11.6121i 0 0 0 -1.8896 +11.6121i 0 0 0 10.3091
21
Chapter 1 0.1869 -0.2133 -0.6060 -0.6466 0.7510 -0.4463 -0.2256 0.3991 0.7187 0.7187 -0.4106 0.0528 + 0.2510i 0.0528 - 0.2510i -0.4042 -0.0845 - 0.0753i -0.0845 + 0.0753i 0.6726 -0.2465 + 0.1043i -0.2465 - 0.1043i 0.3808
1.25 A MATLAB file to calculate and plot ( r , ) is given below % Plots the function LAMBDA(r,zeta) as a function of r for several values of % zeta % Specify four values of zeta zeta1=0.1; zeta2=0.4; zeta3=0.8; zeta4=1.5; % Define values of r for calculations for i=1:400 r(i)=(i-1)*.01; % Calculate function LAMBDA1(i)=r(i)^2/((1-r(i)^2)^2+(2*zeta1*r(i))^2)^0.5; LAMBDA2(i)=r(i)^2/((1-r(i)^2)^2+(2*zeta2*r(i))^2)^0.5; LAMBDA3(i)=r(i)^2/((1-r(i)^2)^2+(2*zeta3*r(i))^2)^0.5; LAMBDA4(i)=r(i)^2/((1-r(i)^2)^2+(2*zeta4*r(i))^2)^0.5; end plot(r,LAMBDA1,'-',r,LAMBDA2,'.',r,LAMBDA3,'-.',r,LAMBDA4,'--') xlabel('r') ylabel('\Lambda') str1=['\zeta=',num2str(zeta1)]; str2=['\zeta=',num2str(zeta2)]; str3=['\zeta=',num2str(zeta3)]; str4=['\zeta=',num2str(zeta4)]; legend(str1,str2,str3,str4) title('\Lambda vs. r') The resulting output from execution of the .m file is the following plot
22
Chapter 1
1.26 The MATALB .m file Problem1_26 which determines and plots the step response of an underdamped mechanical system is shown below. % Problem1_26.m % Step response of an underdamped mechanical system % Input natural frequency and damping ratio clear disp('Step response of underdamped mechanical system') disp('Please input natural frequency in r/s') om=input('>> ') disp('Please input the dimensionless damping ratio') zeta=input('>> ') % Damped natural frequency omd=om*(1-zeta^2)^0.5; C1=zeta*om/omd; C2=1/om^2; C3=zeta*om; tf=10*pi/omd; dt=tf/500; for i=1:501 t(i)=(i-1)*dt; x(i)=C2*(1-exp(-C3*t(i))*(C1*sin(omd*t(i))-cos(omd*t(i)))); end plot(t,x) xlabel('t (s)') ylabel('x (m)')
23
Chapter 1 str1=['Step response of underdamped mechancial system with \omega_n=',num2str(om),'and \zeta=',num2str(zeta)] title(str1) str2=['x(t)=',num2str(C2),'[1-e^-^',num2str(C3),'^t(',num2str(C1),'sin(',num2str(omd),'t)cos(',num2str(omd),'t))]'] text(tf/4,C2/2,str2) Output from execution of Problem1_26 follows >> Problem1_26 Step response of underdamped mechanical system Please input natural frequency in r/s >> 100 om = 100 Please input the dimensionless damping ratio >> 0.1 zeta = 0.1000
24
Chapter 1 h(t ) = qR(1 e t /( RA) ) (a) (a) Since the argument of a transcendental function must be dimensionless the dimensions of the product of resistance and area must be time. Thus the dimensions of T resistance must be 2 L (b) Note that the steady-state value of the liquid-level perturbation is qR . The MATLAB file Problem1_27.m which calculates and plots h(t) from t=0 until h is within 1 percent of its steady-state value is given below disp('Please enter resistance in s/m^2 ') R=input('>> ') % Final value of h hf=0.99*q*R; dt=0.01*R*A; h1=0; h(1)=0; t(1)=0; i=1; while h1<hf i=i+1; t(i)=t(i-1)+dt; h(i)=q*R*(1-exp(-t(i)/(R*A))); h1=h(i); end plot(t,h) xlabel('t (s)') ylabel('h (m)') title('Perturbation flow rate vs time') str1=['A=',num2str(A),' m^3/s'] str2=['R=',num2str(R),' s/m^2'] str3=['q=',num2str(q),' m^3/s'] text(0.5*t(i),0.5*h(i),str1); text(0.5*t(i),0.4*h(i),str2); text(0.5*t(i),0.3*h(i),str3); Sample output from execution of Problem1_27.m is given below >> Please enter tank area in m^2 >> 100 A= 100 Please enter flow rate in m^3/s 25 1.27 The perturbation in liquid level is
Chapter 1 >> 0.2 q= 0.2000 Please enter resistance in s/m^2 >> 15 R= 15 str1 = A=100 m^3/s str2 = R=15 s/m^2
26
Chapter 2
2. Mechanical Systems
2.1 The mass moment of inertia of the annular cylinder is obtained by subtracting the moment of inertia of the hollow part from the moment of inertia of a solid cylinder. I = Is Ih (a) The mass of a solid cylinder of radius r and length L is m = r 2 L The mass of the solid cylinder is kg 2 ms = 7600 3 (0.4 m ) (1.2 m ) m = 4.58 x103 kg The mass missing from the hollow part of the cylinder is kg 2 m sh = 7600 3 (0 .3 m ) (1 . 2 m ) m = 2 .58 x10 3 kg The total mass of the cylinder is
( )
(b)
(c)
(d)
m = ms mh (e) = 2.01x10 3 kg (a) The mass moment of inertia of a solid cylinder about a centroidal axis parallel to the x axis is obtained using Table 2.1 as 1 I xx = m (3r 2 + L2 ) (f) 12 The moment of inertia of the solid cylinder about the cetnroidal axis is 1 I xxs = 4.58 x10 3 kg 3(0.4 m) 2 + (1.2 m) 2 12 = 733.5 kg m 2 (g) The moment of inertia of the hollow part of the cylinder abut the centroidal x axis is 1 I xxh = 2.58 x10 3 kg 3(0.3 m) 2 + (1.2 m) 2 12 = 367.5 kg m 2 (h) The moment of inertia of the cylinder about the centroidal x axis is I xx = I xxs I xxh
)[
)[
= 366.0 kg m 2 (i) The parallel axis theorem is used to determine the moment of inertia about the x axis as shown in Figure P2.1 is
27
Chapter 2
L = I xx + m 2
2
I xx
= 1.09 x10 3 kg m 2 (j) (b) The mass moment of inertia of a cylinder about a longitudinal axis through its center is 1 I yy = mr 2 (k) 2 The mass moment of inertia of the solid cylinder about the y axis is 1 I yys = 4.58 x10 3 kg (0.4 m) 2 2 = 366.7 kg m 2 (l) The moment of inertia of the hollow part of the cylinder about the y axis is 1 I yyh = 4.58 x10 3 kg (0.3 m) 2 2 = 116.0 kg m 2 (l) The moment of inertia of the annular cylinder about the y axis is I yy = I yys I yyh
= 250.7 kg m 2
(m)
2.2 The total mass moment of inertia about an axis through A is the sum of the moments of inertia of the plate, the bar, and the cylinder, I A = I Ap + I Ab + I Ac (a) (a) The moment of inertia of the thin plate about an axis perpendicular to the page through its centroid is determined using Table 2.1 as 1 I zp = m( w 2 + h 2 ) 12 1 2 = (4.1 kg ) (0.2 m ) + (0.4 m) 2 12 = 6.83 x10 2 kg m 2 (b) The parallel-axis theorem is used to determine its moment of inertia about an axis through A as I zAp = I zp + md 2
(c)
28
Chapter 2
I zb = 1 mL2 12 1 2 = (0.8 kg )(1.1 m ) 12 = 8.07 x10 2 kg m 2
2
(d)
I zAb = I zb + md
= 0.531 kg m 2
]
(f)
I zAc = I zc + md 2
= 0.091 kg m 2 (g) The moment of inertia of the total assembly about an axis perpendicular to the page at A is obtained using Equations (a), (c), (e), and (g) as I zA = 8.10 kg m 2 + (0.531 kg m 2 ) + 0.091 kg m 2
= 8.73 kg m 2 (h) (a) A horizontal axis through A is a centroidal axis for each component. Thus application of the parallel axis theorem is not required. Using Table 2.1 for the plate 1 I xp = mh 2 12 1 2 = (4.1 kg )(0.4 m ) 12 = 0.0547 kg m 2 (i) The moment of inertia of the slender bar about a horizontal axis is approximately zero. For the cylinder 1 I xc = mr 2 2 1 2 = (2.8 kg )(0.1 m ) 2 = 0.014 kg m 2 (j)
Thus the moment of inertia of the assembly about a horizontal axis through A is
29
Chapter 2
) (
) (
)
(k)
2.3 The total moment of inertia of the assembly of Figure P2.3 about an axis through O is the sum of the moment of the inertia of the bar about O and the moments of inertia of the particles about O. I O = I Ob + I Op1 + I Op 2 (a) The moment of inertia of the bar about its own centroidal axis is 1 I b = mL2 12 1 = (20 kg )(4 m) 2 12 = 26.7 kg m 2 (b) The parallel axis theorem is used to calculate the moment of inertia of the bar about O as I Ob = I b + md 2
= 26.7 kg m 2 + (20 kg )(1 m) 2 = 46.7 kg m 2 (c) The moment of inertia of a particle about its own centroidal axis is zero. The parallel axis theorem is used to calculate the moment of inertia of a particle about another axis. For the 5-kg particle I Op1 = (5 kg)(1 m) 2
= 5 kg m 2 The moment of inertia of the 10-kg particle about O is I Op 2 = (10 kg)(3 m) 2 = 90 kg m 2 Combining Equations (a),(b),(c), and (d) leads to I O = 46.7 kg m 2 + 5 kg m 2 + 90 kg m 2 (c)
(d)
= 141.7 kg m 2
2.4 The velocity of the 2-kg particle is given as
(e)
m (a) s (a) Noting that the particle is at the origin at t=0, the position vector of the particle at an arbitrary time t is obtained as
v = [2 sin(3t )i + 4 cos(3t ) j + 2k ]
30
Chapter 2
t
r = vdt
0 t
4 2 = cos(3t )i + sin(3t ) j + 2tk 3 3 t =0 4 2 = [1 cos(3t )]i + sin(3t ) j + 2tk m 3 3 (b) The acceleration of the particle at an arbitrary time is dv a= dt m = [6 cos(3t )i 12 sin(3t ) j] 2 s (a) The velocity of the particle at t=1 is obtained from Equation (a) as m v (t = 1 s) = [2sin(3)i + 4 cos(3) j + 2k ] s m = [0.282i 3.96 j + 2k ] s (b) The magnitude of the velocity at t=1 s is
v(t = 1 s) = (0.282) + ( 3.96) + (2) 2
2 2
(b)
(c)
(d)
1 2
m s
(e)
(f)
(g)
2.5 (a) The velocity of the mass center of the bar at the instant shown is obtained using the relative velocity equation
v G = v O + k rG / O
(a)
31
Chapter 2 where O is located at the pin support. Since O is a fixed axis of rotation v o = a o = 0 . Thus r v G = 0 + 10 k (1.5 m )i s m (b) = 15 j s (b) The acceleration of the mass center is obtained using the relative acceleration equation a G = a O + k rG / O + k (k rG / O )
r r m = 0 + 3 2 k (1.5 m)i + 10 k 15 j s s s m = ( 150 i 4.5 j) 2 s (c) The kinetic energy of the bar at this instant is 1 1 2 T = m v G + I 2 2 2 r 1 1 1 m 2 = (6 kg )15 + (6 kg )(5 m ) 10 2 12 2 s s = 1300 N m (d)The angular momentum of the bar about the centroidal axis is H G = I 1 (6 kg )(5 m) 2 10 r 12 s = 125 N s m =
2 2
(c)
(d)
(e)
2.6 Since the center of the fan is a fixed axis of rotation the kinetic energy of each blade is 1 Tb = I O 2 (a) 2 The parallel axis theorem is used to determine the moment of inertia of each blade about O I O = I + md 2
= 0.4 kg m 2 + (1.2 kg)(0.45 m) 2 = 0.592 kg m 2 Using the result of Equation (b) in Equation (a) leads to (b)
2
1 Tb = 0.592 kg m 2 2
= 202.9 N m The fan consists of four blades, thus the total kinetic energy of the fan is
32
(c)
2.7 (a) The total kinetic energy of the engine is the sum of the kinetic energies of the components T = Tcrsnk + Trod + T piston (a) The crank rotates about a fixed axis at O, thus its kinetic energy is 1 2 Tcrank = I O crank (b) 2 where I O = 0.4 kg m 2 and
crank = 250
= 26.2
r (c) s Substituting these values into Equation (b) leads to 2 1 r 2 Tcrank = (0.5 kg m ) 26.2 2 s = 171.3 N m (d) The connecting rod does not rotate about a fixed axis. Its kinetic energy is calculated by 1 1 Trod = mv 2 + I 2 (e) 2 2 The connecting rod is a slender rod of mass 0.4 kg and length 30 cm. Its centroidal mass moment of inertia is obtained using Table 2.1 as 1 I rod = mL2 12 1 2 = (0.4 kg )(0.3 m ) 12 = 3.0 x10 3 kg m 2 (f) Rigid-body kinematics is used to determine the angular velocity and velocity of the mass center of the connecting rod. The angle made by the connecting rod with the vertical is determined using the law of sines as illustrated below sin(b) sin(10) = 0.3 m 0.1 m sin(b) = 3 sin(10) = 0.5209 b = 31.4 (g) Let A be point at which the crank is pinned to the connecting rod. Application of the relative velocity equation leads to
33
Chapter 2
v A = v O + crank k rA / O r = 0 + 26.2 k (0.1 m )[ sin(31.4)i + cos(31.4) j] s m = (2.23i + 1.36 j) (h) s The piston, P, is constrained to move in the y direction. Application of the relative velocity equation between A and P on the connecting rod gives v P j = v A + rod k rP / A
m + rod k (0.3 m )[sin(10)i + cos(10) j] s = (2.23 - 0.295 rod )i + (1.36 + 0.0521 rod ) j Equating the x components of the vectors on both sides of Equation (i) leads to 0 = 2.23 0.295 rod = (2.23i + 1.36 j) r s Equating the y components of the vectors on both sides of Equation (i) leads to v P = 1.36 + 0.0521 rod
(i)
rod = 7.56
(j)
m (k) s The velocity of the mass center of the connecting rod is calculated using the relative velocity equation on the connecting rod between A and the mass center v rod = v A + rod k rG / A = 1.76
m r + 7.56 k (0.15 m )[sin(10)i + cos(10) j] s s m = (1.18i + 1.56 j) s The magnitude of the velocity of the mass center of the connecting rod is = (2.23i + 1.36 j) m m v rod = 1.18 + 1.56 s s m = 1.92 s The use of Equations (f), (j), and (m) in Equation (e) gives 2 2 1 m 1 r 3 2 Trod = (0.4 kg )1.92 + (3.0 x10 kg m ) 7.56 2 s 2 s = 0.823 N m The kinetic energy of the piston is
2 2
(l)
(m)
(n)
34
Chapter 2
T piston = = 1 2 mv P 2 1 (0.3 kg )1.76 m 2 s
2
= 0.465 kg m 2 (o) The total kinetic energy of the engine is calculated by substituting Equations (d), (n) and (o) in Equation (a) leading to T = 171.3 N m + 0.823 N m + 0.465 N m = 172.6 N m (p) (b) The acceleration of the piston is obtained using the relative acceleration equations. The crank rotates with a constant angular velocity, thus its angular acceleration is zero. Application of the relative acceleration equation between O and A on the crank leads to 2 a A = a O + crank k rA / O crank rA / O
r = 0 + 0 26.2 (0.1 m )[ sin(31.4)i + cos(31.4) j] s m (q) = (35.7 58.5 j) 2 s Application of the relative acceleration equation on the connecting rod between A and P leads to 2 a P j = a A + rod k rP / A rod rP / A
2
= (35.7i 58.5 j)
r 7.56 (0.3 m )[sin(10)i + cos(10) j] s = (32.7 0.295 rod )i + ( 75.4 + 0.0521 rod )j Equating x components of the vectors on both sides of Equation (r) leads to 32.7 0.295 rod = 0 r s2 Equating the y components of the vectors on both sides of Equation (r) leads to a P = 75.4 + (0.0521)(110.7 )
(r)
rod = 110.7
(s)
= 69.6
m s2
(t)
2.8 The total kinetic energy of the system is the sum of the kinetic energy of the three system components (a) T = T1 + T2 + T3 where T1 is the kinetic energy of the 5-kg block, T2 is the kinetic energy of the disk and T3 is the kinetic energy of the 10-kg block.
35
Chapter 2 (a) At an arbitrary instant the kinetic energy of the 5-kg block is 1 & T1 = m1 x 2 2 1 & = (5 kg) x 2 2 & (b) = 2.5 x 2 The kinetic energy of the disk is 1 & T2 = I 2 (c) 2 & where is the angular velocity of the disk. The velocity of the particle on the disk where the cable which connects the 5-kg block is attached must be the same as the velocity of the cable and thus the 5-kg block. Thus & (0.1 m )&= x & & x = 10 x & = (d) 0.1 Substitution of Equation (d) into Equation (c) leads to 1 &2 T2 = (0.2)(10 x ) 2 & = 10 x 2 (e) Letting y be the upward displacement of the 10-kg block, its kinetic energy is 1 & T3 = m3 y 2 (f) 2 Using the same kinematic arguments as above & & y = (0.15 m) & = (0.15)(10 x) & = 1.5 x (g) Substitution of Equation (g) into Equation (f) leads to 1 &2 T3 = (10 kg )(1.5 x ) 2 & = 11.25 x 2 (h) Thus the total kinetic energy of the system is & & & T = 2.5 x 2 + 10 x 2 + 11.25 x 2
& = 23.75 x 2 (b) Substitution of Equation (d) Into Equation (i) leads to
(i)
(j)
(a)
36
Chapter 2 The angular velocity of shaft AB of 250 rpm is converted to r/s as rev 2 r 1 min AB = 250 min rev 60 s = 26.2 The angular velocities of meshing gears are related by n11 = n 2 2
(b)
n1 (c) n2 where n represents the number of teeth of a gear. The angular velocities of the shafts connected by gears to shaft AB are calculated using Equation (c). The angular velocity of the shaft with the 25 tooth gear is r 10 1 = 26.2 s 25 r = 10.47 (d) s The angular velocity of the remaining shaft is r 30 3 = 26.2 s 10 r = 78.5 (e) s The total kinetic energy of the system is 2 r 1 T = 10 kg m 2 + 2.1 kg m 2 10.47 s 2
2 = 1
= 2.67 x10 4 J 2.10 The parallel combination of springs is replaced by a spring of stiffness N N N k1 = 2 x10 5 + 1x10 5 = 3 x10 5 m m m The two series combinations are replaced by springs of stiffnesses 1 N k2 = = 1.71x10 5 1 1 m + N N 3 x10 5 4 x10 5 m m
(f)
37
N N 8 x10 5 m m The remaining springs are in parallel leading to an equivalent stiffness of k eq = k 2 + k 3 N N = 1.71x10 5 + 3.08 x10 5 m m N = 4.79 x10 5 m 2.11 Define as the clockwise angular displacement of the bar, measured from its equilibrium position. Let k t ,eq be the torsional stiffness of a torsional spring attached to a particle on the bar such that the potential energy of the torsional spring is equal to the potential energy of the system of springs at any instant assuming small . The potential energy of the equivalent torsional spring at any instant is 1 V = k t ,eq 2 (a) 2 The total potential energy of the existing springs is 1 1 1 2 2 V = (4000 )(3 ) + (3000 )(2 ) + (1800 ) 2 2 2 2 1 = (4.98 x10 4 ) 2 (b) 2 Equating the potential energies of Equations (a) and (b) leads to Nm k t ,eq = 4.98 x10 4 (c) r 2.12 Let be the clockwise angular displacement of bar AB and let be the counterclockwise angular displacement of bar CD. It is desired to replace the springs by a single spring of an equivalent stiffness located at A as illustrated.
38
Chapter 2
The potential energy of this spring, for an arbitrary is 1 2 V = k eq (0.1 ) (a) 2 The potential energy of the springs as illustrated in Figure P2.12 is N 1 N 1 2 2 V = 5000 (0.3 ) + 8000 (1.0 ) (b) m 2 m 2 The two bars are connected by a rigid massless link. The displacements of the particles at each end of the link are equal, 0.8 = 0.9 8 = (c) 9 Substitution of Equation (c) into Equation (b) gives 2 1 8 V = 5000(0.3) 2 + 8000 2 2 9 1 = 1.13x10 4 2 (d) 2 Equating the left-hand sides of Equations (a) and (d) leads to 1.13 x10 4 = .01k eq
k eq = 1.13 x10 6
N m
(e)
Let x1 represent the change in length of the spring of stiffness k1 and let x 2 represent the change in length of the spring of stiffness k 2 . The total potential energy of the two springs is 1 1 2 V = k1 x12 + k 2 x 2 (a) 2 2 The displacement of the block, x, is the sum of the changes in length of the springs x = x1 + x2 (b)
39
k1 x1 = k 2 x2
Equation (c) is rearranged as
(c)
(d)
k1 x1 k2 Substitution of Equation (d) into Equation (b) leads to k x = 1 + 1 x1 k 2 Equation (e) is rearranged resulting in k2 x1 = x k1 + k 2 Equation (f) is used in Equation (f) giving k1 x2 = x k1 + k 2 Substitution of Equations (f) and (g) in Equation (a) leads to x2 =
1 k1 x + k2 2 k1 + k 2 2 2 1 k k + k 2 k1 2 = 1 2 x 2 (k1 + k 2 )2 1 k k (k + k ) = 1 2 1 22 x 2 2 ( k1 + k 2 ) 1 k k = 1 2 x 2 2 k1 + k 2 The equivalent stiffness is obtained from Equation (h) as 1 k2 V = k1 2 k1 + k 2
2
(e)
(f)
(g)
x
2
(h)
keq =
k1k2 k1 + k2
(i)
2.14 The spring of stiffness k1 is in series with the parallel combination of the spring of stiffness k 2 and the viscous damper. The force in each series component is the same as F, as
illustrated below
40
Chapter 2
Let x1 represent the change in length of the spring of stiffness k1 and let x 2 represent the change in length of the parallel combination. Thus F = k1 x1 (a) and & F = k 2 x2 + cx2 (b) The change in length of series components is the sum of the changes in lengths of the individual components x = x1 + x 2 (c) Equating the force in Equations (a) and (b) leads to & k1 x1 = k 2 x 2 + cx2 (d) Solving Equation (c) for x 2 and substituting into Equation (d) leads to & & k1 x1 = k 2 ( x x1 ) + c( x x1 ) & & (k1 + k 2 ) x1 + cx1 = k 2 x + cx (e) F Noting from Equation (a) that x1 = , Equation (e) is rewritten as k1 & & (k1 + k 2 ) F + c F = k 2 x + cx k1 k1 k2 1 + k1 c & F + F = k 2 x + cx & k1 (f)
41
Chapter 2
F = ma
F + P = ma Summing moments about the mass center of the disk leads to M G = I
Fr = 1 2 mr 2
(a)
(b)
a = r (c) Substituting Equations (b) and (c) into Equation (a) leads to 1 mr + P = mr 2 3 mr = P 2 2P = (d) 3mr Substitution of Equation (d) into Equation (b) gives 1 2P F = mr 2 3mr 2 (e) = P 3 (a) For the assumption of no slip to be valid, F < mg . Thus the criterion for no slip is 2 P < mg 3 3 P < mg (f) 2 Substitution of given values into Equation (f) leads to the largest value of P such that the disk rolls without slip as 3 m Pmax = (0.25)(1.8 kg) 9.81 2 2 s = 4.41 N (g) (b) The angular acceleration of the disk is calculated using Equation (d) as 2P = 3mr 2(4.41 N) = 3(1.8 kg)(0.2 m)
= 8.18
r s2
(h)
42
Chapter 2 2.16 Let be the counterclockwise angular rotation of the disk measured from the systems equilibrium position. The work done by a viscous damper of damping coefficient ceq attached to the block of mass m A is & W = c r d (r )
( ) & = (c r )d
eq 2 eq
(a)
It is shown in Example 2.4 that the two viscous dampers in series can be replaced by a single viscous damper of damping coefficient cc (b) c4 = 1 2 c1 + c 2 The work done by this equivalent viscous damper as the disk rotates through an angle is 3 & 3 W = c 4 r d r 2 2 9 & (c) = c 4 r 2 d 4 The work done by the viscous damper of damping coefficient c3 as the disk rotates through an angle is 3 & 3 W = c 3 r d r 2 2 9 & = c3 r 2 d (d) 4 The total work done by the attached viscous dampers is obtained by adding Equations (c) and (d) 9 cc & W = c3 + 1 2 r 2 d (e) c1 + c 2 4 The equivalent viscous damping coefficient is obtained by requiring that the work done by the equivalent viscous damper, Equation (a) is equal to the work done by that attached & viscous dampers, Equation (e), for all and . Thus cc 9 (f) ceq = c3 + 1 2 4 c1 + c 2 2.17 A free-body diagram of the block is illustrated below.
43
Chapter 2 (a) In order to initiate motion the applied force P must be greater than the friction force developed using the static coefficient of friction, P > s mg m = (0.30)(45 kg) 9.81 2 s = 132.4 N When motion occurs the friction force is F = k mg Application of Newtons second law to the free-body diagram leads to P = F = ma The acceleration of the particle is obtained from Equation (c) as 1 a = (P k mg ) m 1 m = 132.4 N (0.27)(45 kg) 9.81 s 2 45 kg = 0.294 m s2
(a) (b)
(c)
(d)
2.18 The solution follows that of Example 2.6, but with n1 =3 (a) n2 (a) The power delivered to the system is the same as calculated in Example 2.6 P = 209.4 kW (b) (b) Use of Equation (c) of Example 2.6 with = 1000 rpm gives CD = 3000 rpm . Then from Equation (g) of Example 2.6 TCD =
AB T AB CD
1000 rpm = 3000 rpm (2000 N m ) = 666.7 N m 2.19 The force of Figure P2.19 has the following superposition
(c)
44
Chapter 2
The mathematical representation of the force is F (t ) = 20tu (t ) 20tu (t 1) + ( 40 20t )u (t 1) ( 40 20t )u (t 2) = 20tu (t ) + ( 40 40t )u (t 1) ( 40 20t )u (t 2) (a) Assuming that F (t ) is the only force applied to the particle, application of Newtons second law to the particle leads to Noting that a =
t
dv dt dt = [20tu (t ) + (40 40t )u(t 1) (40 20t )u(t 2)]dt 0 0 Noting that
(b)
(c)
t t =1
u (t 2) 40t 10t 2
2
t t =2
v(t ) = 2.5 + 10t u (t ) + ( 40t 20t 20)u (t 1) ( 40t 10t 40)u (t 2) (a) Evaluation of Equation (a) for t=0.5 s gives v(0.5) = 2.5 + 10(0.5) 2
(d)
= 5.0
m s
(e)
(b) Evaluation of Equation (d) at t=2.5 s gives v(2.5) = 2.5 + 10(2.5) 2 + 40(2.5) 20(2.5) 2 20 40(2.5) 10(2.5) 2 40
]
(f)
= 22.5
m s
2.20 The superposition of the displacement over one period is illustrated below
45
Chapter 2
(a) The mathematical representation of the displacement over one period is x(t ) = 0.02tu (t ) 0.02tu (t 0.1) + 0.002u (t 0.1) 0.002u (t 0.3) + (0.008 0.02t )u (t 0.3) (0.008 0.2t )u (t 0.4) = 0.02tu (t ) + (0.002 0.02t )u (t 0.1) + (0.006 0.02t )u (t 0.3) (0.008 0.02t )u (t 0.4) (b) The velocity over one period is obtained as v= dx dt = 0.02 + 0.02t (t ) 0.02u (t 0.1) + (0.002 0.02t ) (t 0.1) 0.02u (t 0.3) + (0.006 0.02t ) (t 0.3) + 0.02u (t 0.4) (0.008 0.02t) (t 0.4)
(a)
(b)
Note that the terms multiplying the unit impulse functions are zero at the times where the impulses are applied and thus these terms are zero. Thus the velocity reduces to v (t ) = 0.02 0.02u (t 0.1) 0.02u (t 0.3) + 0.02u (t 0.4) (c) 2.21 A free-body diagram of the system at an arbitrary instant is shown below. It is assumed the block slides on a frictionless surface.
46
Chapter 2
(b)
2.22 Let x be the distance traveled by the vehicle along the incline after the driver spots the stalled car. Thus from the definition of x and the information given appropriate initial conditions are (a1) x(0) = 0 m & x(0) = 60 (a2) s A free-body diagram of the vehicle, drawn at an arbitrary instant after the braking force is applied, is illustrated below.
Application of Newtons second law to the free-body diagram of the vehicle in the direction of motion leads to Fx = ma x
& x FB + mg sin = m& Rearranging Equation (b) and substituting given values leads to
(b)
47
= 5 x10 4 FB + 1.365 (c) Since the braking force is constant, Equation (c) shows that the acceleration is a constant. Integration of Equation (c) leads to the time-dependent velocity as & x = (1.365 5 x10 4 FB )t + C1 (d) The constant of integration is evaluated through application of Equation (a2) leading to m C1 = 60 . Integration of Equation (d) then gives s 1 x = (1.365 5 x10 4 FB )t 2 + 60t + C 2 (e) 2 Application of Equation (a1) leads to C 2 = 0 . Thus 1 x(t ) = 1.365 5 x10 4 FB t 2 + 60t (f) 2 & x(t ) = 1.365 5 x10 4 FB t + 60 (g) The vehicle stops when its velocity is equal to zero. Thus the minimum braking force & required is the value of FB such that x = 0 when x = 300 m . Define 4 a = 1.365 5 x10 FB . The time required for the vehicle to stop is obtained from Equation (b) as 60 ts = (h) a Requiring x(t s ) = 300 , using Equation (h) in Equation (f) gives
1 60 60 300 = a + 60 2 a a 1 3600 = (i) 2 a m Equation (i) is solved giving a = 6 2 . Thus the minimum braking force is determined s from 6 = 1.365 5 x10 4 FB
FB = 1.47 x10 4 N
2.23 The free-body diagram of the bolt is illustrated below
(j)
48
Chapter 2
As it turns the bolt rotates about a fixed axis through its center. Application of the moment equation to the free-body diagram of the bolt, assuming positive moments counterclockwise, leads to M O = I O
500 N m 480 N m = (0.035 kg m 2 ) r = 571.4 2 (a) s The angular acceleration is constant. The angular velocity of the bolt at an arbitrary time, noting that the angular velocity is zero at t=0, is t t d dt dt = dt 0 0
= 571.4t
The angular displacement of the bolt is d dt dt = dt 0 0
t t
(b)
= 285.7t 2
It takes 50 revolutions to remove the bolt, thus
(c)
2.24 The projectile motion equations obtained in Example 2.11 are x = (v0 cos )t
(a) (b)
1 y = gt 2 + (v0 sin )t 2
(a) The projectile lands when y = 100 m . Substituting given values into Equation (b) leads to 1 100 = (9.81)t 2 + 800 sin(30)t 2 2 4.90t 400t 100 = 0 (c) The quadratic formula is applied to solve Equation (c) for the time
49
Chapter 2 400 (400) 2 4(4.90)(100) t= 2(4.90) = 0.249, 81.9 (d) The negative root is discarded implying that the particle impacts the ground at t=81.9 s. (a) The horizontal range of the projectile is the value of x at the time the particle impacts the ground. From Equation (a) m x(81.9) = 800 cos(30)(81.9 s) s = 5.67 x10 4 m (e) (b) The projectile achieves its maximum altitude when the y component of velocity is zero, 0 = gt + v0 sin
t= v0 sin g
m 800 sin(30) s = m 9.81 2 s t = 40.77 s The maximum altitude is the value of y at 40.77 s. From Equation (b) m 1 m 2 y (40.77) = 9.81 2 (40.77 s ) + 800 sin(30)(40.77 s) s 2 s = 8.15 x10 3 m
(f)
(g)
2.25 A free-body diagram of the projectile at an arbitrary instant shows the gravity force and the drag force.
m (a) s2 Assuming the projectile is fired at the origin of the coordinate system the initial conditions for determining x(t) are
50
m & x(0) = 600 cos(40) s m = 459.6 (c) s Integrating Equation (a) with respect to time leads to & x(t ) = 10t + C1 (d) Application of Equation (c) to Equation (d) leads to C1 = 459.6 and & x = 10t + 459.6 (e) Integration of Equation (e) with respect to time and application of Equation (b) leads to x(t ) = 5t 2 + 459.6t (f) Summing forces in the y direction leads to Fy = ma y & y mg = m& &= 0 & y The initial conditions which are applied to determine y(t) are y ( 0) = 0 (g) (h)
m & y (0) = 600 sin( 40) s m = 385.7 (i) s Integration of Equation (g) with application of Equation (i) leads to & y (t ) = 9.81t + 385.7 (j) Integration of Equation (j) with application of Equation (h) leads to (k) y (t ) = 4.9t 2 + 385.7t (b) The maximum altitude occurs & when y = 0 which from Equation (j) occurs at t = 39.3 s . The maximum altitude is obtained from Equation (k) as y (39.3) = 4.9(31.3) 2 + 385.7(31.3)
= 7.59 x103 m (l) The time at which the projectile lands is when y=0, which is obtained using Equation (k) as t = 78.6 s . The range of the projectile is obtained from Equation (f) as x(78.6) = 5(78.6) 2 + 459.6
= 5.24 x103 m 2.26 A free-body diagram of the disk, drawn at an arbitrary instant, is shown below
51
Chapter 2
Application of Newtons second law to the free-body diagram gives & x & F kx cx = m& Noting that the centroidal mass moment of inertia of the thin disk is I =
(a)
1 2 mr and that 2 since the disk rolls without slip its angular acceleration is related to the acceleration of & x the mass center of the disk by &= r , summation of moments about the mass center of the disk leads to M G = I
& x 1 & M (t ) + Fr = mr 2 2 r Elimination of F between Equations (a) and (b) leads to 3 & & mr&+ crx + krx = M (t ) x 2
2.27 The free-body diagram of the system at an arbitrary instant is shown below.
(b)
(c)
Application of Newtons second law in the vertical direction to the free-body diagram leads to F = ma
& & & kx + k ( y x) + c( y x) = m& x Rearrangement of Equation (a) leads to & & & m&+ cx + 2kx = cy + ky x
(a)
(b)
52
Chapter 2
2.28 Let y(t) represent the displacement of the center of the disk, measured downward from the systems equilibrium position. Let represent the clockwise angular rotation of the disk, measured from the systems equilibrium position
When the center of the disk has moved distance of y downward, the cable has decreased in length by 2y. When the block has moved downward a distance x, the cable has increased in length by x. Since the cable is inextensible, x 2y = 0 x y= (a) 2 It is further assumed that there is no slip between the disk and the cable, thus x y = r
x = r 2 x = 2r x
(b)
Free-body diagrams of the disk and the block, drawn at an arbitrary instant are shown below.
53
Chapter 2
Application of Newtons second law to the free-body diagram of the block leads to F = ma & T2 kx = m& x & T2 = m& kx x Application of Newtons second law to the free-body diagram of the disk leads to & ky + T1 + T2 = m& y Application of the moment equation to the free-body diagram of the disk gives M G = I & T r T r = I&
2 1
(c)
(d)
(e)
Substitution of Equations (b) and (c) into Equation (e) leads to & & & ( m& kx )r T1r = I x x 2r I & T1 = m + 2 & kx x 2r Sub stitu & I x & x & & x x tion k 2 m + 2r 2 & kx m& kx = m 2 of Equ 5 m + I &+ 5 kx = 0 & x 2r 2 2 atio 2 ns (a),(c), and (f) into Equation (d) gives
(f)
(g)
54
Chapter 2 2.29 A free-body diagram of the system at an arbitrary instant is shown below. Assuming small the system is linear and the static spring forces cancel with the gravity forces and neither are included on the free-body diagrams.
The bar rotates about a fixed axis through O, thus the moment equation may be applied as (a) M O = I O with clockwise moments taken as positive. Application of Equation (a) leads to & 2k (3 )(3) k () = I &
O
&+ (b) I O& 19k = 0 The mass moment of inertia of abut O is the sum of the moments of inertia of each of the particles. The moment of inertia of a particle about an axis through the particle is zero. The moment of inertia of each particle about O is then obtained using the parallel axis theorem. The moment of inertia of the particle of mass 2m about O is (c) I P1 = 2m(3) 2 = 18m2 The moment of inertia of the particle of mass m about an axis through O is I p 2 = m2 (d)
2
Thus
I O = I p1 + I pw (e)
= 19m2 Substitution of Equation (e) into Equation (b) leads to &+ 19m2& 19k2 = 0 &+ m& k = 0
(f)
2.30 A free-body diagram of the system at an arbitrary instant is shown below. Assuming small the system is linear and the static spring forces cancel with the gravity forces and neither are included on the free-body diagrams.
55
Chapter 2
The bar rotates about a fixed axis through O, thus the moment equation may be applied as (a) M O = I O with clockwise moments taken as positive. Application of Equation (a) leads to & 2k (3 )(3) k () = I O& &+ I & 19k2 = 0
O
(b) The moment of inertia of the assembly about O is the sum of the moments of inertia of the two particles and the moment of inertia of the bar, I O = I Ob + I Op1 + I Op 2 (c) The moment of inertia of the bar about its cenroidal axis is obtained from Table 2.1 as 1 I b = m(4) 2 12 4 = m2 (d) 3 The parallel axis theorem is used to determine the moment of inertia of the bar about O, I Ob = I b + md 2
4 2 m + m2 3 7 = m2 3 = The moment of inertia of a particle about an axis through the particle is zero. moment of inertia of each particle about O is then obtained using the parallel theorem. The moment of inertia of the particle of mass 2m about O is I OP1 = 2m(3) 2 = 18m2 The moment of inertia of the particle of mass m about an axis through O is I Op 2 = m2 Substitution of Equations (e), (f), and (g) into Equation (c) gives 7 I O = m2 + 18m2 + m2 3 64 2 = m 3 Use of Equation (h) in Equation (b) leads to 64 2 & m & 19k2 = 0 + 3
(g)
(h)
(i)
56
Chapter 2 2.31 For the system shown in Figure P2.31, let be the counterclockwise angular rotation of the disk, measured from the systems equilibrium position, and let y be the displacement of the particle on the outer radius of the disk. Kinematics is used to give x x = r = (a) r x y = 3r = 3r = 3x (b) r Free-body diagrams of the disk and the block, drawn at an arbitrary instant are shown below
Application of Newtons second law to the free-body diagram of the block leads to F = ma
& T 2kx = m& x (c) Application of the moment equation, assuming moments are positive counterclockwise, to the free-body diagram of the disk gives M G = I & x & Tr 3kx(3r ) = I r Elimination of T between Equations (c) and (d) leads to I & x mr + &+ 11krx = 0 r (d)
(e)
2.32 Define as the clockwise angular rotation of the upper bar measured from the systems equilibrium position and define as the counterclockwise rotation of the lower bar, measured from its equilibrium position. The link connecting the two bars is rigid.
57
Chapter 2 Thus the displacements of the particles on each bar where the link is attached must be the same. Assuming small displacements this leads to L 2L = 2 3 3 = (a) 4 Free-body diagrams of each bar, drawn at an arbitrary instant, are shown below.
1 mL2 , 12 application of the moment equation, assuming positive moments clockwise, to the upper bar gives M A = I
L L L : 1 & k + F = mL2& (b) 2 2 2 12 The parallel-axis theorem is used to determine the moment of inertia of the lower bar
1 1 L about its pin support at B as I B = mL2 + m = mL2 . Application of the moment 12 3 2 equation to the lower bar, assuming positive moments are clockwise, gives M B = I B
2
3 & 2L 1 3 & cL (L ) F = mL2 & 4 3 3 4 Equation (a) is used to solve for the force in the rigid link as 1 &+ 1 F = mL2& kL 6 2 Substitution of Equation (c) into Equation (d) leads to 1 2 L 1 &+ 1 &+ 3 & mL2& kL = mL2& cL2 2 4 6 3 4 Equation (e) is rearranged to give
(c)
(d)
(e)
58
Chapter 2
13 2 & 3 2 & 1 2 mL & cL + kL = 0 + 36 4 3
(f)
Since the pin support is a fixed axis of rotation the appropriate form of the moment equation is (a) M O = I O where moments are taken positive counterclockwise to be consistent with the chosen positive direction for . Application of Equation (a) to the free-body diagram, assuming small , leads to L & k t mg = I O& (b) 2 The moment of inertia of a slender bar about its centroidal axis is obtained from Table 2.1 as 1 I = mL2 (c) 12 Application of the parallel axis theorem leads to
1 L mL2 + m 12 2 1 = mL2 3 Use of Equation (d) in Equation (b) and subsequent rearrangement leads to 1 2& L + mL & k t + mg = 0 3 2 IO =
2
(d)
(e)
2.34 Free-body diagrams of the blocks, drawn at an arbitrary instant, are illustrated below
59
Chapter 2
& & 3000 x1 200 x1 + 1000( x 2 x1 ) = 30 & x1 Application of Newtons second law to the right block gives & 1000( x2 x1 ) 4000x2 = 20& x2 Equations (a) and (b) are rearranged and written in a matrix form as & x1 30 0 & 4000 1000 x1 0 0 20 & + 1000 5000 x = 0 & x2 2
(a)
(b)
(c)
2.35 Free-body diagrams of the bar and the block, drawn at an arbitrary instant are shown below assuming small . Static spring forces and gravity cancel are not included as they will cancel with each other in the governing equations.
Summing forces on the free-body diagram of the block assuming positive forces act downward leads to F = ma
& k ( x L ) = m& x & m&+ kx kL = 0 x (a) Since the bar rotates about a fixed axis through the pin support the appropriate moment equation to be applied to the free-body diagram of the bar is
60
Chapter 2
= I O (b) The moment of inertia of the slender bar about its own centroidal axis is obtained from 1 Table 2.1 as I = mL2 . Application of the parallel axis theorem gives 12 2 1 L I O = mL2 + m 12 2 1 = mL2 (c) 3 Application of Equation (b) to the free-body diagram of the bar, taking clockwise moments as positive, leads to 1 & kL ( L) + k ( x L )( L) = mL2& 3 1 2& mL & kLx + 2kL2 = 0 (d) 3 Equations (a) and (d) provide the mathematical model for the system.
O
2.36 Free-body diagrams of each of the blocks at an arbitrary instant are illustrated.
Application of Newtons second law to each of the blocks, assuming forces are positive to the right, leads to & kx1 + k ( x 2 x1 ) = m& x1 (a) & & & k ( x 2 x1 ) + 2k ( x 3 x 2 ) + c ( x 3 x 2 ) = 2m& x2 (b) & & & 2k ( x3 x 2 ) c ( x3 x 2 ) + F (t ) = m& x3 Equations (a), (b), and (c) are rearranged as & m& + 2kx1 kx2 = 0 x1 & & & x2 2m& + cx 2 cx3 kx1 + 2kx2 kx3 = 0 & & & m& cx 2 + cx3 kx2 + kx3 = F (t ) x3 Equations (d), (e), and (f) are summarized in matrix form as & & x1 0 x1 2k k 0 x1 0 m 0 0 & 0 0 0 2 m 0 & + 0 c c x + k 2 k k x = 0 & & x2 2 2 0 0 m & 0 c c x3 0 k k x3 F (t ) & & x3 2.37 Free-body diagrams of the link and the block are illustrated assuming small x. (c)
(g)
61
Chapter 2
Since the link is massless, application of the moment equation to its free-body diagram leads to MO = 0 F (t ) + G (t )(3) = 0 1 G (t ) = F (t ) 3 Application of Newtons second law to the free-body diagram of the block gives & x & G (t ) kx cx = m& 1 & m&+ cx + kx = F (t ) x & 3 (a)
(b)
2.38 Assuming small displacements and using similar triangles, as illustrated below, the displacement of the end of the spring attached to the bar of length 3 is 2 y (t ) such that the spring is stretched.
A free-body diagram of the bar at an arbitrary instant, assuming small , is illustrated. Note that the since is assumed positive clockwise rotation of the bar in the positive direction of leads to an increase in length of the spring. The bar rotates about a fixed axis at O, thus the appropriate form of the moment equation is (a) M O = I O The moment of inertia of the slender bar about its own centroidal axis is obtained from 1 Table 2.1 as I = m2 . Application of the parallel axis theorem gives 12
62
Chapter 2
1 m2 + m 12 2 1 = m2 (b) 3 Application of Equation (a) to the free-body diagram of the bar, taking clockwise moments as positive, leads to 1 & & k (2 y + ) c () + mg = m2& 2 3 1 2& & + m & c2 + k2 mg = 2k2 y (t ) (c) 3 2 IO =
2
2.39 Define (t ) as the clockwise angular displacement of the crank from the aligned position, as illustrated below.
Define x p (t ) as the distance between the piston and the center of the crank at any instant. From the above diagram x p = 0.3 sin + 0.1 cos (a) The velocity of the piston is calculated as & & & x p = 0.3 cos 0.1 sin (b) & Noting that = , which is a constant, differentiation of Equation (b) leads to & & & = 0.3 2 sin 0.1 2 cos & x (c)
p
(a) Since the angular velocity is constant and the system is aligned at t=0, = t Use of Equation (d) in Equations (a), (b), and (c) leads to x p = 0.3 sin(t ) + 0.1 cos(t ) v p = 0.3 cos(t ) 0.1 sin(t )
(d)
(e) (f)
a p = 0.3 2 sin(t ) 0.1 2 cos(t ) (g) (b)Assume the inertia effects of the crank and connecting rod are negligible. A free-body diagram of the engine, drawn at an arbitrary instant, is illustrated below 63
Chapter 2
Application of Newtons second law, in the form of DAlemberts principle to the freebody diagram leads to Fext = Feff & & & & 9000 x 1000 x = 10 &+ 0.3( &+ & ) x x xp & & & x xp 10.3&+ 1000 x + 9000 x = 0.3& Use of Equation (g) in Equation (h) leads to & & 10.3&+ 1000 x + 9000 x = 0.3( 0.3 2 sin(t ) 0.1 2 cos(t ) ) x Noting that = 250 rpm = 26.2 r / s , Equation (i) becomes & & 10.3&+ 1000 x + 9000 x = 61.8 sin(t ) 20.6 cos(t ) x (h) (i)
(j)
64
Chapter 3
3. Electric Systems
3.1(a) The capacitance for a parallel plate capacitor is
A d Equation (a) is rearranged to solve for the separation distance as K 0 A d= C Substitution of given numerical values into Equation (b) gives C2 0 .5 m 2 (4.6) 8.85 x10 12 N m2 d= 8 5 x10 F = 0.410 mm (b) The charge on this parallel-plate capacitor when connected to a 12 V battery is q = Cv C = K 0 (a)
(b)
(c)
= 5 x10 8 F (12 V) = 6.0 x10 C (c) The energy stored in the capacitor after being charged by a 12 V battery is 1 E = Cv 2 2 1 2 = 5 x10 8 F (12 V ) 2 = 3.60 J
7
(d)
(e)
P = vi
= (20 V )(1.5 A ) = 30 W
(a)
(b)
3.3 (a) The power in the CD player is given by P = vi Thus for a 200 W CD player run from a 120 V source,
(a)
65
Chapter 3
i= P v 200 W = 120 V = 1.67 A
(b)
(c)
3.4 The power delivered by the heating coil is E P= t J 1 hr = 1500 x10 3 hr 3600 s = 416.7 W The power is related to the resistance of the coil through v2 P= R The resistance is calculated as v2 R= P (120 V )2 = 416.7 W = 34.6 3.5 The current in a series LRC circuit is i (t ) = 10 sin(100t + 0.4) A (a)The voltage across a 100 resistor is v R = iR = [10 sin(100t + 0.4) A ](100 ) = 1x10 3 sin(100t + 0.4) V (b)The voltage change across the capacitor, assuming it is uncharged at t=0, is
(a)
(b)
(c)
(a)
(b)
66
Chapter 3 1 idt C 0
t t
vC =
= 5 x10 5 [0.921 cos(100t + 0.4)] V (c) (c)The voltage across the inductor is di vL = L dt d = (0.25) [10 sin(100t + 0.4)] dt = (0.25)(10)(100)cos(100t + 0.4) = -250cos(100t + 0.4) V (d) (d) Application of KVL around the series LRC circuit shows that the voltage provided by the source is the sum of the voltage changes across the resistor, capacitor, and inductor. Thus from Equations (b), (c), and (d), v (t ) = vR + vC + vL
= 1x10 3 sin(100t + 0.4) + 5 x10 5 [0.921 cos(100t + 0.4)] 250 cos(100t + 0.4)
(e)
1 2 Cv c (a) 2 Substitution of Equation (c) from the solution of Problem 3.5 into Equation (a) leads to 2 1 E c = 0.2 x10 6 F 5 x10 5 [0.921 cos(100t + 0.4)]V 2 = 2.5 x10 4 0.8482 1.842 cos(100t + 0.4) + cos 2 (100t + 0.4) Ec =
)(
= 2.5 x10 4 [1.3482 1.842 cos(100t + 0.4) + 0.5 cos(200t + 0.8)] N m Equation (b) is plotted below
(b)
67
Chapter 3
Substituting given values into Equation (c) leads to 1 2 E L = (0.25 H )[10 sin(100t + 0.4 A] 2 = 12.5 sin 2 (100t + 0.4) N m Equation (d) is plotted below
(d)
68
Chapter 3
(e)
= 5 x10 3 [t 0.05 sin( 200t + 0.8) + sin( 0.8)] N m Equation (f) is plotted below.
(f)
69
Chapter 3
Application of KCL at node B leads to i1 i2 i3 = 0 Application of KCL at node C gives i3 i 4 i 5 = 0 KVL is applied around each loop in a clockwise direction. For loop ABGH 3i1 5i3 + 12 = 0 For loop BCFG (b) (c) (a)
70
Chapter 3 2i2 4i5 + 5i3 = 0 For loop CDEF 3i4 4i4 + 4i5 = 0 (e) Equations (a)-(e) are summarized in a matrix form as 0 i 0 1 1 1 0 0 0 1 1 1 i2 0 3 0 (f) 5 0 0 i3 = 12 0 4 i4 0 0 2 5 0 0 0 7 4 i 5 0 The equations represented by Equation (f) are solved simultaneously leading to (g) i1 = 2.29 A, i2 = 1.26 A, i3 = 1.03 A, i4 = 0.373 A, i5 = 0.654 A (b) The voltage drops across the resistors are determined using Ohms law. The subscripts correspond to the labels in the figure v1 = 3ii = 6.87 V, v 2 = 5i3 = 5.15 V, v3 = 2i2 = 2.52 V, v 4 = 4i5 = 2.62 V, (d)
(h) v5 = 3i4 = 1.12 V, v6 = 4i4 = 1.49 V (c) The power in each resistor is calculated as 2 P1 = 3i12 = 15.73 W, P2 = 5i32 = 5.30 W, P3 = 2i2 = 3.18 W, P4 = 4i52 = 1.71 W,
2 2 P5 = 3i4 = 0.417 W, P6 = 4i4 = 0.557 W
(i)
3.8 (a) The reduction of the circuit to a single resistor of equivalent resistance is illustrated below
71
Chapter 3 The reduction shows that the resistors may be replaced by a single resistor of equivalent resistance 1.3 . (b) The power delivered to the circuit is v2 P= Req 1.3 = 0.111 kW 3.9 The reduction of the resistive circuit is illustrated below.
(12 V )2 =
(a)
The equivalent resistance of the combination of resistors is 157 Req = 23 = 6.83 3.10 The circuit may be reduced as illustrated below
(a)
72
Chapter 3 Application of KCL at node B leads to i1 i2 i3 = 0 Application of KVL around loop ABEF gives 127 10 i1 9i3 = 0 8 Application pf KVL around loop BCDE leads to 21i2 + 9i3 = 0 Equations (a)-(c) are summarized in matrix form as 1 1 1 i1 0 127 0 9 i2 = 10 8 0 21 9 i3 0 The solution of Equation (d) is i1 0.415 A i = 0.135 A 2 i3 0.316 A (a)
(b)
(c)
(d)
(e)
The positive sign for i2 indicates the current in loop BCDE flows in the direction assumed. Thus the current is flowing in the direction from the + to - terminals placed across the resistor, indicating a voltage decrease of v = (0.135 A)(11 ) = 1.49 V (f) 3.11 From KVL the voltage change across each pair of series resistors is v1 . Let i1 be the current through R1 and R2 . Then v1 = i1 ( R1 + R2 )
v1 R1 + R2 v1 . Let i2 be the current through R3 and R4 . Then v1 = i2 ( R3 + R4 ) i1 = i2 = v1 R3 + R4 (a)
(b)
73
Chapter 3 Define v A , v B , and vC as in the above diagram. Then v B = v A i1 R1 vC = v A i2 R3 v0 = vC v B Substitution of Equations (c) and (d) into Equation (e) leads to v0 = v A i2 R3 v A + i1 R1 = i1 R1 i2 R3 (a) Use of Equations (a) and (b) in Equation (f) yields vR vR v0 = 1 1 1 3 R1 + R2 R3 + R4
R1 R4 R2 R3 = v1 ( R1 + R2 )( R3 + R4 ) (b) The output voltage is zero if R1 R4 = R2 R3 (h) (c) The resistance in a wire is proportional to its length , R = , where is the resistivity of the wire. The strain of a wire initially of length a which has changed to length b is a = b a
Rb Ra (i) Ra Suppose the strain gauge is calibrated such that v0 = 0 when the wire is unstretched =
(f)
(g)
R R Ra = R4 = 2 3 . Then, when stretched the measured voltage is given by Equation (g) R1 with R4 = Rb .
3.12 Let i be the current through the resistors. Since the resistors are in series Ohms law gives v1 = i( R1 + R2 ) (a) Equation (a) is rearranged to v1 (b) i= R1 + R2
74
Chapter 3
Using the definitions of voltages in the above diagram, v A = vC + v1 v B = v A iR2 v 2 = v B vC = v A iR2 v A + v1 = v1 iR2 Substitution of Equation (b) in Equation (e) leads to v 2 = v1 v1 R2 R1 + R2 v1
(c) (d)
(e)
R1 = R +R 2 1
(f)
3.13 The relation between the voltage across a capacitor and the charge on the capacitor is q = Cv (a) Application of conservation of charge at node B leads to (b) q1 q 2 q3 = 0 Application of KVL around loop ABEF gives q3 q1 q1 +6=0 (c) 6 6 3x10 5 x10 2 x10 6 Application of KVL around loop BCDE results in q3 q2 + =0 (d) 6 4 x10 2 x10 6
75
Chapter 3
1 8 15 0 The solution of Equation (e) is 1 1 q1 0 1 6 0 q 2 = 6 x10 2 1 1 q3 0 4 2
(e)
0.271 C q1 q = 10 6 0.081 C (f) 2 0.189 C q3 The voltage across each capacitor is calculated using Equation (a). The voltage across the 3 F capacitor is q v1 = 1 C1
0.271x10 -6 C 3x10 -6 F = 0.0902 V = The voltage across the 5 F capacitor is
(g)
v2 = = The voltage across the 4 F capacitor is v3 = = The voltage across the 2 F capacitor is v4 = =
q1 C 21
(h)
(i)
(j)
76
(a)
3.15 Define v R as the voltage drop across the resistor and vC as the voltage drop across the capacitor.
77
Chapter 3 Application of KVL around loop BCDE leads to vC + v R = 0 Application of KCL at node be results in i (t ) i2 i3 = 0 The voltage drop across the resistor is calculated using Ohms law as v R = i3 R i3 = The voltage drop across the capacitor is obtained as
vC = 1 i2 dt C 0
t
(b) (c)
vR R
(d)
(e)
(f)
3.16 Application of KVL in a clockwise direction around the circuit leads to di L R1 i R 2 i + v(t ) = 0 dt Equation (a) is rearranged to di L + ( R1 + R 2 )i = v(t ) dt
(a)
(b)
3.17 Let i (t ) be the current in the circuit. Application of KVL over the loop, including the effect of mutual inductance, results in di di di di v L1 + M L2 + M Ri = 0 (a) dt dt dt dt Equation (a) is rearranged as (L1 + L2 + 2M ) di + Ri = v (b) dt
78
Chapter 3
3.18 (a) When the switch is open the circuit is the resistive circuit shown below.
Application of KCL at node B gives i1 i2 i3 = 0 Application of KVL around loop ABEF leads to 10 6i1 4i3 = 0 Application of KVL around loop BCDE leads to 5i2 + 4i3 = 0 Equations (a), (b), and (c) are summarized in matrix form as 1 1 1 i1 0 6 0 4 i2 = 10 0 4 5 i3 0 The solution of Equation (d) is (a) (b)
(c)
(d)
(e)
Application of KCL at node B and KVL around loops ABEF BCDE leads to
79
Chapter 3 10 6i1 4(i1 i2 ) = 0 5i2 Equations (f) and (g) are rewritten as 1 i2 dt + 4(i1 i2 ) = 0 5 x10 6 0
t
(f) (g)
(h)
t
(i)
Application of KCL at node B gives i1 i 2 i3 = 0 Application of KVL around loop ABEF leads to 8i1 4i3 + v(t ) = 0 Application of KVL around loop BCDE gives di 0.003 2 + 4i3 = 0 dt Equations (b) is rearranged to give 1 i 3 = v(t ) 2i i 4 Use of Equation (d) in Equation (a) leads to 1 i1 i 2 v(t ) 2i1 = 0 4 1 1 i1 = i 2 + v(t ) 3 12 Equations (d) and (e) are used to give 1 2 i3 = v(t ) i 2 12 3 Use of Equations (e) and (f) in Equation (c) leads to di 2 1 0.003 2 + 4 v(t ) i2 = 0 3 dt 12 di 8 1 0.003 2 + i2 = v(t ) dt 3 3 80 (a) (b)
(c)
(d)
(e)
(f)
(g)
Chapter 3
Ri1 L
Equation (a) is rearranged as
di1 + R(i i1 ) = 0 dt
(a)
di1 + 2 Rii = Ri dt
(b)
3.21 The currents are defined in the circuit diagram shown below
Application of KCL at nodes B and C gives i1 i2 i4 = 0 i 2 i3 i5 = 0 Application of KVL around loop ABGH results in di v1 L1 1 R1i4 = 0 dt Application of KVL around loop BCFG gives di L2 2 R2 i5 + R1i4 = 0 dt Application of KVL around loop CDEF leads to
(a) (b)
(c)
(d)
81
Chapter 3
1 i3 dt + v 2 + R2 i5 = 0 (e) C 0 Substituting Equations (a) and (b) into Equations (c), (d), and (e) and rearranging leads to di L1 1 + R1i1 R1i2 = v1 (f) dt di L2 2 R1i1 + ( R1 + R2 )i2 R2 i3 = 0 (g) dt t 1 R2 i2 + R2 i3 + i3 dt = v 2 (h) C0
3.22 Let i(t) be the current in the circuit after the switch is closed. Noting that the initial Q voltage in the initially charged capacitor is v(0) = application of KVL around the C1 circuit leads to
Ri
Q 1 1 idt C1 idt + C1 = 0 C2 0 0
(a)
The appropriate initial condition is i (0) = 0 . 3.23 The currents are defined in the diagram below.
Application of KCL at node B gives i1 i2 i3 = 0 Application of KVL around loop ABDE leads to di R1i1 L1 3 + v = 0 dt Application of KVL around loop BCDE gives di di R2 i 2 L2 2 + L1 3 = 0 dt dt Equation (a) is rearranged as i 3 = i1 i 2 Use of Equation (d) in Equations (b) and (c) along with rearrangement gives 82 (a)
(b)
(c)
(d)
Chapter 3 3.24 The currents are defined in the circuit diagram shown below di di L1 1 L1 2 + R1i1 = v dt dt (L1 + L2 ) di 2 L1 di1 + R 2 i2 = 0 dt dt
(e) (f)
Application of KCL at node B leads to i(t ) i1 i2 = 0 Application of KVL around loop ABEF leads to t 1 3000i i2 dt = 0 0.2 x10 6 0 Application of KVL around loop BCDE results in t di1 1 0.4 6000i1 + i2 dt dt 0.2 x10 6 0 3.25 Noting, from application of KCL, that the current through the inductor is i1 i2 , application of KVL around each loop leads to d R1i1 L (i1 i2 ) = v dt t 1 d i2 dt R2 i2 + L (i1 i2 ) C0 dt Equations (a) and (b) are rearranged as di di L 1 L 2 + R1i1 = v dt dt t di di 1 L 1 + L 2 + R2 i2 + i2 dt = 0 dt dt C0 3.26 Currents are defined in the diagram below.
(a)
(b)
(c)
(a) (b)
(c) (d)
83
Chapter 3 Application of KCL at nodes B and C gives i1 i2 i4 = 0 i 2 i3 i5 = 0 Equations (a) and (b) are rearranged to give i 4 = i1 i 2 i5 = i 2 i3 Application of KVL around loop ABGH gives t di 1 L1 1 i 4 dt R1i1 + v = 0 dt C1 0 Application of KVL around loop BCFG gives t di 1 1 L2 2 i5 R2 i5 + i 4 dt = 0 dt C 2 C1 0 Application of KVL around loop CDEF gives t t di 1 1 L3 3 i3 dt + R 2 i5 + i 4 dt = 0 dt C 3 C2 0 0 Substitution of Equations (c) and (d) into Equations (e)-(g) and rearranging leads to t t di 1 1 L1 1 + R1i1 + i1 dt i2 dt = v dt C1 C1 0 0
t t t 1 di2 1 1 1 i1 dt +L2 + R2 i 2 + + C C i2 dt R2 i3 C i3 dt = 0 dt C1 2 0 2 0 1 0 t t 1 di 1 1 i3 dt = 0 i2 dt + L3 3 + R2 i3 + R2 i 2 + C C2 dt C3 2 0 0
(e)
(f)
(g)
84
Chapter 3
(a) (b)
i5 i 6 i 7 = 0 (c) The resistor of resistance R2 is in parallel with the capacitor of capacitance C2 . Thus the potential change across each is the same t 1 R2 i 6 = i7 dt (d) C2 0 Application of KVL around loops ABGJ, BCFG, and CDEF leads to t di 1 i4 dt R1i4 + v1 = 0 L1 1 (e) dt C1 0
t di2 1 i4 dt L2 R2 i6 + R1i4 dt C1 0
(f) (g)
di3 + v 2 + R2 i6 = 0 dt Differentiation of Equation (d) with respect to time leads to di R2 6 = C 2 i7 dt di i7 = R2 C 2 6 dt Substitution of Equation (h) into Equation (c) gives L3
(h)
i5 = i 6 + R 2 C 2
Substitution of Equation (i) into Equation (b) gives
di6 dt
(i)
di6 =0 dt Use of Equation (a) in Equations (g) and (h) leads to t t di 1 1 i1 dt R1i2 i2 dt = v1 L1 1 + R1i1 + dt C1 C1 0 0 i 2 i3 i 6 R2 C 2
t t di2 1 1 R1i1 i1 dt +L2 + R1i2 + i2 dt + R2 i6 = 0 C1 dt C1 0 0 Equations (g), (h), (k), and (l) form the mathematical model of the system.
(j)
(k) (l)
85
Chapter 3
The resistor of resistance R2 is in parallel with the capacitor of capacitance C2 . Thus the potential change across each is the same t 1 R2 i 6 = i7 dt (d) C2 0 Application of KVL around loops ABGJ, BCFG, and CDEF, including the mutual inductance between adjacent inductors leads to t di di 1 i4 dt R1i4 + v1 = 0 L1 1 M 2 (e) dt dt C1 0
t di3 di2 di1 1 i4 dt L2 M M R2 i6 + R1i4 dt dt dt C1 0
(f) (g)
L3
di3 di M 2 + v 2 + R2 i6 = 0 dt dt
Differentiation of Equation (d) with respect to time leads to di R2 6 = C 2 i7 dt di i7 = R2 C 2 6 dt Substitution of Equation (h) into Equation (c) gives
(h)
86
Chapter 3
i5 = i 6 + R 2 C 2
Substitution of Equation (i) into Equation (b) gives
di6 dt
(i)
di6 =0 dt Use of Equation (a) in Equations (g) and (h) leads to t t di di 1 1 L1 1 + R1i1 + i1dt + M 2 R1i2 i2 dt = v1 dt C1 dt C1 0 0 i 2 i3 i 6 R2 C 2
di di1 di 1 1 R1i1 i1dt +L2 dt2 + M dt3 + R1i2 + C1 i2 dt + R2i6 = 0 dt C1 0 0 Equations (g), (h), (k), and (l) form the mathematical model of the system. M
t t
(j)
(k) (l)
87
Chapter 3
88
Chapter 3
89
Chapter 3 3.33 The electric circuit analogy for the mechanical system is illustrated below.
3.34 The electric circuit analogy for the mechanical system is illustrated below
3.35 The circuit is the non-inverting circuit of Example 3.17(b).Use of Equation (g) of Example 3.17 leads to R v 2 = 1 + 2 v1 R1 4 = 1 + v1 2 (a) = 3v1 3.36 Assume the amplifier is an ideal amplifier. Thus referring to the diagram below,
90
Chapter 3
i3 = i 7 = 0 Applications of KCL at nodes A, B, and C give respectively i1 i2 i3 = 0 i2 i4 = 0 i 4 i5 i 6 = 0 Application of Ohms law across the circuit components leads to v i1 = A 400 v A vB i2 = 800 v vC i4 = B 600 v vC i5 = F 300 Application of KVL around loop EFCD leads to 100i6 + 300i5 + v1 = 0 Equation (h) is used in Equation (i), which is then rearranged to v 300 v F vC i6 = 1 + 100 100 300 v vC v = 1 + F 100 100 Substitution of Equations (e)-hi) and Equation (j) in Equations (b)-(d) leads to v v vB =0 A A 400 800 v A v B v B vC =0 800 600 v B vC v F vC v vC v 1 F =0 600 300 100 100 The amplifier is ideal, thus the potentials at each of its terminals is the same,
(i)
(j)
91
Chapter 3
vF = v A
Using Equation (n), Equations (k)-(l) are rearranged as 3v A + v B = 0 3v A 7v B + 4vC = 0 v B 10vC = 6v1 The solution of Equations (o)-(q) is 1 v A = v1 7 3 v B = v1 7 9 vC = v1 14 The output v2 is v 2 = v B vC = 3.37 Currents are as defined below. 3 v1 14
(n)
(u)
Assuming the amplifier is ideal, (a) i3 = i 4 = 0 Since there is no current through the 400 resistor, the voltage input to the amplifier is v1 which is also the voltage at its negative terminal. Application of KCL at A gives i1 i2 = 0 (b) Use of Ohms law across the resistors gives v vA i1 = 1 (c) 600 v v3 (d) i2 = A 1800 Substitution of Equations (c) and (d) into Equation (b) and noting that v A = v 2 leads to v1 v2 v2 v3 =0 600 1800 3v1 3v2 v2 + v3 = 0 92 v3 = 4v2 v1 (e)
Chapter 3
The first amplifier is an inverting amplifier, thus from Equation (c) of Example 3.17 R v B = 2 v1 R1
600 v1 300 = -2v1 (a) Application of KCL at node C assuming ideal amplifiers leads to i1 i2 = 0 (b) Applications of Ohms law across circuit components lead to v vC i1 = B 400 2v1 0 = 400 v = 1 (c) 200 v v2 i2 = C 500 v = 2 (d) 500 Substitution of Equations (c) and (d) in Equation (b) leads to v v 1 + 2 =0 200 500 5 v 2 = v1 (e) 2 3.39 The circuit is that of the integrating circuit of Figure 3.28 whose output is given in Equation (3.66) =
93
= 8.33x10
3
sin(200t )dt
0
= 4.17 x10 [cos(200t ) 1] V 3.40 Currents and potentials are defined below.
(a)
Since the amplifier is ideal i4 = 0 v A = v2 Application of KCL at A gives i1 + i3 i2 = 0 Use of component laws leads to i1 = v1 v A 16 (c) (d) (e) (f) (a) (b)
d (v A v2 ) dt d i3 = 30 x10 6 ( v A ) dt Use of Equations (b), and (d)-(f) in Equation (c) results in v1 v 2 dv 30 x10 6 2 = 0 16 dt dv 4.8 x10 4 2 + v 2 = v1 dt i2 = 50 x10 6 94
(g)
Chapter 3 The output v2 is the solution of the differential equation of Equation (g). 3.41 The currents and potentials are as defined below.
(a)
dvi dt
(b)
(c)
i 2 i 4 i5 = 0
Noting that the parallel circuit components have the same voltage change, v A v2 = v2 , application of component laws give v i4 = o (d) R dv i 5 = C 2 o (e) dt Substitution of Equations (d) and (e) in Equation (c) leads to v dv i2 = o C 2 o (f) R dt Substitution of Equations (b) and (f) into Equation (a) leads to dv v dv C1 i + o + C 2 o = 0 dt R dt dv dv RC 2 o + vo = RC1 i (g) dt dt The output vo (t ) is the solution of Equation (g).
95
Chapter 3 3.42 Currents and potentials are defined in the diagram below
Since the amplifier is ideal, i4 = 0 and v B = vo . Applications of KCL at nodes A and B give i1 i2 i3 = 0 (a) i3 + i5 = 0 Applications of component laws lead to
vi v A R d i2 = C 2 (v A v o ) dt v A vo i3 = R dv i5 = C1 o dt Substitution of Equations (c)-(f) in Equations (a) and (b) leads to vi v A v vo d C 2 (v A vo ) A =0 R dt R dv v A vo C1 o = 0 dt R Equation (h) is solved for v A resulting in dv v A = RC1 o + v0 dt Substitution of Equation (i) in Equation (g) gives d 2 vo dv dv v vi C1 o o RC1C 2 C1 o = 0 2 dt dt R R dt 2 d vo dv 1 1 RC1C 2 + 2C1 o + vo = vi 2 dt R R dt The output vo (t ) is the solution of Equation (j). i1 = (c) (d) (e) (f)
(b)
(g) (h)
(i)
(j)
96
leads to i1 i2 i3 = 0 where v1 v A (b) R d i2 = C 2 (v A vo ) (c) dt v vB i3 = A (d) R Assuming an ideal amplifier, v B = vo and substitution of Equations (b)-(d) in Equation (a) leads to v v v1 v A d C2 (v A vo ) A 0 = 0 (e) R dt R i1 = Application of KCL at node B gives i3 i4 i5 = 0 Since the amplifier is ideal, i5 = 0 . Additionally (f) (a)
i4 = C1
dvo dt
(g)
Substitution of Equations (d) and (g) in Equation (f) gives v A vo dv C1 o = 0 R dt dv v A = vo + RC1 o dt Substituting Equation (h) into Equation (e) leads to d 2 vo dv dvo v1 vo C1 C 2 RC1 C1 o = 0 2 dt dt R R dt 2 d v dv v v C1C 2 R 2o + 2C1 o + o = 1 dt R R dt 97
(h)
(i)
Chapter 3 3.44 The differential equation governing the motion of the system shown is & & 2 &+ 10 x + 12 x = F (t ) x (a) The differential Equation derived in Example 3.30 for the amplifier circuit of Figure 3.47 is d 2 v 0 R3 C 2 dv 0 R5 v v0 = 1 (a) C1 C 2 R 3 + + 2 R 2 dt R 4 R6 R1 dt
which is of the form of Equation (a). The circuit is designed to simulate Equation (a) if (b) R1 = 1
C1C 2 R3 = 2 R3 C 2 = 10 R2 (c) (d)
R5 = 12 (e) R 4 R6 There are many solutions of Equations (b)-(e). One solution is C1 = 2 F, C = 1 F, R1 = 1 , R2 = 0.1 , R3 = 1 , R 4 = 1 , R5 = 12 , R6 = 1 (f)
i1 i2 = 0
However i1 = i2 = 1 (v1 v A )dt L1 0
t
(a)
(b)
v A vB (c) R1 Assuming the amplifier is ideal, v A = 0 , then substitution of Equations (b) and (c) into Equation (a) leads to
98
Chapter 3
vB 1 v1dt + R1 = 0 L1 0 vB = R1 t v1 dt L1 0 (d)
t
(e)
(f) (g) (h)
Assuming the amplifier is ideal, vC = 0 and substitution of Equations (d), (f),(g), and (h) in Equation (e) leads to t R t v 1 1 v1 dt + 2 + v 2 dt = 0 L1 R2 0 R3 L 2 0
R3 t R3 R1 t v2 + v 2 dt = v1 dt L2 L1 R2 0 0 Differentiation of Equation (i) with respect to time leads to RR R & v 2 + 3 v 2 = 3 1 v1 L1 R2 L2 If R3 = 10 , L1 = 1 H, L2 = 1 H, R1 = 3 and R2 = 10 , Equation (j) becomes & v2 + 10v2 = 3v1 (i)
(j)
(k)
Referring to the circuit diagram application of KCL at node Q1 leads to i1 i2 i3 = 0 Assuming an ideal amplifier, i3 = 0 and vQ1 = v1 . Thus
(a)
99
Chapter 3
i1 = i2 =
v1 R1
(b) (c)
v1 v A d + C1 (v1 v A ) R2 dt Substitution of Equations (b) and (c) into Equation (a) leads to v v vA d 1 1 C1 (v1 v A ) = 0 dt R1 R2
1 dv A dv1 1 1 v A = v1 + + R R + C1 dt dt R2 2 1 Application of KCL at node Q2 gives i 4 i5 i 6 = 0 Assuming an ideal amplifier, i6 = 0 and v Q2 = 0 . Thus C1
vA R3 dv vB i5 = C2 B R4 dt Substitution of Equations (f) and (g) in Equation (e) leads to vA v d + C 2 (v B ) + B = 0 R3 dt R4 i4 =
(d)
(e)
(f)
(g)
dv B R3 vB dt R4 The final amplifier is simply an inverter, such that R vC = 6 v B R5 Substitution of Equation (i) in Equation (h) leads to R C R dv RR v A = 5 2 3 C + 3 5 vC R6 dt R6 Substitution of Equation (j) in Equation (d) leads to 1 C1C 2 R3 R5 d 2 vC C1 R3 R5 R5 C 2 R3 dv A R3 R5 dv 1 v1 + C1 1 + vA = + + + 2 R R R R6 R2 dt R 2 R6 dt dt 6 2 1 Equation (e) simulates a differential equation of the form & & & & vC + 10vC + 4vC = 3v1 + 2v 2 when the circuit components are chosen with values such that v A = R3 C 2
(h)
(i)
(j)
(k)
(l)
100
Chapter 3 C1C 2 R3 R5 =1 R6 C1 R3 R5 R5 C 2 R3 + = 10 R6 R2 R3 R5 =4 R2 R6 1 1 + =2 R1 R2 C1 = 3
3.47 The differential equations for a field controlled DC motor are di f Lf + Rf if = v (a) dt d J eq + ct ,eq K f i f = 0 (b) dt The kinetic energy of the system at an arbitrary instant is 1 1 1 2 T = ( J r1 + J G1 )12 + ( J G 2 + J G 3 ) 2 + ( J G 4 + J r 4 ) 32 (c) 2 2 2 where the gear ratios are used to determine the angular velocities of the shafts in terms of the angular velocity of the motor shaft 1 n11 = n 2 2
n1 1 n2 n3 2 = n 4 3
2 =
(d)
nn n3 2 = 1 3 n n n4 2 4 Substitution of Equations (d) and (e) into Equation (c) and rearranging leads to 2 n1 2 n1 n3 2 1 T = J r1 + J G1 + ( J G 2 + J G 3 ) n + ( J G 4 + J r 4 ) n n 1 2 2 2 4 The equivalent moment of inertia is determined from Equation (f) as 2 n1 2 n1 n3 J eq = J r1 + J G1 + ( J G 2 + J G 3 ) n + ( J G 4 + J r 4 ) n n 2 4 2 The work done by all torsional viscous damping moments is 2 2 n1 n1 n3 W = c t 1 + c t 2 + c t 3 n n n 1 d1 2 2 4 from which the equivalent torsional viscous damping coefficient is determined as
3 =
(e)
(f)
(g)
(h)
101
Chapter 3
n = c t1 + c t 2 1 n 2 nn + ct 3 1 3 n n 2 4
2
ct ,eq
(i)
3.48 The differential equations for an armature controlled dc servomotor are di (a) La a + R a i a + K b = v a dt d J eq + ct ,eq K a ia = 0 (b) dt The kinetic energy of the system at an arbitrary instant is 1 1 1 2 T = ( J r1 + J G1 )12 + ( J G 2 + J G 3 ) 2 + ( J G 4 + J r 4 ) 32 (c) 2 2 2 where the gear ratios are used to determine the angular velocities of the shafts in terms of the angular velocity of the motor shaft 1 n11 = n 2 2
n1 1 n2 n3 2 = n 4 3
2 =
(d)
nn n3 2 = 1 3 n n n4 2 4 Substitution of Equations (d) and (e) into Equation (c) and rearranging leads to 2 n1 2 nn 1 + ( J G 4 + J r 4 ) 1 3 12 T = J r1 + J G1 + ( J G 2 + J G 3 ) n n 2 2 4 n2 The equivalent moment of inertia is determined from Equation (f) as 2 n1 2 n1 n3 J eq = J r1 + J G1 + ( J G 2 + J G 3 ) n + ( J G 4 + J r 4 ) n n 2 2 4 The work done by all torsional viscous damping moments is 2 2 n1 n1 n3 W = c t 1 + c t 2 + c t 3 n n n 1 d1 2 2 4 from which the equivalent torsional viscous damping coefficient is determined as
3 =
(e)
(f)
(g)
(h)
ct ,eq
n = c t1 + c t 2 1 n 2
nn + ct 3 1 3 n n 2 4
(i)
3.49 If the shafts are supported by elastic bearings then the appropriate dependent variables are the current in the field circuit i f and the angular rotation of the shaft . The differential equations are
102
Chapter 3 di f
+ Rf if = v (a) dt d 2 J eq 2 + ct ,eq + k t ,eq K f i f = 0 (b) dt The kinetic energy of the system at an arbitrary instant is 1 1 1 2 T = ( J r1 + J G1 )12 + ( J G 2 + J G 3 ) 2 + ( J G 4 + J r 4 ) 32 (c) 2 2 2 where the gear ratios are used to determine the angular velocities of the shafts in terms of the angular velocity of the motor shaft 1 n11 = n 2 2 Lf
n1 1 n2 n3 2 = n 4 3
2 =
(d)
nn n3 2 = 1 3 n n n4 2 4 Substitution of Equations (d) and (e) into Equation (c) and rearranging leads to 2 n1 2 n1 n3 2 1 T = J r1 + J G1 + ( J G 2 + J G 3 ) n + ( J G 4 + J r 4 ) n n 1 2 2 2 4 The equivalent moment of inertia is determined from Equation (f) as 2 n1 2 n1 n3 J eq = J r1 + J G1 + ( J G 2 + J G 3 ) n + ( J G 4 + J r 4 ) n n 2 4 2 The work done by all torsional viscous damping moments is n 2 n n 2 W = ct 1 + 1 + 1 3 1 d1 n2 n2 n4 from which the equivalent torsional viscous damping coefficient is determined as n 2 n n 2 ct ,eq = ct 1 + 1 + 1 3 n2 n2 n4 The total potential energy stored in the bearings at an arbitrary instant is 2 1 n1 n1 n3 2 V = k t 1 + + 2 n 2 n 2 n4 The equivalent torsional stiffness is obtained from Equation (j) as n 2 n n k t ,eq = k t 1 + 1 + 1 3 n 2 n 2 n4
3 =
(e)
(f)
(g)
(h)
(i)
(j)
(k)
3.50 The inductors in the circuit are in parallel and can be replaced by a single inductor. The motion of the mass induces a back emf, eb in loop BCDE of the circuit. 103
Chapter 3
1 R1i1 + (i1 i2 )dt = v1 (a) C0 Application of KVL to loop BCDE leads to t 1 R2 i2 + Leq i2 + eb + (i2 i1 )dt = 0 (b) C0 Application of Newtons second law to the free-body diagrams of the blocks, drawn at an arbitrary instant, leads to & & F kx1 cx1 + k ( x 2 x1 ) = m1 x1
& & m1 & + cx1 + 2kx1 kx 2 = F x1 & k ( x 2 x1 ) = m2 & x2 & m2 & kx1 + kx2 = 0 x2 3.51 Consider n resistors in parallel as illustrated below (c) (d)
KVL applied to each loop in the circuit shows that the voltage change across each resistor is the same (a) v = v1 = v 2 = = v n The sum of the currents through each resistor is equal to the current entering the parallel combination (b) i = i1 + i2 + i3 + + in The voltage change across each resistor is given by Ohms law
104
Chapter 3
v k = v = i k Rk v (c) Rk If the parallel combination were to be replaced by a single resistor of resistance Req then ik =
Substitution of Equations (c) and (d) into Equation (b) leads to v v v v = + + + Req R1 R2 Rn Equation (e) is rearranged leading to 1 Req = 1 1 1 + + + R1 R2 Rn = 1 1 R k =1 k
n
(e)
(f)
KVL applied to each loop in the circuit shows that the voltage change across each capacitor is the same (a) v = v1 = v 2 = = v n The sum of the currents through each capacitor is equal to the current entering the parallel combination (b) i = i1 + i2 + i3 + + in The voltage change across each capacitor is given by t 1 vk = v = ik Ck 0
ik = C k dv dt (c)
105
Chapter 3 If the parallel combination were to be replaced by a single capacitor of capacitance C eq then
v= 1 idt C eq 0 (d)
t
dv dt Substitution of Equations (c) and (d) into Equation (b) leads to dv dv dv dv = C1 + C2 + + Cn C eq dt dt dt dt Equation (e) is rearranged leading to C eq = C1 + C 2 + + C n i = C eq
(e)
= Ck
k =1
(f)
106
KVL applied to each loop in the circuit shows that the voltage change across each inductor is the same (a) v = v1 = v 2 = = v n The sum of the currents through each inductor is equal to the current entering the parallel combination (b) i = i1 + i2 + i3 + + in The voltage change across each inductor is given by di v k = v = Lk k dt t 1 ik = vdt (c) Lk 0 If the parallel combination were to be replaced by a single inductor of inductance Leq then
v = Leq i= di dt
t
1 vdt Leq 0 Substitution of Equations (c) and (d) into Equation (b) leads to t t t t 1 1 1 1 idt = idt + idt + idt Leq L1 0 L2 Ln 0 0 0 Equation (e) is rearranged leading to 1 Leq = 1 1 1 + + + L1 L2 Ln
(d)
(e)
1 1 L k =1 k
n
(f)
107
Chapter 4
= h sin(20) + 0.0015 = 0.342h + 0.0015 Also the distance from the tip of the truncated cone to the hole is = r2 cos
(a)
= 0.0015 cos(20) = 0.00141 m (b) The volume of a right cone of height z and base radius r is 1 V = r 2 z (c) 3 Consider a control volume of the clock. Let z be the instantaneous height of the column of liquid in the control volume. Assume both ends of the control volume are open to the atmosphere and that the flow rate of air into the control volume is much smaller than the flow rate of the water leaving the control volume. At a given instant of time the volume of liquid in the clock is 1 1 V ( z ) = r 2 ( z ) z r12 3 3
= =
[0.117 z 3
(d)
The velocity of the liquid leaving the hole is approximated using Toricellis law as v = 2gz Thus the volumetric flow rate of the liquid leaving is
(e)
107
Chapter 4
Q = vA m 2 = 2 9.81 2 z (0.0015) s
= 3.13 x10 5 z (f) The application of the control volume form of Conservation of Mass to the clock is of the form: Rate of change of liquid leaving the control volume =Rate of change of volume of liquid in the control volume dV Q= dt d 3.13 x10 5 z = 0.117 z 3 + 1.03 x10 3 z 2 + 2.25 x10 6 z 3.17 x10 9 dt 3 dz 3 5 z = 0.351z 2 + 2.03 x10 3 z + 2.25 x10 6 (g) 3.13 x10 dt Equation (g) is rearranged to dz 3 x10 5 = (0.351z 1.5 + 2.03 x10 3 z 0.5 + 2.25 x10 6 z 0.5 ) (h) dt Separation of variables in Equation (h) leads to 3 x10 5 dt = 0.351z 1.5 + 2.03 x10 3 z 0.5 + 2.25 x10 6 z 0.5 dz (i) Integration of Equation (i), requiring that the clock drains in 3600 s leads to
) (
3600
3x10 t
z =h z =0
0.108 = 0.141z + 1.35 x10 z + 1.13x10 z Equation (j) can be written as a 5th order polynomial in z 0.5 as
5 3
(j)
0.141(z 0.5 ) + 1.35 x10 3 (z 0.5 ) + 1.13x10 6 z 0.5 0.108 = 0 (k) The following MATLAB workspace is used to determine the roots of the polynomial >> clear >> p=[0.141 0 1.35e-3 0 1.13e-6 0 -0.108] p= 0.1410 0 0.0014 0 0.0000 0 -0.1080
108
Chapter 4 0.4774 + 0.8298i 0.4774 - 0.8298i -0.4774 + 0.8298i -0.4774 - 0.8298i -0.9549 >> aa(1)^2 ans = 0.9118 >> Thus the water clock will drain in one hour if
h = 0.912 m
(l)
Application of conservation of mass to the control volume leads to dm & min = dt where the total mass in the control volume is m = V
(a)
4 = r 3 (b) 3 where r is the radius of the balloon. Substituting Equation (b) into Equation (a) leads to 4 d 3 & min = r 3 dt 4 dr = r 2 (c) 3 dt Rearrangement of Equation (c) gives & dr 3min = (d) r2 dt 4 Integration of Equation (d) with respect to time is
( )
109
Chapter 4
r
2 r dr = 0
& 3min dt 4 0
t
& r 3 3min = t 3 4 Equation (e) is rearranged to solve for the time required to achieve a radius r as 4r 3 t= & 9min The time required to achieve a radius of 10 cm is kg 3 4 0.169 3 (0.1 m ) m t= kg 9 0.1 s 3 = 2.36 x10 s
(e)
(f)
(g)
4.3 Application of Bernoullis equation from the surface of the reservoir to the exit of the reservoir assuming no losses h1 = h2
2 p2 v2 p1 v12 + +z = + +z g 2 g 1 g g 1
(a)
The pressure at the surface of the reservoir is given as p1 = 200 kPa , the exit jet is open to the atmosphere, thus p1 = 0 . Assuming the reservoir is large v1 0 . Taking the datum at the tank exit leads to z1 = 40 m and z 2 = 0 . The density of water is taken kg as = 1000 3 . Substitution into Equation (a) leads to m
110
kg m m 2 9.81 2 1000 3 9.81 2 m s s m v 2 = 34.4 (b) s The water in the jet has a horizontal velocity of 34.4 m/s as it leaves the tank. The range of the jet is obtained by applying the projectile motion equations to the jet. Let x and y be the position of a fluid particle as it leaves the tank at t=0. The projectile motion equations are 1 y = gt 2 (c) 2 x = vt (d) The water hits the surface when y = 10 m . The time at which the fluid particle hits the surface is obtained using Equation (c) as
2y 2 t = g 2 2( 10 m ) = m 9.81 s 2 = 1.43 s The range of the fluid jet is obtained using Equation (d) as m x = 34.4 (1.43 s ) s = 49.1 m
1 1
+ 40 m =
2 v2
(e)
(f)
4.4 Application of extended Bernoullis equation between the surface of the reservoir (1) and the exit of the pipe (2) leads to h1 = h + h2 (a) Choosing the datum at the free surface of the reservoir and assuming the velocity at the surface is small, the head at the surface of the reservoir is p1 v12 h1 = + +z g 2 g 1 =0 (b) The head at the exit of the pipe is
111
Chapter 4
2 p2 v2 + + z2 g 2 g 2 v2
h2 =
= 0 Pa +
m 2 9.81 2 s 2 = 0.0510v 2 40
+ ( 40 m )
(d)
(d)
v 2 = 23.8
m s
(e)
4.5 Application of extended Bernoullis equation between the surface of the reservoir (1) and the exit of the pipe (2) leads to h1 = h + h2 (a) Choosing the datum at the free surface of the reservoir and assuming the velocity at the surface is small, the head at the surface of the reservoir is p1 v12 h1 = + + z1 g 2 g N 5 x10 4 2 m = kg m 1000 3 9.81 2 m s = 5.10 m (b) The head at the exit of the pipe is p v2 h2 = 2 + 2 + z 2 g 2 g
= 0 Pa +
2 v2
m 2 9.81 2 s 2 = 0.0510v 2 40
+ ( 40 m )
(d)
112
Chapter 4 v 2 L h = f 2 2 g D
2 26 m = (0.003)(0.0510v 2 ) 0.2 m 2 = 0.0199v 2 Substitution of Equations (b)-(d) into Equation (a) leads to 2 2 5.10 = 0.0510v 2 40 + 0.0199v 2
(d)
v 2 = 25.2
m s
(e)
4.6 Application of extended Bernoullis equation between the surface of the reservoir (1) and the exit of the pipe (2) leads to (a) h1 = h + h2 Choosing the datum at the free surface of the reservoir and assuming the velocity at the surface is small, the head at the surface of the reservoir is p v2 h1 = 1 + 1 + z1 g 2 g =0 (b) The head at the exit of the pipe is p v2 h2 = 2 + 2 + z 2 g 2 g
= 0 Pa +
2 v2
m 2 9.81 2 s 2 = 0.0510v 2 40
+ ( 40 m )
(d)
(d)
2 40 v2 = 6.63 f + 0.0510
(e)
113
Chapter 4 The friction factor is unknown as is the velocity. They are related from the Colebrook formula on which the Moody diagram is based. The relative roughness ratio is given as = 0.002 . An iterative solution using the Moody diagram is required. The procedure is D 1. Guess a value of the velocity vD kg , assuming = 1000 3 and 2. Calculate the Reynolds number Re = m N s = 1.31x10 3 2 . In terms of the velocity the Reynolds number is calculated as m kg 1000 3 v(0.2 m ) m Re = N s 1.31x10 3 2 m 5 (f) = 1.53x10 v 3. Determine a friction factor from the Moody diagram (Fig. 4.5) 4. Calculate the velocity using Equation (e) 5. Compare the calculated velocity to the guessed value. Use the calculated value as the new initial guess, continuing the procedure until convergence is achieved. The procedure, using an initial guess of v 2 = 20 Iteration 1 2 Guess for v2 (m/s) 20 14.02
m is summarized in the Table below. s Re (Eq. f) v2 (m/s) (Eq.e) f (Moody diagram) 6 0.023 14.02 3.06 x10 6 0.023 14.02 2.15 x10
4.7 The velocity of the flow through the pipe is calculated in terms of the diameter of the pipe as Q v= A m3 0 .2 s =
D2 4 0.255 = D2
(a)
114
Chapter 4 Application of extended Bernoullis equation from the pressure head to the flow exit, assuming they are at the same elevation and the exit pressure is atmospheric gives h1 = h2 + h 0.255 0.255 2 2 D + f 10 m D 18 m = m D 2 9.81 m 2 9.81 2 s s2 3.31x10 3 3.31x10 2 f 18 = + (b) D4 D5 Since the friction factor is a function of the Reynolds number, which is a function of the diameter, vD Re =
2 2
kg 0.255 1000 3 (D ) m D 2 = N s 1.31x10 3 2 m 5 1.95 x10 = (c) D an iterative process using the Moody diagram is necessary to solve for the required pipe diameter. The process is to guess a diameter, calculate the Reynolds number using Equation (c), use the Moody diagram to determine the friction factor for a smooth pipe, and use Equation (b) to calculate a new diameter. Use this new diameter as an initial guess for the subsequent iteration. The process is summarized in the table below using an initial guess of 10 cm. Iteration 1 D (m) 0.1 Re (Eq. c)
1.95 x10 6 f (Moody diagram) 0.0105
0.0109
4.8 Before the rocket is filled with water it contains air at atmospheric pressure. The total mass of the air is m a = aV a
= a D 2 L
The ideal gas law implies
p 0 = a Ra T
(a)
a =
p0 Ra T
(b)
115
Chapter 4 where p 0 is the atmospheric pressure. Substitution of Equation (b) into Equation (a) leads to p 0D 2 L ma = (c) Ra T As the rocket is being filled with water, the mass of the air remains constant. Its density is calculated as ma ( h) = Va ( z )
ma (d) D ( L h) (a) Assuming the process is isothermal use of the ideal gas law and Equation (c) in Equation (d) leads to p 0D 2 L p ( h) = Ra T Ra TD 2 ( L h) =
2
L p0 (e) Lh (b) The pressure on the surface when the water level is z is obtained by replacing h by z in Equation (e) leading to L p0 (f) p( z ) = Lz (c) Let 1 represent the free surface of the water and 2 represent the exit of the rocket. Assume acceleration of the particles on the surface of the water is negligible. Application of Bernoullis equation leads to p1 v2 p v2 + 1 + z1 = 2 + 2 + z 2 1 g 2 g 2 g 2g p ( h) =
2 p0 v2 1 dz L p0 + + +z= L z g 2 g dt g 2 g Equation (g) is rearranged as
1
(g)
2 z p 0 dz 2 (h) + + 2 gz v 2 = 2 L z dt Equation (h) is a nonlinear first-order differential equation whose solution is z(t). The solution must satisfy z ( 0) = h (i) (d) Consider a control volume consisting of the rocket. The principle of conservation of mass, assuming the water is incompressible, is written as: rate at which water leaver the control volume through its exit=rate at which water accumulates in rocket
116
Chapter 4
v 2d 2 = d D 2 z dt
)
2 1 2
dz d z p 0 dz + + 2 gz = 2 dt D L z dt
(h)
4.9 Application of extended Bernoullis equation from the free surface of the tank to the exit of the pipe assuming the velocity of the fluid on the surface of the reservoir is small leads to h1 = h2 + h (a) where p1 v12 + + z1 h1 = g 2 g = 0 + 0 + 20 m (b)
h2 =
2 p2 v2 + + z2 g 2 g 2 v2
m 2 9.81 2 s 2 = 0.510v 2
= 0+
+0
(c)
The head loss through the pipe is v L h = f 2 2 g D 2 60 m = f (0.0510v 2 ) 0.3 m 2 = 15.3 fv 2 Substitution of Equations (b)-(d) into Equation (a) leads to 2 2 0 = 0.0510v 2 40 + 6.63 fv 2
2 1
(d)
2 20 (e) v2 = 15.3 f + 0.0510 The friction factor is unknown as is the velocity. They are related from the Colebrook formula on which the Moody diagram is based. The relative roughness ratio is calculated as 1.5 x10 4 m = D 0.3 m = 5 x10 4 (f) An iterative solution using the Moody diagram is required. The procedure is 1. Guess a value of the velocity
117
Chapter 4 Re =
vD , assuming
= 1000
kg and m3
= 1.31x10 3
N s . In terms of the velocity the Reynolds number is calculated as m2 kg 1000 3 v 2 (0.3 m ) m Re = N s 1.31x10 3 2 m 5 = 2.29 x10 v 2 (g)
3. Determine a friction factor from the Moody diagram (Fig. 4.5) 4. Calculate the velocity using Equation (e) 5. Compare the calculated velocity to the guessed value. Use the calculated value as the new initial guess, continuing the procedure until convergence is achieved. The procedure, using an initial guess of v 2 = 20 Iteration 1 2 Guess for v2 (m/s) 20 8.11
m is summarized in the Table below. s Re (Eq. g) v2 (m/s) (Eq.e) f (Moody diagram) 6 0.0165 8.11 4.58 x10 6 0.017 8.02 1.86 x10
m 4 = 8.02 (0.3 m ) s 4 m3 = 0.567 (h) s (b) Since the Reynolds number is much greater than 2300, the flow is turbulent and the resistance in the pipe is h R=2 Q
20 m = 2 m3 0.567 s s = 70.6 2 m
(i)
118
Chapter 4 4.10 The pipes resistance is calculated as R=2 The head loss across the pipe is
v 2 L h = f 2g D
2 m3 0.075 s (0.3 m )2 30 m 4 = f 2 9.81 m 0.3 m s2
h Q
(a)
(b)
m3 0.075 kg s (0.3 m) 917 3 2 m ( 4 0.3 m ) = N s 0.290 2 m 3 = 1.01x10 (c) The friction factor is determined using the smooth pipe curve on the Moody diagram as f = 0.07 . Thus from Equation (a) (5.74)(0.07) m R=2 m3 0.075 s s = 10.71 2 (d) m
4.11 The tanks are operating at steady state. Assuming turbulent flow, the resistance in a pipe is h R=2 (a) q
119
Chapter 4 The head loss in the first pipe is the difference in the levels between the two tanks, h1 = (15 m ) (8 m ) = 7 m and the flow rate through the pipe is the flow rate into the first tank, q1 = 1.8
m3 . Thus s
7m s = 7.78 2 (b) 3 m m 1.8 s The head loss in the second pipe is equal to the level of the liquid in the second tank, h2 = 8 m . The flow rate through the second pipe is the sum of the flow rates into the R1 = 2
(c)
4.12 The three pipes are in parallel, thus their equivalent resistance is the sum of the reciprocals of the individual resistances 1 Req = 1 1 1 + + R1 R2 R3
= 1 7.0 + 1 1 + 1
s s s 4.0 2 5.0 2 2 m m m s = 1.69 2 (a) m The head loss across each pipe is the same for all pipes in a parallel combination while the total flow rate is the sum of the individual flow rates, 1 1 1 h = R1Q1 = R2 Q2 = R3 Q3 (b) 2 2 2 m3 Q = 4 .9 = Q1 + Q2 + Q3 (c) s The flow rate is related to the equivalent resistance through
120
Chapter 4
Req = 2 h= h Q
1 Req Q 2 1 s = 1.69 2 2 m
m3 4.9 s (d)
(e)
(f)
(g)
4.13 The appropriate differential equation is dh 1 (a) A + h = q p (t ) dt R The steady-state resistance is obtained as h 15 m s = 12.5 2 (b) R=2 s =2 3 qs m m 2 .4 s From the given information the perturbation in flow rate is q p (t ) = 0.2u (t ) . Thus Equation (1) becomes
121
Chapter 4
1 dh h = 0.2u (t ) + dt 12.5 dh 650 + 0.08h = 0.2u (t ) dt 650
(c)
4.14 The mathematical model for the two-tank system is dh 1 1 A1 1 + h1 h2 = q p1 (t ) dt R1 R1 1 dh2 1 1 h1 + + R R h2 = q p 2 (t ) dt R1 2 1 The resistances are calculated using the steady-state values h h2 s R1 = 2 1s q1s A2 9.7 m 6.8 m m3 4.0 s s = 1.45 2 m h2 s R2 = 2 q1s + q 2 s =2 6.8 m m3 m3 4.0 +0 s s s = 3.4 2 m =2 The perturbation in flow rates are
q p1 (t ) = 0 q p 2 (t ) = 0.2u (t ) m3 s
(a) (b)
(c)
(d)
(e) (f)
Substitution of Equations (c)-(f) in Equations (a) and (b) leads to dh 1 1 450 1 + h1 h2 = 0 dt 1.45 1.45 dh 1 1 1 h1 + 790 2 + h2 = 0.2u (t ) dt 1.45 1.45 3.40 Equations (g) and (h) are rearranged as
(g)
(h)
122
Chapter 4
dh1 + 0.690h1 0.690h2 = 0 dt dh 790 2 0.690h1 + 0.984h2 = 0.2u (t ) dt 450
(i) (j)
4.15 The mathematical model for the two-tank system is dh 1 1 A1 1 + h1 h2 = q p1 (t ) dt R1 R1 A2 The resistances are calculated by R1 = 2 R2 = 2 h 1 q1s h2 1 dh2 1 1 h1 + + R R h2 = q p 2 (t ) dt R1 2 1
(a) (b)
(c)
(d) q1s The head loss in the first pipe is obtained by application of extended Bernoullis equation between the free surface of the reservoir and the exit of the pipe h1 = h2 + h1 N m2 + 13 m = 15 m + h1 kg m 1000 3 9.81 2 m s h1 = 4.12 m 6 x10 4 Substitution of Equation (e) into Equation (c) leads to
(e)
R1 = 2
(f)
(g)
123
Chapter 4
1 1 h1 h2 = 0.6u (t ) dt 2.06 2.06 4 (1.8 m )2 dh2 1 h1 + 1 + 1 h2 = 0 dt 2.06 4 2.06 7.5 Equations (h) and (i) are simplified to dh 3.14 1 + 0.485h1 0.485h2 = 0.6u (t ) dt dh 2.54 2 0.485h1 + 0.619h2 = 0 dt
(2 m )2 dh1 +
(h) (i)
(j) (k)
4.16 Application of conservation of mass to each tank leads to an equation of the form dH i Ai = Qi ,in Qi ,out (a) dt With the exception of the first tank both the inlet and outlet flow rates are related to the pipe resistances. It is shown in the text that linearization of the nonlinear equations obtained through application of Equation (a) is consistent with approximating the change in head across a pipe and the flow rate through the pipe by h R= (b) Q where R is the resistance of the pipe at steady state. Consider the piping system connecting the three tanks. Let Q1 , Q2 and Q3 be the flow rates through the pipes. Then from conservation of mass (c) Q1 = Q2 + Q3 Defining hm as the head at the junction of the three pipes, the resistances are related to the flow rates and perturbations in level by h hm R1 = 1 (d) Q1 R2 = R3 = hm h2 Q2 (e)
hm h3 (f) Q3 Solving Equations (d)-(f) for the flow rates and substituting into Equation (c) leads to
124
Chapter 4
h1 h m h h 2 h m h3 = m + R1 R2 R3 1 h3 h2 1 1 h1 hm R + R + R = R + R + R 2 3 1 2 3 1 1 h1 + h 2 + h3 hm = 1 1 1 R1 R 2 R 3 + + R1 R 2 R 3
Then Q1 = h1 hm R1
(g)
1 1 h1 + h2 + h3 = h1 1 1 1 R1 R2 R3 R1 + + R1 R2 R3 h1 h2 h1 h3 1 = + R1 R2 + R2 R3 + R3 R1 R3 R2 In a similar fashion
Q2 = Q3 = h1 h2 h3 h2 1 + R1 R2 + R2 R3 + R3 R1 R3 R1
(h)
(i)
h1 h3 h2 h3 1 + (j) R1 R2 + R2 R3 + R3 R1 R2 R1 Using Equations (h)-(i) the appropriate linear equations for the liquid levels in the tanks are h1 h2 h1 h3 dh 1 = qi (k) A1 1 + + dt R1 R2 + R2 R3 + R3 R1 R3 R2
A2 A3
h1 h2 h3 h2 dh2 h2 1 =0 + + dt R5 R1 R2 + R2 R3 + R3 R1 R3 R1 h1 h3 h2 h3 dh3 h3 1 =0 + + dt R4 R1 R2 + R2 R3 + R3 R1 R2 R1
(l) (m)
4.17 If the process is isentropic then the pressure and density are related by p = C (a) where C is a constant and is the ratio of specific heats. The constant is evaluated from the initial state 125
Chapter 4
p 0 = C 0
C=
p0
0
(b) (c)
The ideal gas law relates pressure, temperature, and density by p = Ra T Substitution of Equation (c) into Equation (a) leads to p Ra T = 0
0 Ra
p0
1
T (d)
1
= BT
where
(e)
p0
0
= DT (f) The mass of the air in the pressure vessel is as given by Equation (d) in the solution of Example 4.10, ma = V (g) Use of Equation (d) in Equation (g) leads to
1
ma = BVT (h) As in Example 4.10 the mass flow rate into the vessel is related to the resistance, pressure, and upstream pressure by p p & (i) m= 0 R Substitution of Equation (f) into Equation (h) leads to
p 0 DT R The control volume form of conservation of mass for the pressure vessel is & m=
1
(j)
126
BVT dT p0 DT = 1 dt R
2 1
(k)
4.18 The mass of the air in the pressure vessel is as given by Equation (d) in the solution of Example 4.10, ma = V (a) As in Example 4.10 the mass flow rate into the vessel is related to the resistance, pressure, and upstream pressure by p p & (b) m= 0 R The control volume form of conservation of mass for the pressure vessel is dma & =m dt p R a T d (V ) = 0 dt R p d R a T + = 0 (c) dt RV RV 4.19 The appropriate mathematical model is that of Equation (h) of Example 4.10 V dp p p s + = R g T dt R R RV dp + p = ps R g T dt The volume of the tank is 4 V = r 3 3 4 3 = (2 m ) 3 = 33.5 m 3 The resistance is determined from the given information by (a)
(b)
127
Chapter 4
ps p & m 10 x10 3
R=
(c)
Thus 1 4 2 1.25 x10 (33.5 m ) RV ms = Rg T m2 (313 K ) 2.077 2 s K = 6.44 x10 2 s Substituting Equation (d) into Equation (a) leads to dp 6.44 x10 2 + p = 10 x10 3 dt
(d)
(e)
4.20 Application to conservation of mass to the control volume containing the tank dm & & (a) = min mout dt where the mass of the gas in the chamber is m = V and from the ideal gas law, p = . Since the process is isothermal T is constant and Rg T
dm V dp = (b) dt R g T dt Application of extended Bernoullis equation between the exit of the tank and the end of the pipe leads to p p = e + hl (c) g g Equation (c) is rearranged to p p e = ghl (d) Assuming laminar flow
128
Chapter 4
64 v 2 L hl = Re 2 g D 64 v 2 L = vD 2 g D 32 L = v gD 2 Substitution of Equation (e) into Equation (d) leads to
(e)
(f)
Substitution of Equations (b) and (f) into Equation (a) leads to D 4 V dp & ( p pe ) = min R g T dt 128L
(g)
129
Chapter 4 Summation of moments about the pin support, noting that the bar is massless leads to MO = 0
F (b) F1 (a) = 0 b F a Summation of forces on the free-body diagram of the piston of mass m1 leads to F = m1a F1 =
b & F pA1 = m1 & y a & y b F m& p = A A a 1 1
(a)
(b)
where y is the displacement of the piston. Summation of forces on the free-body diagram of the piston of mass m2 leads to & & pA2 kx cx = m2 & x (c) Substitution of Equation (b) into Equation (c) leads to A FbA2 & & m2 &+ m1 2 &+ cx + kx = x y & (d) A1 aA1 Assuming the hydraulic liquid is incompressible the volume of the air displaced by each piston must be equivalent. Thus A1 y = A2 x A y= 2 x (e) A1 Use of Equation (e) in Equation (d) leads to 2 bA A2 & m2 + m1 &+ cx + kx = 2 F (t ) (f) x & aA A 1 1
130
Chapter 4
Noting that the pressure in the hydraulic chamber is constant and the pistons are massless, summing forces on each of the pistons leads to F = 0
F (t ) pA1 = 0 pA2 + F1 = 0 A2 F (t ) (a) A1 Summation of moments about the pin support applied to the free-body diagram of the bar leads to M O = I O F1 =
F1 a + F2 b = I O (b) The mass moment of inertia of the bar about O, assuming a>b is obtained using Table 2.1 and the parallel axis theorem 2 a+b 1 m m 2 I O = (a + b) + a 12 2 2 2
1 m 1 = (a 2 2ab + b 2 ) + (a 2 + 2ab + b 2 ) 12 2 4 m = (a 2 ab + b 2 ) (c) 6 Assuming small displacements the angular acceleration of the bar is related to the acceleration of the block by
131
Chapter 4 & b = & x & & x = b Substitution of Equations (a), (c), and (d) in Equation (b) leads to A2 & x m & F (t ) a + F2 b = (a 2 ab + b 2 ) A 6 b 1
(d)
A a m & (e) x F2 = 2 a 2 ab + b 2 & 2 F (t ) + 6b A1b Application of Newtons second law to the free-body diagram of the block, drawn at an arbitrary instant leads to & & F2 cx kx = m& x (f) Substituting Equation (e) in Equation (f) and rearranging leads to 2 A2 a m a a & x & (g) + 7 &+ cx + kx = A b F (t ) 6 b b 1
4.23 The potential energy of the system at an arbitrary instant is simply 1 V = kx 2 (a) 2 The kinetic energy of the system is the sum of the kinetic energies of the block, the bar and the two pistons. Let represent the angular rotation of the bar, y the displacement of the piston of mass m1 and z the displacement of the piston of mass m2 . The kinetic energy of the system is 1 1 & 1 1 & & & T = mx 2 + I o 2 + m1 y 2 + m2 z 2 (b) 2 2 2 2 The mass moment of inertia of the bar about O, assuming a>b is obtained using Table 2.1 and the parallel axis theorem 2 a+b 1 m m 2 I O = (a + b) + a 12 2 2 2 1 m 1 = (a 2 2ab + b 2 ) + (a 2 + 2ab + b 2 ) 12 2 4 m = (a 2 ab + b 2 ) 6
(c)
The displacement of the end of the bar which is a distance b from the support is equal to the displacement of the block of mass m. Assuming small displacements b = x x = (d) b
132
Chapter 4 The displacement of the end of the bar which is a distance a from the pin support is equal to the displacement of the piston of mass m2 . Assuming small displacements and using Equation (d) z = a a = x (e) b Conservation of mass applied to the fluid reservoir implies that the volumetric flow rate is constant in the reservoir. To this end, & & A1 y = A2 z & y= A2 & z A1 a x b &
2 2
(f)
& 1 1m 1 Aa 1 a x & & T = mx 2 + a 2 ab + b 2 + m1 2 x + m2 x Ab & 2 2 6 2 1 2 b b Equation (g) is simplified to 2 2 A2 a m a2 a 1 a 2 & T = m + 2 + 1 + m1 A b + m2 b x b 2 6 b 1 The work done by the viscous damping force is
(g)
(h)
& W = cxdx
x1
x2
(i)
A a = F (t ) 2 dx Ab 1 x1 Application of the energy method leads to a differential equation of the form & & meq &+ c eq x + k eq x = Feq (t ) x where the equivalent mass is determined from the kinetic energy, Equation (g), as
x2
(j)
(k)
A a m a2 a a meq = m + 2 + 1 + m1 2 + m 2 b Ab 6 b b 1 The equivalent stiffness is determined using the potential energy, Equation (a), as k eq = x
(l)
(m)
The equivalent damping coefficient is determined using the work done by the viscous damping force as c eq = c (n)
133
A a Feq (t ) = F (t ) 2 Ab 1
(o)
4.24 Let w(t) represent the displacement of the particle to which the spring and viscous damper are attached, as illustrated in the figure below.
z+w y+w = a+b b bz a + b w= y a a Summing forces to zero at the end of the walking beam leads to & cw + k ( w x) = 0 The motor equation is & Cy = A p x
(a)
(b)
(c)
(d)
Substitution of Equation (c) into Equation (a) leads to bz a + b A p & w= x a a C Use of Equation (d) in Equation (b) leads to bz a + b A p bz a + b A p & & + k & & x kx = 0 x c a a C a C a a + b A p c a C a + b A p &+ k & x a C cb kb x + kx = & z+ z & a a
(e)
134
Chapter 4 4.25 Let z(t) represent the displacement of the point on the walking beam connected to the valve, w(t) represent the displacement of the end of the walking beam, and q(t) represent the displacement of the rod. Assume small displacements.
Using similar triangles on the walking beam y+w z+w = + a a + a a z w= y a Using similar triangles on the bar leads to q x = a+ a a q= x a+ The servomotor equation is & Cz = A p x
(a)
(b)
z=
Use of Equation (c) in Equation (a) leads to
w=
Ap & x C
(c)
+ a Ap a & x y a C Summing forces to zero at the end of the walking beam leads to & & c ( w q ) + kw = 0 Substitution of Equations (b) and (d) in Equation (e) leads to a + a Ap + a a a & & & & x x + k y x =0 c y a+ a C
(d)
(e)
(f)
Chapter 4
4.26 Consider a control volume consisting of the plate. The control volume form of conservation of energy applied to the plate is Rate at which energy accumulates in the control volume=Rate at which energy enters the control volume through its boundary rate at which energy leaves the control volume through its boundary. The rate of accumulation of energy is the rate of change of internal energy, dU dT = wdc p (a) dt dt Energy is transferred into the control volume due to the friction between the plate and the surface. The rate of energy dissipation due to friction is x & d W = (wdgx )dx dt 0 = wdgV (b) Energy is transferred from the control volume due to heat transfer with the ambient. Using Newtons law of cooling & Q = hwd (T T0 ) (c) Thus application of Conservation of Energy to the control volume leads to dT = wdgV hwd (T T0 ) wdc p dt dT c p + hT = gV hT0 (d) dt 4.27 A control volume consisting of the aquarium is illustrated below.
The control volume form of conservation of energy applied to this control volume is Rate at which energy enters the control volume- rate at which energy leaves the control volume +rate at which heat is transferred to the control volume=rate at which energy accumulates in control volume
& Rate at which energy enters control volume = min c p Tin
136
Chapter 4
& Rate at which heat is transferred to the control volume = Q Rate at which energy accumulates in control volume = Vc p
dT where T is the dt temperature of the water in the aquarium under the lumped property assumption
& & Application of the control volume form of conservation of mass leads to min = mout It is assumed that the water in the aquarium is well mixed such that Tout = T Thus the energy equation is written as
Vc
(a) In the steady state
(a)
(b)
The mass flow rate is calculated as m3 kg & m = 1.03 x10 3 3 1.8 s m kg = 1.85x10 3 s Substitution into Equation (a) leads to & & Q = mc p (T Tin )
kg J 3 (30 C 20 C ) = 1.85 x10 3 4.0 x10 s kg C MJ = 74.2 s &= 0 and Equation (a) becomes (b) If the heater stops working, Q dT & & Vc p + m c p T = m c p T in dt Substitution of given values leads to
(c)
(d)
(e)
137
Chapter 4
3 kg 600 m 2 1.03 x10 3 m
) 4.0 x10
kg J 3 (293 K ) = 1.85 x10 3 4.0 x10 s kg K dT 2.47 x10 9 + 7.4 x10 6 T = 2.17 x10 9 dt Equation (f) is supplemented by the initial condition T (0) = 303 K 4.28 A control volume consisting of the aquarium is illustrated below.
(f)
(g)
The control volume form of conservation of energy applied to this control volume is Rate at which energy enters the control volume- rate at which energy leaves the control volume +rate at which heat is transferred to the control volume=rate at which energy accumulates in control volume
& Rate at which energy enters control volume = min c p Tin & Rate at which energy leaves control volume= mout c p Tout
& Rate at which heat is transferred to the control volume = Q Rate at which energy accumulates in control volume = Vc p
dT where T is the dt temperature of the water in the aquarium under the lumped property assumption
& & Application of the control volume form of conservation of mass leads to min = mout It is assumed that the water in the aquarium is well mixed such that Tout = T Thus the energy equation is written as
Vc
(a)
138
Chapter 4 When the cooling and filtration stop working the flow is recirculated such that the temperature of the stream entering is the temperature of the tank 15 s before. Thus the rate at energy enters the control volume is & mc p T (t 15) (b) Equation (a) is modified as dT & & & Vc (c) + mc p T mc p T (t 15) = Q dt The rate of heat transfer is equal to the rate such that the tank is in steady state when the system is working properly. This steady-state equation is & & & m c p Tin mc pT + Q = 0 (d) The mass flow rate is calculated as m3 kg & m = 1.03 x10 3 3 1.8 s m kg = 1.85x10 3 s Substitution into Equation (a) leads to & & Q = mc p (T Tin )
(e)
kg J 3 (30 C 20 C ) = 1.85 x10 3 4.0 x10 s kg C MJ = 74.2 (f) s Substitution of given and calculated values in Equation (c) leads to 3 kg (600 m 2 ) 4.0 x10 3 kg J K dT + 1.85 x10 3 kg 4 .0 x10 3 kg J K T 1 .03 x10 3 dt m s
kg J J 3 1 .85 x10 3 T ( t 15 ) = 7 .42 x10 7 4 .0 x10 s kg K s dT 2 .47 x10 9 + 7 .4 x10 6 [T T ( t 15 )] = 7 .42 x10 7 dt 4.29 A control volume consisting of the aquarium is illustrated below.
(g)
139
Chapter 4
The control volume form of conservation of energy applied to this control volume is Rate at which energy enters the control volume- rate at which energy leaves the control volume +rate at which heat is transferred to the control volume=rate at which energy accumulates in control volume
& Rate at which energy enters control volume = min c p Tin & Rate at which energy leaves control volume= mout c p Tout
& Rate at which heat is transferred to the control volume = Q Rate at which energy accumulates in control volume = Vc p
dT where T is the dt temperature of the water in the aquarium under the lumped property assumption
& & Application of the control volume form of conservation of mass leads to min = mout It is assumed that the water in the aquarium is well mixed such that Tout = T Thus the energy equation is written as
dT & & & + mc p T = Q + mc p Tin (a) dt The rate of heat transfer is equal to the rate such that the tank is in steady state when the system is working properly. This steady-state equation is & & & m c p Tin mc pT + Q = 0 (b)
Vc
The mass flow rate is calculated as m3 kg & m = 1.03 x10 3 3 1.8 s m kg = 1.85x10 3 s Substitution into Equation (b) leads to & & Q = mc p (T Tin )
(c)
kg J 3 (30 C 20 C ) = 1.85 x10 3 4.0 x10 s kg C MJ = 74.2 (d) s m3 If the flow rate suddenly changes to 1.6 a transient perturbation in temperature is s induced in the aquarium. Defining
140
Chapter 4
Q = Qs + Q p m3 m3 = 1 .8 0.2u (t ) s s (e)
(f)
and
T (t ) = Ts + T p (t )
& where Ts = 30 C = 303 K . Noting that m = Q . Assuming no resistance in the system & & min = mout . Substitution of Equations (e) and (f) into Equation (a) leads to d & c pV (Ts + T p ) + (Q s + Q p )c p (Ts + T p ) = (Qs + Q p )c p Tin + Q (g) dt dTs = 0 and using Equation (b), Equation (g) is reduced to Noting that dt dT p c pV + (Qs + Q p )c p T p = Q p c p Tin (h) dt Substitution of known values in Equation (h) leads to kg J dT kg m 3 J 4.0 x10 3 T 1.03x10 3 3 (600 m 2 ) 4.0 x10 3 + 1.03x10 3 3 1.6 kg K dt s kg K m m
kg J (293 K ) = 1.03x10 3 3 ( 0.2u (t ) ) 4.0 x10 3 kg K m dT + 6.59 x10 6 T = 2.41x10 8 2.47 x10 9 dt
k
(i)
4.30 During the reaction 2 A B 2 moles of reactant A are consumed to make 1 mole of reactant B. The rate of change of the number of moles of component A due to the reaction is dn A 2 = kVC A (a) dt reaction while the rate of change of number of moles of reactant B made during the reaction is dn B 2 = kVC A (b) dt reaction Application of the conservation of species equation to reactant A gives dC A 2 V = qC Ai qC A kVC A dt dC A 2 V + qC A + kVC A = qC Ai (c) dt Application of the conservation of species equation to reactant B gives dC B 2 V = qC Bi qC B + kVC A dt dC B 2 V kVC A + qC B = qC Bi (d) dt 141
Chapter 4 Equations (c) and (d) form the mathematical model for the system. 4.31 If the reaction A B is of order 1.5 then the rate of consumption of reactant A is dC A = kVC 1.5 (a) A dt (a) Application of the conservation of species equation to a CSTR with this reaction to species A leads to dC A V = qC Ai qC A kVC 1.5 A dt dC A V + qC A + kVC 1.5 = qC Ai (b) A dt (b) The concentration of the reactant A in the inlet stream is suddenly perturbed according to C A,i = C A,is + C A,ip (c) which leads to a perturbation in the concentration of species A in the tank according to C A = C As + C Ap (d) The steady-state condition before the perturbation occurs is obtained using Equation (b) as qC As + kVC 1.5 = qC Ais (e) As Substitution of Equations (c) and (d) into Equation (b) leads to dC Ap V + q(C As + C Ap ) + kV (C As + C Ap )1.5 = q (C Ais + C Aip ) (f) dt The nonlinear term is rewritten as 1 .5 C Ap 1 .5 1 .5 (g) kV (C As + C Ap ) = kVC As 1 + C As Since C Ap / C As << 1 the binomial expansion is used and truncated after the linear term to provide a linearization of Equation (g) of C Ap kV (C As + C Ap )1.5 kVC 1.5 1 + 1.5 As C As 1.5 0.5 = kVC As + 1.5kVC As C Ap Use of Equation (h) in Equation (f) gives dC Ap 0. V + q(C As + C Ap ) + kVC 1.5 + 1.5kVC As5 C Ap = q (C Ais + C Aip ) As dt Subtracting Equation (e) from Equation (i) leads to dC Ap 0. V + qC Ap + 1.5kVC As5 C Ap = qC Aip dt
k
(h) (i)
(j)
4.32 When the reaction A + B C occurs in a CSTR the rate at which the reactants are absorbed are 142
Chapter 4
dn A = kV 2 C A C B dt dn B = kV 2 C A C B dt dnC = kV 2 C A C B dt Application of conservation of species to each component in the CSTR leads to dC A V + qC A + kV 2 C A C B = qC Ai dt dC B + qC B + kV 2 C A C B = qC Bi V dt dC C + qC C kV 2 C A C B = qC Ci V dt
4.33 The equations for the species are unaffected by the presence of the jacket, except of course that the jacket affects the temperature. The equations for the species concentrations are the same as those of Equations (a) and (b) of Example 4.17, dC A V + (q + Ve E /( RT ) )C A = qC Ai (a) dt dC B V Ve E /( RT ) C A + qC B = qC Bi (b) dt The equation for the temperature in the reactor is modified to take into account the heat transfer between the mixture and the jacket, & 1 (T T ) Q= (c) w Rc Thus Equation (c) of Example (4.17) is modified as dT 1 & (T Tw ) + qc p Ti qc p T Q + Ve E /( RT ) C A = Vc p (d) Rc dt Consider a control volume encasing the jacket. Application of conservation of energy to the jacket leads to dT 1 1 & mw c w w = c w mw (Twi Tw ) + (T Tw ) Tw (e) dt Rc R1 + R2 + Rw In deriving Equation (e) it is assumed that the resistances are in series, m w is the total & mass of the water in the jacket and m w = q w is the mass flow rate of the water into the jacket. 4.34 Let ma (t ) be the mass of the drug that has yet to be absorbed and let k a is the rate of absorption. The rate at which the mass of the drug to be absorbed changes is dma = k a ma (a) dt Equation (a) is rearranged to
143
Chapter 4
dma + k a ma = 0 (b) dt (a) Considering the one-compartment model, application of conservation of mass to the drug in the control volume leads to Rate of accumulation =rate of absorption rate of elimination dC = k a m a k eVC V dt dC + k eVC k a ma = 0 (c) V dt Equations (b) and (c) constitute the mathematical model for the one-compartment model. (b) The two-compartment model is changed from that of Example 4.30 only by replacing the infusion term by the absorption term. The three equations are dma + k a ma = 0 (d) dt dC p Vp + (k e + k1 )V p C p k 2Vt C t k a ma = 0 (e) dt dC t Vt k1V p C p + k 2Vt C t = 0 (f) dt
144
Chapter 5
5. Laplace Transforms
5.1 Recall sinh(t ) =
1 t e e t . Thus 2 1 L{sinh(t )} = e st e t e t 2 0
)dt
= =
1 ( s ) t e dt e ( s + ) t dt 0 2 0
1 1 1 ( s ) t + e ( s + ) t (a) s e 0 0 2 s + The exponential functions in Equation (a) approach zero as t if Re(s ) > . Thus, 1 1 1 L(sinh(t ) ) = 2 s s + 1 s + (s ) = 2 ( s + )( s )
s 2
2
(b)
5.2 By definition
(a )
(b) (c)
sin(t )dt
v = g (t )
where
e st [s sin(t ) + cos(t )] (d ) g (t ) = 2 s +2 Equation (d) is obtained through integration by parts. Substitution into Equation (a) gives
st te st [s sin(t ) + cos(t )] 2e 2 [s sin(t ) + cos(t )]dt (e) F ( s ) = 2 2 s + t =0 0 s + t =
(f )
(g )
145
Chapter 5 The definition of the transform, applying integration by parts, is used in Example 5.3 leading to L {sin(t )} = 2 (h ) s +2 A similar procedure is used to show that s L {cos(t )} = 2 (i) s +2 Substitution of Equations (h) and (i) in Equation (g) leads to s s F ( s) = 2 + 2 2 2 2 2 2 2 s + s + s + s + 2s = ( j) 2 (s + 2 )2 2 5.3 Recall that L {t 2 } = 3 . Thus application of the first shifting theorem leads to s L {t 2 e 3t } = L {t 2 }
s s 3
= =
2 s3
s s 3
2 ( s 3) 3
(a )
L {cos(2t )} =
146
Chapter 5
(g )
1 the First Shifting Theorem is used to determine s2 1 L te 3t = ( s + 3) 2 Linearity and the Second Shifting Theorem are used leading to L te 3t [u (t ) u (t 2)] = L te 3t u (t ) L te 3t u (t 2)
{ }
(a)
} { te = L{ te = L{
=
3t 3t
1 e 2s e 2s e 6 2e 6 s+3 ( s + 3) 2 ( s + 3) 2
(b)
L {cos( 4t )} =
(s
s 2 16
2
+ 16)
( b)
5.7 The function is written using unit step functions as f (t ) = 20t[u (t ) u (t 1)] + 20[u (t 1) u (t 4)] = 20[tu (t ) (t 1)u (t 1) u (t 4)] Application of linearity of the transform leads to L { f (t )} = L {20[tu (t ) (t 1)u (t 1) u (t 4)]} = 20[L {tu (t )} L {(t 1)u (t 1)} L{u (t 4)}] Application of the Second-Shifting Theorem to each of the transforms lead to 1 e s e 4 s L{ f (t )} = 20 2 2 s s s
(a )
(b)
(c )
147
Chapter 5 5.8 The function of Figure P5.8 has a mathematical formulation in terms of unit step functions as F (t ) = 20t [u (t ) u (t 1)] + 20[u (t 1) u (t 4)] = 20tu (t ) 20(t 1)u (t 1) + 20u (t 4) (a) Thus, using Equation (a) and linearity of the transform, L{F (t )} = 20L{tu (t )} 20L{(t 1)u (t 1)} + 20L{u (t 4)} (b) Recall L{t} = e as 1 , L{u (t a)} = and from the Second Shifting Theorem s s2 e s L{(t 1)u (t 1)} = 2 . Thus, Equation (b) becomes s 20 20e s 20e 4 s L{F (t )} = 2 2 + s s s
(c)
5.9 Application of transform pair 1 from Table 5.1 leads to 1 L {e 2t } = (a ) s+2 Linearity of the transform shows that L {2e 2t [u (t ) u (t 3)]} = 2L {e 2t u (t )} 2L {e 2t u (t 3)} ( b) The first term on the right-hand side of Equation (c) is evaluated directly using Equation (a). Determination of the transform of the second term on the right-hand side of Equation (b) requires application of the Second Shifting Theorem. However it is not in the form for direct application of the shifting theorem. Properties of the exponential function and linearity of the transform are used to rewrite Equation (b) as L {2 e 2 t [u (t ) u (t 3) ]} = 2 L {e 2 t u (t )} 2 L {e 2 ( t 3+ 3 ) u (t 3)}
= 2 L e 2 t u (t ) 2 L e 6 e 2 ( t 2 ) u (t 3)
2t 6 2(t 2)
(c) The Second Shifting Theorem is applied to the second-term on the right-hand side of Equation (c) leading to 3 s 2 2t 6 e 2e L 2e [u (t ) u (t 3)] = (d ) s+2 s+2
{ = 2 L {e
} { u (t )} 2 e L {e
} u (t 3)}
5.10 The transform is evaluated using the convolution property. Noting that the convolution between two functions, f(t) and g(t) is defined as
f (t ) g (t ) = f ( ) g (t ) f
0
(a )
L{ f (t ) g (t )} = F ( s)G ( s)
( b)
148
Chapter 5
t L e ( t ) sin[2(t )]d = L{t}L e 2t sin(4t ) 0 4 1 = 2 ( s + 2) 2 + 16 s
4 s s + 4s + 20
2
(c )
1 L t 2 = s
(a )
1 1 (a) Note that t 2 = (t ) t 2 . Thus from the property of differentiation of the transform using Equation (a) 1 1 d L t 2 = L t 2 ds d = ds s 1 d 2 s = ds
1 = s 2 2 3
3 2
( b)
2s (b) Another application of the property of differentiation of the transform, using Equation (b) leads to
149
Chapter 5
1 3 d L t 2 = L t 2 ds 3 d 2 s = ds 2
2 ds 5 3 2 s = 2 2 3 2 s 4 (c) Use of the First Shifting Property and Equation (b) leads to 1 1 L t 2 e 3t = L t 2 s s +3
5
d 2 s
3
(c)
= 3 2 2s =
s s +3
3 2
2(s + 3)
(d )
5.12 The final value theorem is applied to the transform leading to lim x(t ) = lim sX ( s )
t s 0
(a )
150
Chapter 5 5.13 The initial value theorem is applied to the transform leading to x(0) = lim sX ( s )
s
151
Chapter 5
d L {J 1 (t )} = L ( J 0 (t )) dt = [sF ( s ) J 0 (1)] s = 1 s 2 +1 s = 1 s 2 +1
(c)
5.15 The definitions of the hyperbolic trigonometric functions are 1 cosh (t ) = (e t + e t ) (a ) 2 1 sinh(t ) = (e t e t ) ( b) 2 Substitution of Equations (a) and (b) into the transform and using linearity of the transform leads to 1 1 L{sin (t ) cosh(t ) cos(t )sinh (t )} = L (e t + e t )sin (t ) (e t e t )cos(t ) 2 2 1 = L {e t sin (t )}+ L {e t sin (t )} 2 1 L {e t cos(t )} L {e t cos(t )} (c ) 2
Applications of transform pairs from Table 5.1 and the First Shifting Theorem give
(s )2 + 2 L{e t sin (t )} = (s + )2 + 2 s L{e t cos(t )} = (s )2 + 2 s + L{e t cos(t )} = (s + )2 + 2 Substituting Equations (d)-(g) in Equation (c) leads to
L e t sin (t ) =
152
Chapter 5
L{sin (t ) cosh (t ) cos(t )sinh (t )} = 1 + 2 2 2 2 2 (s ) + (s + ) +
1 s s + 2 2 2 2 2 (s ) + (s + ) + 1 s = 2 2 2 2 2 (s ) + (s ) + 1 s + + + 2 2 2 2 2 (s + ) + (s + ) + 1 s 2 s + 2 = + 2 2 2 2 2 (s ) + (s + ) +
1 ( s 2 ) (s + ) + 2 + ( s + 2 ) (s ) + 2 = 2 (s + )2 + 2 (s )2 + 2
2 2
[ [
] ][
(h)
4 3 s 2 + 4 2
(i)
F ( s) = f (t )e st dt
0
(a)
F (s)ds = f (t )e
s s 0
st
st
dtds
(b)
F (s)ds = f (t )e
s 0 s
dsdt
= f (t ) e st ds dt s 0 s = 1 st = f (t ) e dt s=s t 0
(c)
153
Chapter 5
(d)
2 2 ds s (s 2 + 2 ) (e)
s2 + 2 = ln s2
F ( s) = f (t )e st dt
0
(a)
(b)
(c)
(d)
1 L sin t = 2 ds 2 t s s +
= tan 1 s (e)
154
t L f = a 0
Let
v=
t f e st dt a
(a)
t a t = av
(b) (c)
(d) dt = adv Changing the variable of integration in Equation (a) through introduction of Equations (b)-(d) leads to t L f = f ( v ) e asv ( adv ) a 0
= a f ( v ) e ( as ) v dv
0
= aF ( as ) Equation (e) is a statement of the change of scale property. 5.19 The property of differentiation of the transform states dx L = sX ( s ) x(0) dt Rearrangement of Equation (a) leads to dx sX ( s ) = L + x(0) dt Taking the limit of Equation (b) for large s leads to dx lim sX ( s ) = lim L + x(0) s s dt
(e)
(a)
(b)
dx lim e st dt + x(0) (c) = s dt 0 Since the integral on the right-hand side of Equation (c) exists the limiting process may be interchanged with the integration, dx lim sX ( s) = lim e st dt + x(0) (d) s s dt 0 Since the exponential functions approaches zero for large s, the limit in Equation (d) is zero leading to the statement of the Initial Value Theorem lim sX ( s ) = x(0) (e)
s
155
dF ( s ) d = f (t )e st dt ds ds 0 = f (t ) te st dt
0
= tf (t )e st dt
0
(b)
L{tf (t )} =
Use of Equation (d) and the definition of the transform in Equation (e) leads to d ( n +1) (F ( s) ) = d (1) n t n f (t )e st dt ( n +1) ds ds 0
= (1)
ds (t
d
0
f (t )e st dt
= (1) n t n f (t )(t )e st dt
0
= (1)
n +1
t
0
n +1
f (t )e st dt
n +1
L t n +1 f (t )
(g) (h)
d ( n +1) (F ( s ) ) ds ( n +1)
156
Chapter 5 Equation (c) shows that the property of differentiation of the transform holds for n=1. Equation (h) shows that if the property holds for any value of n, then it also holds for n+1. Thus the property is proved by induction. 5.21 The Laplace transform of a periodic function of period T is calculated by T 1 F ( s) = f (t )e st dt Ts 1 e 0
(a)
The function for this problem is periodic of period T = t 0 whose definition over one period is t F0 0<t< 0 2 F (t ) = (b) t0 2 F0 < t < t0 2 3 Use of Equation (b) in Equation (a) leads to t0 t0 1 2 2 st st F ( s) = F0 e dt + F0 e dt 3 1 e t 0 s t0 0 2 t0 1 F0 st t = 2 2 F0 st t =t0 e = + e t t =0 t= 0 3s 1 e t 0 s s 2 t0 t0 s 2 t 0 s 2 2 s 1 F0 2 = + e e 1 e 3 3 1 e t 0 s s t0 1 F0 5 2 s 2 t0 s (c) + e = 1 e 3 1 e t 0 s s 3
3s + 1 s2 + 9 s 1 3 = 3 2 (a ) + 2 s +9 3 s +9 The inverse transform is obtained by applying linearity and use of transform pairs from Table 5.1 s 1 1 3 x(t ) = 3L1 2 + L 2 s + 9 s + 9 3 1 = 3 cos(3t ) + sin(3t ) ( b) 3 X ( s) =
(a)
157
A=
Thus
s(3s + 1) s( s 2 + 9)
=
s =0
1 9
(b)
1 Bs + C 3s + 1 = 9+ 2 (c) 2 s s( s + 9) s +9 Evaluation of Equation (c) at s=1 leads to 4 1 1 = + (B + C ) 10 9 10 9 B + 9C = 26 (d) Evaluation of Equation (c) at s=-1 leads to 1 1 2 = + ( B + C ) 9 10 10 (e) 9 B + 9C = 28 1 Equations (d) and (e) are solved simultaneously yielding B = C = 3 . Thus 9 1 s+3 1 X ( s) = + 9 (f ) 9s s2 + 9 Linearity and transform pairs from Table 5.1 are used to invert Equation (f) leading to 1 1 x(t ) = u (t ) cos(3t ) + sin(3t ) (g ) 9 9
5.24 The denominator of the transform has poles at s=3, -3. Thus the transform has a partial fraction decomposition of the form A A 3s + 1 = 1 + 2 (a ) X ( s) = 2 s 9 s3 s+3 The residues are determined as ( s 3)(3s + 1) A1 = s2 9 s =3 3(3) + 1 5 = = ( b) 3+3 3 ( s + 3)(3s + 1) A2 = s2 9 s = 3 3(3) + 1 4 = = (c) 33 3 Thus 1 5 4 (d ) X ( s) = + 3 s 3 s + 3
158
Chapter 5 The inverse transform of Equation (d) is obtained using the transform L {e at } =
x(t ) = 1 3t (5e + 4e 3t ) 3 1 sa (e )
5.25 The poles of the transform are s = 0,3,3 . Thus a partial fraction decomposition is assumed as A A A 3s + 1 = 1+ 2 + 3 (a ) 2 s s+3 s3 s ( s 9) The residues are determined as s (3s + 1) A1 = s (s 2 9 ) s =0
3(0) + 1 1 = 09 9 ( s + 3)(3s + 1) A2 = s s2 9 s = 3 = (b)
( c)
A3 = =
s =3
3(3) + 1 5 = (3)(3 + 3) 9 Substituting Equations (b)-(d) into Equation (a) leads to 1 1 4 5 X ( s) = + 9 s s + 3 s 3 The inverse transform of Equation (d) is obtained using the transform L {e at } =
x(t ) = 1 u (t ) 4e 3t + 5e 3t 9 1 sa
(d )
( e)
(f )
5.26 A partial fraction decomposition is assumed as 3s + 1 A Bs + C = + 2 2 s ( s + 9) s s + 9 The residue, A, is determined as s (3s + 1) 1 A= = 2 s( s + 9) s =0 9 Thus 1 Bs + C 3s + 1 = 9+ 2 2 s s( s + 9) s +9
(a )
(b)
(c)
159
Chapter 5 Evaluation of Equation (c) at s=1 leads to 4 1 1 = + (B + C ) 10 9 10 9 B + 9C = 26 Evaluation of Equation (c) at s=-1 leads to
(d)
2 1 1 = + ( B + C ) 10 9 10 9 B + 9C = 28 (e) 1 Equations (d) and (e) are solved simultaneously yielding B = C = 3 . Thus 9 1 1 s + 3 (1 e 2 s ) X ( s) = + 9 (f ) 9s s2 + 9 3 s 1 It is noted that L1 = u (t ) L1 2 L1 2 = sin(3t ) . These = cos(3t ) s + 9 s + 9 s transform pairs and the second shifting theorem are used to invert the transform of Equation (f) leading to
1 1 1 1 x (t ) = cos(3t ) + sin(3t ) u (t ) cos(3(t 2) ) + sin(3(t 2)) u (t 2) 9 9 9 9 (g)
5.27 The appropriate partial fraction decomposition is s2 + 2 A B Cs + D = + + 2 2 ( s + 3)( s + 1)( s + 16) s + 3 s + 1 s + 16 Residues are used to obtain s2 + 2 A= ( s + 1)( s 2 + 16) s = 3
(3) 2 + 2 (3 + 1)[(3) 2 + 16] 13 = 50 =
(a)
(b)
B= =
s2 + 2 ( s + 3)( s 2 + 16)
2
s = 1
(c)
160
Chapter 5 Thus Equation (a) becomes s2 + 2 13 1 3 1 Cs + D = + + 2 50 s + 3 34 s + 1 s 2 + 16 ( s + 1)( s + 3)( s + 16) Evaluation of Equation (d) at s=0 gives 2 13 1 3 1 D = + + (1)(3)(16) 50 3 34 1 16 D = 1.283 Evaluation of Equation (d) at s=2, using Equation (e) leads to 6 13 1 3 1 2C + 1.283 = + + (3)(5)(20) 50 5 34 3 20 C = 0.2621 Thus 13 1 3 1 0.2621s + 1.283 X (s) = + + 50 s + 3 34 s + 1 s 2 + 16 Inversion of Equation (g) gives x(t ) = 0.26e 3t + 0.0882e t + 0.2621cos(4t ) + 0.3208 sin(4t )
(d)
(e)
(f) (g)
(h)
5.28 Completing the square of the denominator leads to 2s + 3 X ( s) = 2 s + 6s + 13 2s + 3 (a ) = ( s + 3) 2 + 4 For purposes of application of the First Shifting Theorem to invert the transform, Equation (a) is rewritten as 2( s + 3 3) + 3 X (s) = ( s + 3) 2 + 4 3 s+3 =2 2 ( s + 3) + 4 ( s + 3) 2 + 4 3 2 s+3 =2 ( b) 2 ( s + 3) + 4 2 ( s + 3) 2 + 4 The First Shifting Theorem, linearity of the inverse transform and transform pairs from Table 5.1 are used to invert Equation (b) to obtain 3 x(t ) = 2e 3t sin( 2t ) e 2t cos( 2t ) (c) 2
(a)
161
(b)
Use of Equation (b) in Equation (a) leads to 5s 2 + 4 2 1 Bs + C = + 2 2 s + 2 s + 26 13 s s + 2 s + 26 2 5s 2 + 4 = (s 2 + 2 s + 26 ) + ( Bs + C ) s 13 4 2 = + B s 2 + + C s + 4 13 13 Equating coefficients of like powers of s in Equation (c) leads to 2 5= +B 13 63 B= 13 4 0 = +C 13 4 C= 13 Use of Equations (c), (d), and (e) in Equation (a) leads to 1 2 63s 4 X ( s) = + 2 13 s s + 2 s + 26 1 1 63s 4 = + s ( s + 1) 2 + 25 13 1 1 63( s + 1) 67 + 2 2 13 s ( s + 1) + 25 ( s + 1) + 25 Inversion of Equation (f) leads to 67 1 x(t ) = u (t ) + 63e t cos(5t ) e t sin(5t ) 5 13 5.30 The appropriate partial fraction decomposition for X (s ) is 2s + 7 A B C = + + 2 s + 2 s + 4 ( s + 4) 2 ( s + 2)( s + 4) The residue A is calculated as =
(c)
(d)
(e)
(f)
(g)
(a)
162
Chapter 5 2s + 7 ( s + 4) 2
A= =
s = 2
2(2) + 7 (2 + 4) 2 3 = 4
(b)
Use of Equation (b) in Equation (a) leads to B C 2s + 7 3 1 = + + 2 4 s + 2 s + 4 ( s + 4) 2 ( s + 2)( s + 4) 3 2 2s + 7 = (s + 4) + B( s + 2)( s + 4) + C ( s + 2) 4 Evaluating Equation (c) at s=-4 leads to 2(4) + 7 = C (4 + 2) 1 C= 2 Evaluating Equation (c) for s=0, using Equation (d) leads to 1 3 2 2(0) + 7 = (4 ) + B (2)(4) + (2) 2 4 3 B= 4 Thus 1 3 3 2 + X (s) = 4 s + 2 s + 4 ( s + 4) 2 Inversion of Equation (f) leads to 1 x(t ) = 3e 2t 3e 4t + 2te 4t 4 5.31 The appropriate partial fraction decomposition is 1 X ( s) = 2 (s + 2s + 4)(s + 1) 2 As + B C D = 2 + + s + 2 s + 4 s + 1 (s + 1)2 Multiplication of Equation (a) by the denominator leads to 2 1 = ( As + B )(s + 1) + C ( s 2 + 2 s + 4)( s + 1) + D ( s 2 + 2 s + 4)
(c)
(d)
(e)
(f)
(g)
(a)
= ( As + B )( s 2 + 2 s + 1) + C s 3 + 3s 2 + 6 s + 4 + D ( s 2 + 2 s + 4) = ( A + C ) s 3 + ( 2 A + B + 3C + D ) s 2 + ( A + 2 B + 6C + 2 D ) s + ( B + 4C + 4 D ) (b)
Equating coefficients of like powers of s in Equation (b) leads to the following set of equations to solve for the unknown coefficients
163
Chapter 5 0 A 0 1 B 0 = (c) 2 C 1 4 D 0 1 1 1 The solution of Equation (c) is A = , B = 0, C = , D = . Thus the transform of the 3 3 3 response is 1 1 s 1 X ( s) = 2 + 3 s + 2s + 4 s + 1 ( s + 1) 2 0 1 2 1 1 3 6 4 = = 1 1 1 s + 2 3 ( s + 1) + 3 s + 1 ( s + 1) 2 (d) 1 2 1 0
( )
( )
(e)
5.32 (a) The MATLAB worksheet to determine the partial fraction decomposition of
X ( s) = 3s + 1 s ( s 2 + 9) (a)
>> D=[1 0 9 0] D= 1 0 9 0
Chapter 5 0.1111
p= 0 + 3.0000i 0 - 3.0000i 0
k= [] >> syms s >> Num=r(2)*(s-p(1))+r(1)*(s-p(2)) Num = (-1/18+1/2*i)*(s-3*i)-(1/18+1/2*i)*(s+3*i) >> Num1=vpa(simplify(Num),3) Num1 = -.111*s+3. >> Thus the appropriate partial fraction decomposition is 0.111 0.111s + 3 X (s) = + (b) s s2 + 9 (b) The MATLAB workspace to help determine the partial fraction decomposition for A A A 3s + 1 (c) = 1+ 2 + 3 2 s s+3 s3 s( s 9) is provided below >> clear >> N=[3 1] N= 3 1
Chapter 5
0 -9
p= 3 -3 0
k= [] >> The partial fraction decomposition determined from the above is 0.1111 0.5556 0.4444 X ( s) = + s s3 s+3 (c) The MATLAB worksheet to determine the partial fraction decomposition of 5s 2 + 4 X ( s) = s( s 2 + 2 s + 26) is provided below >> clear >> N=[5 0 4] N= 5 0 4
(d)
(e)
>> D=[1 2 26 0] D= 1 2 26 0
Chapter 5
k= [] >> syms s >> Num=r(1)*(s-p(2))+r(2)*(s-p(1)) Num = (63/26+67/130*i)*(s+1+5*i)+(63/26-67/130*i)*(s+1-5*i) >> Num1=vpa(simplify(Num),3) Num1 = 4.85*s-.308 >> The partial fraction decomposition is determined from the MATLAB work as 0.1538 4.85s 0.308 X (s) = + 2 s s + 2 s + 26 (d) The MATLAB worksheet to determine the partial fraction decomposition for 2s + 7 X ( s) = ( s + 2)( s + 4) 2 is provided below >> clear >> N=[2 7] N= 2 7 167
(f)
(g)
Chapter 5
>> syms s >> Ds=(s+2)*(s+4)^2 Ds = (s+2)*(s+4)^2 >> expand(Ds) ans = s^3+10*s^2+32*s+32 >> D=[1 10 32 32] D= 1 10 32 32 >> [r,p,k]=residue(N,D) r= -0.7500 0.5000 0.7500
k= [] >> The partial fraction decomposition is determined from the MATLAB workspace as 0.75 0 .5 0.75 X ( s) = + + 2 s + 4 ( s + 4) s+2
(h)
168
Chapter 5 5.33 The mathematical problem is & (a ) x + 3x = 2e 3t ( b) x(0) = 0 Taking the Laplace transform of both sides of Equation (a), defining X ( s) = L{x(t )} using the linearity property leads to & L{x} + 3{x} = 2L {e 3t } (c) 1 Noting from Table 5.1 that L {e 3t } = and using the property of transform of s+3 derivatives Equation (c) is rewritten as 2 (d ) sX ( s ) x(0) + 3 X ( s ) = s+3 Substitution of the initial conditions, Equation (b) into Equation (d) and solving for X(s) results in 2 ( e) X ( s) = ( s + 3) 2 Equation (e) is inverted using the first shifting theorem and the known transform pair 1 L{t} = 2 leading to s (f ) x(t ) = 2te 3t 5.34 The mathematical problem is
&+ 2 x + 17 x = 0 & & x
x ( 0) = 0 & x ( 0) = 1 Taking the Laplace transform of both sides of X ( s) = L{x(t )} using the linearity property leads to & & L{& + 2L x + 17L{x} = 0 x}
Equation
(a),
(d )
Using the property of transform of derivatives Equation (d) is rewritten as & s 2 X ( s ) sx(0) x(0) + 2[ X ( s) sx(0)] + 17 X ( s ) = 0 (e) Substitution of the initial conditions, Equations (b) and (c) into Equation (e) and solving for X(s) results in 1 X ( s) = 2 (f ) s + 2 s + 17 Equation (f) is rewritten by completing the square in the denominator of the fraction on its right-hand side 1 X ( s) = (g ) (s 2 + 1)2 + 16 The inverse of the right-hand side of Equation (g) is obtained by using the First Shifting Theorem and known transform pairs leading to 1 x(t ) = e t sin( 4t ) (h ) 4
169
(a )
x ( 0) = 0 ( b) & x ( 0) = 0 (c) Taking the Laplace transform of both sides of Equation (a), defining X ( s) = L{x(t )} using the linearity property leads to & L{& + 25L{x} = 3L{sin(5t )} x} (d ) 5 , using the property of transform of Noting from Table 5.1 that L {sin(5t )} = 2 s + 25 derivatives Equation (d) is rewritten as 15 & s 2 X ( s ) sx (0) x(0) + 25 X ( s ) = 2 (e ) s + 25 Substitution of the initial conditions, Equations (b) and (c) into Equation (e) and solving for X(s) results in 15 X ( s) = (f ) 2 2 s + 25 There are several ways to invert the transform in Equation (f). Note that 3 5 5 X ( s) = 2 (g ) 5 s + 25 s 2 + 25 Application of the convolution property leads to 3 x(t ) = sin(5t ) sin(5t ) 5 t 3 = sin(5 ) sin [5(t )]d 50
Use of trig identities on the integrands leads to t t 3 1 1 x(t ) = sin(5t ) sin(10 )d cos(5t ) [1 cos(10 )]d 5 2 2 0 0
= 3 1 =t sin(5t )cos(10 ) =0 cos(5t ) 10 10
=t =0
1 =t cos(5t )sin(10 ) =0 10
3 1 1 1 sin(5t ) cos(10t ) + sin(5t ) t cos(5t ) + cos(5t ) sin(10t ) (i) 10 10 10 10 Further algebra gives =
170
Chapter 5
3 1 1 t cos(5t ) + 10 sin(5t ) 10 (sin(5t ) cos(10t ) cos(5t ) sin(10t ) ) 10
x(t ) = =
( j)
5.36 The mathematical problem is &+ 49 x = 7e 2t & x (a) x(0) = 0 (b) & x(0) = 0 (c) Taking the Laplace transform of both sides of Equation (a), defining X ( s) = L{x(t )} using the linearity property leads to & L{& + 49L{x} = 7L {e 2t } x} (d ) 1 and using the property of transform of Noting from Table 5.1 that L {e 2 t } = s+2 derivatives Equation (d) is rewritten as 7 & s 2 X ( s ) sx (0) + x(0) + 49 X ( s ) = ( e) s+2 Substitution of the initial conditions, Equations (a) and (b) into Equation (e) and solving for X(s) results in 7 X (s) = (f ) ( s + 2)( s 2 + 49) The appropriate partial fraction decomposition of the right-hand side of Equation (f) is 7 A Bs + C = + 2 (g ) 2 ( s + 2)( s + 49) s + 2 s + 49 The residue A is determined as 7 s + 29 s = 2 7 = 2 (2) + 49 7 = 53 Substitution of Equation (h) in Equation (g) gives 7 Bs + C 7 = 53 + 2 2 ( s + 2)( s + 49) s + 2 s + 49 7 2 s + 49 + (Bs + C ) )( s + 2) 7= 53 A=
2
(h )
(i)
171
( j)
Evaluation of Equation (i) at s=2 leads to 7 14 (53) + Bs + (4) 53 53 14 B= 53 Use of Equations (g), (h), (j), and (k) in Equation (f) leads to 7 1 2 2s X ( s) = + 2 53 s + 2 s + 49 Inversion of Equation (l) using transform pairs found in Table 5.1 leads to 2 7 2t x(t ) = e 2 cos(7t ) + 7 sin(7t ) 53 7=
(k )
( l)
( m)
5.37 The mathematical problem is & + 2 &+ 25 x + 50 x = 0 & & & x& x (a) x ( 0) = 0 (b) & x ( 0) = 0 (c) & 0 ) = 0 .5 & x( (d) Taking the Laplace transform of both sides of Equation (a), defining X ( s) = L{x(t )} using the linearity property leads to & & & L{&} + 25L{& + 25L{x} + 50L{x} = 0 x& x} (e) Using the property of transform of derivatives Equation (d) is rewritten as & & & s 3 X ( s ) sx(0) s 2 x(0) s& 0) + 2 s 2 X ( s) sx(0) x(0) x( + 25[sX ( s ) x(0)] + 50 X ( s) = 0 (f ) Substitution of the initial conditions, Equations (a), (b), and (c) into Equation (f) and solving for X(s) results in 0 .5 X ( s) = 3 (g ) 2 s + 2 s + 25s + 50 The poles of the transform (the roots of the polynomial in the denominator of Equation (g) are s = 2,5.0 j . Thus Equation (g) can be written as 0 .5 X ( s) = (h ) ( s + 2)( s 2 + 25) A partial fraction decomposition leads to
172
Chapter 5
X ( s) = A Bs + C + 2 s + 2 s + 25
(i)
A = lim = 0.5 29 1 = 58
s 2
( j)
Bs + C 0.5 1 1 = + 2 2 ( s + 2)( s + 25) 58 s + 2 s + 25 1 0.5 = (s 2 + 25) + ( Bs + C )( s + 2) 58 Setting s=0 in Equation (k) leads to 1 1 = (25) + 2C 2 58 1 C= 29 Setting s=-1 in Equation (l) leads to 1 1 1 = (26) + B + (1) 2 58 29 1 B= 58 Substitution of Equations (j), (l), and (m) into Equation (h) gives 1 1 s+2 X ( s) = + 2 58 s + 2 s + 25 Inversion of Equation (n) leads to 2 1 2t x(t ) = e cos(5t ) + 5 sin(5t ) 58 5.38 The mathematical problem is
Thus
(k )
( l)
( m)
(n )
(o)
x 2 (0) = 0 (d) Taking the Laplace transforms of Equations (a) and (b) and using linearity of the transform leads to
173
Chapter 5 2 (e) s+3 & L{x 2 } X 1 ( s ) + 3 X 2 ( s) = 0 (f) Using the property of the transform of the first derivative and application of Equations (c) and (d) leads to 2 ( s + 2) X 1 ( s ) X 2 ( s) = (g) s+3 X 1 ( s) + ( s + 3) X 2 ( s) = 0 (h) Equations (g) and (h) are summarized in a matrix form as s + 2 1 X 1 (s) 2 (i) 1 s + 3 X ( s ) = s + 3 2 0 Cramers rule is applied to determine the solution of Equation (i) as 2 1 s +1 0 s+3 X 1 ( s) = s + 2 1 1 s + 3 = 2( s + 1) (s + 3)(s 2 + 5s + 5) s+2 X 2 ( s) = 1 2 s+3 0 (j) & L{x1 } + 2 X 1 ( s ) X 2 ( s ) =
s 2 + 5s + 5 2 (k) = 2 ( s + 3)( s + 5s + 5) MATLAB is used to help determine the partial fraction decompositions for Equations (j) and (k). The MATLAB workspace is >> clear >> syms s >> Ds=(s+3)*(s^2+5*s+5) Ds = (s+3)*(s^2+5*s+5) >> Ds1=expand(Ds) Ds1 = s^3+8*s^2+20*s+15
174
>> N=[2 2] N= 2 2
175
Chapter 5
k= [] >> The partial fraction decompositions determined using MATLAB are 4 3.789 0.2111 X 1 ( s) = s + 3 s + 3.618 s + 1.382 1.447 0.5538 2 X 2 ( s) = + + s + 3 s + 3.618 s + 1.382 Inversion of Equations (l) and (m) leads to x1 (t ) = 4e 3t 3.789e 3.618t 0.2111e 1.382t
(l) (m)
(n) (o)
(a)
x(0) = .001 (b) & x ( 0) = 0 (c) Defining x ( s) = L{x(t )}, taking the Laplace transform of Equation (a), and using linearity of the transform leads to & & 10L{& + 20L{x} + 250L{x} = 0 x} (d) Using the properties of transforms of first and second derivatives and use of the initial conditions of Equations (b) and (c) leads to & 10 s 2 x ( s ) sx(0) x(0) + 20[sx ( s ) x(0)] + 250 x ( s ) = 0
10[ s 2 x ( s ) .001s ] + 20[ sx ( s ) .001] + 250 x ( s ) = 0 (10s 2 + 20s + 250) x ( s ) = .01s + .02 0.01s + 0.02 x ( s) = 10s 2 + 20s + 250 0.001s + 0.002 (e) = 2 s + 2s + 25 The transform of Equation (e) has only complex poles. The technique of completing the square is used to write the denominator as ( s + 1) 2 + 24 . Thus
176
Chapter 5 0.001s + 0.002 ( s + 1) 2 + 24 0.001( s + 1) 0.001 + = 2 ( s + 1) + 24 ( s + 1) 2 + 24 Inversion of Equation (f) leads to 1 t x(t ) = 0.001e t cos 24t + e sin 24t 24 x (s) =
(f)
(g)
(a)
x ( 0) = 0 (b) & x ( 0) = 0 (c) Defining X ( s) = L{x(t )} , taking the Laplace transform of Equation (a), and using linearity of the transform leads to & & 10L{& + 20L{x} + 250L{x} = 200L{sin(400t )} x} (d) Using the properties of transforms of first and second derivatives and use of the initial conditions of Equations (b) and (c) leads to 200(400) & 10 s 2 X ( s ) sx(0) x(0) + 20[sX ( s ) x(0)] + 250 X ( s ) = 2 s + (400) 2
10 s 2 X ( s ) + 20 sX ( s ) + 250 X ( s ) =
8 x10 4 s 2 + 1.6 x10 5 8 x10 4 (10 s 2 + 20 s + 250) X ( s ) = . 2 s + 1.6 x10 5 8 x10 4 X ( s) = 10 s 2 + 20 s + 250 s 2 + 1.6 x10 5
)(
8 x10 3 ( e) s 2 + 2 s + 25 s 2 + 1.6 x10 5 The appropriate partial fraction decomposition is of the form As + B Cs + D X ( s) = 2 + 2 (f ) s + 2 s + 25 s + 1.6 x10 5 Multiplying through by the denominator of the right-hand side of Equation (e) leads to 8 x10 3 = ( As + B )( s 2 + 1.6 x10 5 ) + (Cs + D )( s 2 + 2 s + 25) =
)(
8 x10 3 = ( A + C )s 3 + (B + 2C + D )s 2 + 1.6 x10 5 A + 25C + 2 D s + 1.6 x10 5 B + 25 D Equating coefficients of like powers of s leads to
(g )
177
Chapter 5 0 1 0 A 0 1 0 1 2 1 B 0 = 1.6 x10 5 0 25 2 C 0 1.6 x10 5 0 25 D 8 x10 3 0 Then solution to the set of equations specified in Equation (h) is A = 6.25 x10 7 , B = 0.05, C = 6.25 x10 7 , D = 0.05 Substitution of Equation (i) into Equation (f) leads to 6.25 x10 7 s + 0.05 6.25 x10 7 s + 0.05 X ( s) = s 2 + 2s + 25 s 2 + 1.6 x10 5 Completing the square in the denominator of the first term leads to 6 .25 x10 7 s + 0 .05 6 .25 x10 7 s + 0 .05 X (s) = ( s + 1) 2 + 24 s 2 + 1 .6 x10 5
(h )
(i) ( j)
(k)
Inversion of Equation (k) leads to x(t ) = 6.25 x10 7 e t cos( 24t ) + .05 25 .05 + 6.26 x10 7 cos(400t ) + sin( 400t ) 400 e t sin( 24t ) ( l)
Defining X ( s) = L{x(t )} , taking the Laplace transform of Equation (a), and using linearity of the transform leads to & & 10L{& + 20L{x} + 250L{x} = 200L{u (t ) u (t 2)} x} (d) Using the properties of transforms of first and second derivatives and use of the initial conditions of Equations (b) and (c) leads to 200 & 10 s 2 X ( s) sx(0) x(0) + 20[ X ( s) x(0)] + 250 X ( s) = (1 e 2 s ) s 200 10 s 2 + 20s + 250 X ( s ) = (1 e 2 s ) s 200 X (s) = (1 e 2 s ) 2 s 10s + 20 s + 250 200 = (1 e 2 s ) (e) 2 s ( s + 2s + 25) The appropriate partial fraction decomposition is
178
Equation (g) is inverted using the first shifting theorem, the second shifting theorem and known transforms leading to x(t ) = 8u (t ) 8e t cos(4.90t ) 1.63e t sin(4.90t ) 8u (t 2)
(h)
(a)
i (0) = 0 (b) Define I ( s ) = L{i (t )} . Taking the Laplace transform of Equation (a) and using linearity of the transform gives t di 5 0.1L + 1000L{} + 4 x10 L idt = 1000L{sin(500t )} i (c) dt 0 Use of properties of transform of the first derivative and transform of an integral and application of the initial condition, Equation (b), leads to 4 x10 5 1000(500) I (s) = 2 0.1[sI ( s ) i (0)] + 1000 I ( s ) + s s + (500) 2
4 x10 5 0.1s + 1000 + s 5 x10 5 I (s) = 2 s + (500) 2 5 x10 5 s I (s) = 0.1s 2 + 1000 s + 4 x10 5 ( s 2 + 500 2 )
( (
I (s) =
(d)
3 The roots of s 2 + 10000 s + 4 x10 6 are determined as s1 = 417.4 and s2 = 9.58x10 . Thus Equation (d) can be rewritten as 5 x10 5 s (e) I ( s) = ( s + 417.4)( s + 9.58 x10 3 )( s 2 + 500 2 ) which has a partial fraction decomposition of A1 A2 Bs + C I ( s) = + (f) + 2 3 s + 417.4 s + 9.58 x10 s + 500 2
179
A1 = A2 =
Thus
= 0.0537
s = 417.4
(g) (h)
= 0.5.68 x10 3
s = 9.58 x10
3
0.0537 Bs + C 5 x105 s 5.68x10 3 (i) = + + 2 3 2 2 3 s + 5002 ( s + 417.4)(s + 9.58x10 )(s + 500 ) s + 417.4 s + 9.58x10 Evaluation of Equation (i) at s=0 leads to C = 32.01 . Further evaluation of Equation (i) at s=500 leads to B=0.1221. Thus 0.0537 5.68 x10 3 0.1221s + 32.01 + (j) + I (s) = 3 s + 417.4 s + 9.58 x10 s 2 + 500 2 Inversion of Equation (j) leads to 3 i(t ) = 0.0537e 417.4t + 5.68x10 3 6e 9.58 x10 t + 0.1221cos(500t ) + 0.0640(500t ) (k) 5.43 The mathematical problem is t di 0.1 + 1000i + 4 x10 5 idt = 1000e 0.02t (a) dt 0 i ( 0) = 0 (b) Define I ( s ) = L {i (t )} . Taking the Laplace transform of Equation (a) and using linearity of the transform gives t di 5 0.1L + 1000L{} + 4 x10 L idt = 1000 L{e 0.02t } i (c) dt 0 Use of properties of transform of the first derivative and transform of an integral and application of the initial condition, Equation (b), leads to 4 x10 5 1000 0.1[sI ( s ) i (0)] + 1000 I ( s ) + I ( s) = s s + 0.02 4 x10 5 1000 0.1s + 1000 + I ( s) = s s + .02
I ( s) = I ( s) =
( (
(d)
3 The roots of s 2 + 10000 s + 4 x10 6 are determined as s1 = 417.4 and s2 = 9.58x10 . Thus Equation (d) can be rewritten as 1x10 4 s (e) I ( s) = ( s + 417.4)( s + 9.58 x10 3 )( s + 0.02) which has a partial fraction decomposition of
180
Chapter 5 1.091 1.091 5 x10 5 s + 417.4 s + 9.58 x10 3 s + 0.02 Inversion of Equation (f) leads to 3 i(t ) = 1.091e 417.4t 1.091e 9.58 x10 t 5 x10 5 e 0.02t I ( s) = 5.44 The mathematical problem is t di 0.1 + 1000i + 4 x10 5 idt = 1000[u (t ) u (t 0.25)] dt 0
(f)
(g)
(a)
i (0) = 0 (b) Define I ( s) = L{i(t )}. Taking the Laplace transform of Equation (a) and using linearity of the transform gives t di 0.1L + 1000L{} + 4 x10 5 L idt = 1000 L{u (t ) u (t 0.25)} i (c) dt 0 Use of properties of transform of the first derivative and transform of an integral and application of the initial condition, Equation (b), leads to 4 x10 5 1000 (1 e 0.25s ) I ( s) = 0.1[sI ( s) i (0)] + 1000 I ( s) + s s 5 4 x10 1000 I (s) = 0.1s + 1000 + (1 e 0.25s ) s s 1000(1 e 0.25 s ) I (s) = (0.1s 2 + 1000s + 4 x105 )
(d)
3 The roots of s 2 + 10000 s + 4 x10 6 are determined as s1 = 417.4 and s2 = 9.58x10 . Thus Equation (d) can be rewritten as 10000 1 e 0.25 s I ( s) = ( s + 417.4)( s + 9.58 x10 3 )
A B = + 1 e 0.25 s 3 s + 417.4 s + 9.58 x10 1.091 1.091 = 1 e 0.25 s 3 s + 417.4 s + 9.58 x10 Inversion of Equation (e) leads to
) )
(e)
(1.091e
417.4 ( t 0.25 )
( t 0.25 )
)u(t 0.25)
(f)
181
1 1 i1dt C i2 dt = v(t ) C0 0
t
di 1 1 i1 dt + L 2 + Ri 2 + i2 dt = 0 C0 dt C0 i1 (0) = 0 i 2 ( 0) = 0 Substitution of given values into Equations (a) and (b) leads to 500i1 + 5 x10
t 4
i dt 5 x10 i dt = 500u (t )
4 1 2 0 0
(e) (f)
t di 2 + 500i2 + 5 x10 4 i2 dt = 0 dt 0 0 Taking the Laplace transforms of Equations (e) and (f) and using linearity leads to t t 500 500 I 1 ( s ) + 5 x10 4 L i1 dt 5 x10 4 L i2 dt = s 0 0
5 x10 4 i1 dt + 0.2
(g)
t t di2 4 5 x10 L i1 dt + 0.2L + 500 I 2 ( s) + 5 x10 L i1 dt = 0 (h) dt 0 0 Application of the property of transform of the first derivative, transform of an integral, and the initial conditions, Equations (c) and (d), to Equations (g) and (h) leads to 5 x10 4 5 x10 4 500 500 I 1 ( s) + I1 ( s) I 2 ( s) = (i) s s s 5 x10 4 5 x10 4 + 0.2sI 2 ( s) + 500 I 2 ( s) + I 2 ( s) = 0 (j) s s Equations (i) and (j) are summarized in matrix form as 5 x10 4 5 x10 4 500 + I 1 ( s ) 500 s s = s (k) 5 x10 4 5 x10 4 I 2 ( s ) 0 0.2 s + 500 + s s Solution of Equation (k) using Cramers rule gives
4
182
Chapter 5
5 x10 4 s 5 x10 4 0 0.2 s + 500 + s I 1 ( s) = 4 5 x10 5 x10 4 500 + s s 5 x10 4 5 x10 4 + 0.2 s + 500 + s s 500 s = 5 x10 4 500 0.2 s + 500 + s s
2
5 x10 4 5 x10 4 500 + 0.2s + 500 + s s 500 0.2 s 2 + 500s + 5 x10 4 = s 100s 2 + 2.6 x10 5 s + 5 x10 7
5 x10 4 s
= and
(l)
5 x10 4 500 s s 4 5 x10 0 s I 2 ( s) = 5 x10 4 5 x10 4 500 + s s 5 x10 4 5 x10 4 + 0.2s + 500 + s s 500 + =
4 500 5 x10 s s
5 x10 4 s
2
5 x10 4 5 x10 4 500 + 0.2s + 500 + s s 2.5 x10 7 = s 100s 2 + 2.6 x10 5 s + 5 x10 7
( (
2.5 x10 (m) s s + 2.6 x10 3 + 5 x10 5 The quadratic polynomial in the denominators of Equations (l) and (m) have roots of s1 = 2.39 x10 3 and s 2 = 2.09 x10 2 . Partial fraction decompositions are performed with the aid of MATLAB =
2
>> clear
183
184
k= [] >> The partial fraction decompositions are 0.5 2.09 x10 3 0.502 i1 (t ) = + 3 s s + 2.39 x10 s + 2.09 x10 2 0.5 4.79 x10 2 0.5479 i2 (t ) = + 3 s s + 2.39 x10 s + 2.09 x10 2 Equations (n) and (o) are inverted leading to 3 2 i1 (t ) = 0.5u (t ) 2.09 x10 3 e 2.39 x10 t + 0.502e 2.09 x10 t i2 (t ) = 0.5u (t ) + 4.79 x10 2 e 2.39 x10 t 0.5479e 2.09 x10 The following MATLAB m file is used to plot Equations (p) and (q) t=0:.0001:0.02 i1=0.5-2.09e-3*exp(-2.39e3.*t)+0.502*exp(-2.09e2.*t) i2=0.5+4.79e-2*exp(-2.39e3.*t)-0.5479*exp(-2.09e2.*t) plot(t,i1,'-',t,i2,'.') xlabel('t (s)') ylabel('i (A)') legend('i_1(t)','i_2(t)') title('Solution of Problem 5.45')
3 2
(n) (o)
(p)
t
(q)
185
Chapter 5
5.46 The differential equations obtained from the system modeling are & m1 & + (k1 + k 2 ) x1 k 2 x 2 = 0 x1 & m2 & k 2 x1 + (k 2 + k 3 ) x 2 = 0 x2 Substitution of given values into Equations (a) and (b) leads to & 10 & + 2.5 x10 4 x1 1.5 x10 4 x 2 = 0 x1
(a) (b)
(c) (d)
(e) (f)
& 20 & 1.5 x10 4 x1 + 2.5 x10 4 x 2 = 0 x2 The given initial conditions are x1 (0) = 0.001 & x1 (0) = 0
x 2 (0) = 0 (g) & x 2 (0) = 0 (h) Taking the Laplace transforms of Equations (c) and (d) using the initial conditions of Equations (e)-(h) leads to 10 s 2 X 1 ( s) 0.001s + 2.5 x10 4 X 1 ( s) 1.5 x10 4 X 2 ( s ) = 0 (i)
20s 2 X 2 ( s) 1.5 x10 4 X 1 ( s ) + 2.5 x10 4 X 2 ( s) = 0 Equations (i) and (j) are rearranged and written in matrix form as
(j)
186
Chapter 5
10s 2 + 2.5 x10 4 1.5 x10 4 X 1 ( s) 0.01s = 4 20s 2 + 2.5 x10 4 X 2 ( s) 0 1.5 x10 Cramers rule is used to solve the simultaneous equations of Equation (j) leading to 0.01s 1.5 x10 4 0 20s 2 + 2.5 x10 4 X 1 ( s) = 10s 2 + 2.5 x10 4 1.5 x10 4 20 s 2 + 2.5 x10 4 1.5 x10 4
= =
(k)
(10s
)(
) (
0.2s 3 + 250s 200 s 4 + 7.5 x10 5 s 2 + 4 x10 8 0.0010s 3 + 1.25s = 4 s + 3750s 2 + 2 x10 6 0.0010 s 3 + 1.25s = 2 s + 643.9 s 2 + 3.11x10 3
)(
(l) (m)
Inversion of Equation (l) leads to x1 (t ) = 2.46 x10 3 cos(25.38t ) + 7.54 x10 3 cos(55.74t )
X 2 ( s) =
(10s
+ 2.5 x10 20 s 150 s 4 200 s + 7.5 x10 5 s 2 + 4 x10 8 0.75s 4 s + 3750s 2 + 2 x10 6 0.75s 2 s + 643.9 s 2 + 3.11x10 3 0.0305s 0.0305s 2 2 s + 643.9 s + 3.11x10 3
2 2
)(
4 2
)(
(n)
187
Chapter 5
x1 (t ) = 0.0305 cos(25.38t ) 0.0305 cos(55.74t ) The following MATLAB code is written to generate a plot of the system response t=0:.001:0.5 x1=2.46e-3*cos(25.38.^t)+7.54e-3*cos(55.74.*t); x2=0.0305*cos(25.38.*t)-0.0305*cos(55.74.*t); plot(t,x1,'-',t,x2,'.') xlabel('t (s)') ylabel('x (m)') legend('x_1(t)','x_2(t)') title('Solution of Problem 5.46')
(o)
5.47 Substitution of numerical values into Equations (a) and (b) of the problem statement leads to dh 100 1 + 0.1h1 0.125h2 = 0 (a) dt dh 150 2 0.1h1 + 0.225h2 = 0.5u (t ) (b) dt Since the dependent variables represent perturbations from a steady state, h1 (0) = h2 (0) = 0 . Let H 1 ( s ) = L{h1 (t )} and H 2 ( s ) = L{h2 (t )}. Taking the Laplace
188
Chapter 5 transform of Equation(a), using linearity or the transform and the property of transform of the first derivative leads to
(c)
(e)
(f)
4.17 x10 6 s s 2 + .0025s + 6.67 x10 7 100s + 0.1 0 0.5 .05 0.1 50 + s s H2 = = 4 2 100s + 0.1 0.125 1.5 x10 s + 37.5s + 0.01 =
(g)
0.1 =
150s + 0.225
3.33 x10 3 s + 3.33x10 6 s s 2 + .0025s + 6.67 x10 7 Partial fraction decompositions lead to 6.25 1.00 7.26 H 1 ( s) = + s s + 0.002196 s + .000304 5 .0 2.56 7.55 H 2 ( s) = + s s + 0.002196 s + .000304 Inversion of the transforms in Equations (h) and (i) leads to
(h) (i)
189
Chapter 5
(j) (k)
7.55e
0.000304t
5.48 The equations obtained for the perturbation in temperatures of oil and water in the double pipe heat exchanger of Example 4.25 are d & & o c o ri 2 L o + (2 ro LU + c o m o ) o 2 roUL w = c 0 m o oi (a) dt
d w & & (b) 2ro LU o + (2ro LU + c w m w ) w = c w m w wi dt It is desired to determine the perturbations in temperatures of the oil and water when oi (t ) = 0 , wi (t ) = 3u (t ) and kg kg J J w = 9.60 x10 2 3 , o = 7.50 x10 2 3 c w = 4.19 x10 3 , co = 2.4 x10 3 m m kg C kg C W kg kg & & U = 8.65 x10 2 2 m w = 2.51 mo = 4.85 ri = 2 cm ro = 2.2 cm, r2 = 4.2 cm, L = 2.5 m s s m C The appropriate initial conditions are o (0) = w (0) = 0 . Substitution of given values into Equations (a) and (b) leads to
w c w (r22 ro2 )L
d o + 1.19 x10 4 o 1.36 x10 4 w = 0 dt d 4.04 x10 4 w 1.36 x10 4 o + 2.41x10 4 w = 3.16 x10 4 u (t ) dt Taking the Laplace transforms of Equations (c) and (d), applying initial conditions, summarizing the resulting equations in a matrix form leads to 0 o ( s) 5.65x103 s + 1.19 x10 4 1.36 x10 4 = 3.16 x10 4 4.04 x10 4 s + 2.41x10 4 w ( s) 1.36 x10 4 s 5.65 x10 3
(c) (d)
and
(e)
Cramerr rule is used to determine a solution to the simultaneous equations of Equation (e) leading to
190
Chapter 5 0 1.36 x10 4 4 3.16 x10 4.04 x10 4 s + 2.41x10 4 s o ( s) = 5.65 x10 3 s + 1.19 x10 4 1.36 x10 4 1.36 x10 4 4.04 x10 4 s + 2.41x10 4 3.16 x10 4 (1.36 x10 4 ) s = 3 4 4 5.65 x10 s + 1.19 x10 (4.04 x10 s + 2.4 x10 4 ) (1.36 x10 4 ) 2 4.30 = 2 s (2.48s + 6.17 s + 1.02) 1.73 = 2 s ( s + 2.49s + 0.411)
(f)
5.65 x10 3 s + 1.19 x10 4 1.36 x10 4 w ( s) = 5.65 x10 3 s + 1.19 x10 4 1.36 x10 4
3.16 x10 4 (5.65 x10 3 s + 1.19 x10 4 ) s = 3 4 4 4 (5.65x10 s + 1.19 x10 )(4.04 x10 s + 2.4 x10 ) (1.36 x10 4 ) 2 1.78s + 3.76 = s (2.48s 2 + 6.17 s + 1.02) 0.718s + 1.52 = 2 s ( s + 2.49 s + 0.4113) Partial fraction decompositions lead to 1.70 0.602 2.30 0 (s) = + s s + 1.97 s + 0.517 1.49 0.0360 1.53 w (s) = + s s + 1.87 s + 0.517 Inversion of Equations (h) and (i) leads to o (t ) = 1.70 + 0.602e 1.97 t 2.30e 0.517t
(g)
(h) (i)
(j) (k)
w (t ) = 1.49 + 0.0360e
1.97 t
1.53e
0.517 t
5.49 Then differential equations obtained for the concentrations of the reactant and product in the constant volume CSTR of Example 4.16 are
191
Chapter 5
dC A + ( q + kV ) C A = qC Ai dt dC B V kVC A + qC B = qC Bi dt
(a) (b)
It is desired to determine the concentrations of the reactant A and the product B when 3 mol 6 m 3 3 q = 2.0 x10 k = 2.0 x10 3 s 1 C Ai (t ) = 0.2u (t ) and V = 1.00 x10 m , s L C Bi (t ) = 0 . Substitution of these values into Equations (a) and (b) leads to dC A 1x10 3 + 4 x10 6 C A = 4 x10 7 u (t ) (c) dt dC B 1x10 3 2 x10 6 C A + 2 x10 6 C B = 0 (d) dt Taking the Laplace transforms of Equations (c) and (d), taking initial conditions to be zero, simplifying, and writing the resulting equations in a matrix form leads to 4 s + 4 x10 3 C A ( s ) 4 x10 0 (e) = s s + 2 x10 3 2 x10 3 C B ( s ) 0 Equation (e) is used to determine 4 x10 4 C A ( s) = (f) s ( s + 4 x10 3 )
2x10 -3 C B ( s) = C A ( s) s + 2 x10 3 8 x10 7 = s ( s + 2 x10 3 )( s + 4 x10 3 ) Partial fraction decompositions of Equations (f) and (g) leads to 1 1 C A ( s ) = 0.1 3 s s + 4 x10 2 1 1 + C B ( s ) = 0.1 3 3 s + 4 x10 s s + 2 x10 Inversion of Equations (h) and (i) leads to 3 C A (t ) = 0.1 1 e 4 x10 t
(g)
(h) (i)
2 x10 t CB + e 4 x10 t (k) The following MATLAB program is used to plot the time dependence of the concentrations % Problem 5.49 dt=.1/4e-3; for i=1:101 t(i)=(i-1)*dt; CA(i)=0.1*(1-exp(-4e-3*t(i)));
( (t ) = 0.1( 2e 1
(j)
192
Chapter 5 CB(i)=0.1*(1-2*exp(-2e-3*t(i))+exp(-4e-3*t(i))); end plot(t,CA,'-',t,CB,'.') xlabel('t (s)') ylabel('C (mol/L') title('Response of system of Problem 5.49') legend('C_A(t)','C_B(t)') The resulting MATLAB generated plot is
193
Chapter 5
194
Chapter 6
1.3 x10 5 y1 + 900 y1 & (b) 5 & 1.04 x10 y1 720 y1 Define X ( s ) = L{x(t )} and ( s ) = L { (t )} . Taking the Laplace transforms of Equations (b) and using the properties of linearity of the transforms and transforms of derivatives and assuming all initial conditions are zero leads to 1500s 2 + 1700s + 2.8x105 X (s) 900s + 1.3x105 320s + 9.1x104 Y1 (s) = 320s + 9.1x104 250s 2 + 1928s + 3.367x105 (s) 720s 1.04x105 (c) Cramers rule is used to determine the solution of Equation (c) as 900 s + 1.3x10 5 Y ( s ) 320 s + 9.1x10 4 720 s 1.04 x10 5 Y ( s ) 250s 2 + 1928s + 3.367 x10 5 X ( s) = 1500s 2 + 1700s + 2.8 x10 5 320 s + 9.1x10 4 320 s + 9.1x10 4 250s 2 + 1928s + 3.367 x10 5
s 3 + 3.45 x10 7 s 2 + 6.52 x10 8 s + 5.32 x1010 Y ( s ) 3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 5.78 x10 8 s 2 + 1.05 x1010 s + 8.60 x1011
5
(2.25x10
(d)
( s) =
1500s 2 + 1700 s + 2.8 x10 5 320s + 9.1x10 4 1500s 2 + 1700 s + 2.8 x10 5 320 s + 9.1x10 4
6
s 3 1.58 x10 8 s 2 5.02 x10 8 s 4.10 x1010 Y ( s) 3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 5.78 x10 8 s 2 + 1.05 x1010 s + 8.60 x1011 The transfer functions are determined from Equations (d) and (e) as =
( 1.08x10
(e)
195
Chapter 6
G1 ( s ) =
X (s) 2.25 x10 5 s 3 + 3.45 x10 7 s 2 + 6.52 x10 8 s + 5.32 x1010 Y ( s ) = (f ) Y ( s ) 3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 5.78 x10 8 s 2 + 1.05 x1010 s + 8.60 x1011
G2 ( s ) =
( s) 1.08 x10 6 s 3 1.58 x10 8 s 2 5.02 x10 8 s 4.10 x1010 Y ( s) = Y ( s) 3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 5.78 x10 8 s 2 + 1.05 x1010 s + 8.60 x1011
(g)
(a)
t t di 1 1 i1 dt + i2 dt + L 2 + Ri2 = 0 (b) C C0 dt 0 Taking the Laplace transform of Equations (a) and (b), assuming all initial conditions are zero leads to 1 1 RI 1 ( s ) + I1 (s) I 2 (s) = V (s) (c) Cs Cs 1 1 I1 ( s) + I 2 ( s ) + LsI 2 ( s ) + RI 2 ( s ) = 0 (d) Cs Cs Equations (c) and (d) are summarized in matrix form by 1 1 I 1 ( s ) V ( s ) R + Cs Cs (e) = 1 1 I 2 ( s ) 0 Ls + R + Cs Cs Cramers rule is used to solve for I 1 ( s) as
196
Chapter 6
1 Cs
V ( s) 0 I 1 ( s) = R+ 1 Cs 1 Cs
Ls + R +
1 Cs
1 Cs
Ls + R +
1 Cs
(f)
R+ I 2 (s) =
1 V ( s) Cs 1 0 Cs 2R C
2R C (g)
2R RLs + (R + LC )s + C
2
197
Chapter 6
G1 ( s ) = I1 (s) V ( s) Ls 2 + Rs +
2
2R RLs + R + LC s + C I ( s) G2 ( s ) = 2 V ( s) 1 C = 2R RLs 2 + R 2 + LC s + C
2
1 C
(h)
(i)
6.3 The differential equations derived in Example 3.22 governing the current in the field circuit and angular velocity in a field controlled dc servomotor are di f Lf + Rf if = vf (a) dt d J + ct K f i f = T (b) dt Taking the Laplace transform of Equations (a) and (b), setting T=0 and assuming all initial conditions are zero leads to L f sI f ( s) + R f I f ( s ) = V f ( s) (c)
Js( s ) + ct ( s) K f I f ( s) = 0
Equation (c) is rewritten as
I f ( s) = V f (s) Lf s + Rf
(d)
(e)
G1 ( s ) =
I f ( s) V f (s) (f)
(g)
198
Chapter 6 G2 ( s) = = ( s ) V f ( s) Kf ( L f s + R f )( Js + ct ) (h)
6.4 Equations (s), (t), and (u) of Example 4.17 are dC Ap E + qC Ap + Ve E /( RTs ) C Ap + T p C As = 0 V dt RTs2 dC Bp E Ve E /( RTs ) C Ap + T p C As + qC Bp = 0 V 2 dt RTs
(a) (b)
& & E + qcv T p Ve E /( RTs ) C Ap + T p C As = Q Q1 (c) 2 dt RTs Taking the Laplace transforms of Equations (a)-(c) assuming all initial conditions are zero leads to E VsC ap ( s ) + qC ap ( s ) + Ve E /( RTs ) C Ap ( s ) + C As T p ( s ) = 0 (d) 2 RTs E VsC Bp ( s ) Ve E /( RTs ) C Ap ( s ) + C As T p ( s ) + qC Bp ( s ) = 0 (e) 2 RTs E cV VsT p ( s ) + qcv T p ( s ) Ve E /( RTs ) C Ap ( s ) + C As T p ( s ) = Q ( s ) (f) 2 RTs Equations (d)-(f) are summarized in matrix form as EC E /( RTs ) VeE /( RTs ) As 0 Vs + q + Ve 2 RT s C (s) 0 Ap ECAs E /( RTs ) E /( RTs ) Vs + q Ve Ve CBp (s) = 0 RT 2 T (s) Q(s) s p E /( RTs ) E /( RTs ) ECAs cvVs + qcv Ve 0 Ve RT 2 s
cvV
dT p
199
Chapter 6
ECAs 2 RTs EC Ve E /( RTs ) Ve E /( RTs ) As 0 RT 2 s EC 0 cvVs + qcv VeE /( RTs ) As Q(s) RT 2 s C Ap (s) = EC 0 VeE /( RTs ) As Vs + q + Ve E /( RTs ) RT 2 s EC Ve E /( RTs ) Ve E /( RTs ) As Vs + q RT 2 s EC Ve E /( RTs ) 0 cvVs + qcv VeE /( RTs ) As RT 2 s 0 0
VeE /( RT )
s
[Ve
= where
E /( RTs ) 2
D(s)
] EC RT
As 2 s
Q(s)
(h)
] c Vs + qc
v
EC As Ve E /( RTs ) RT 2 s
(i)
Vs + q + Ve E /( RTs ) Vs + q Ve E /( RTs ) C Bp ( s ) =
Ve E /( RT )
s
[Ve
=
E /( RTs ) 2
D( s)
] EC RT
As 2 s
Q( s )
(j)
200
Chapter 6
Vs + q + Ve E /( RTs ) Vs + q Ve E /( RTs ) 0 Ve E /( RTs ) 0 0 0 Q( s )
T p (s) = =
Ve E /( RTs ) Vs + q + Ve E /( RTs ) Q( s ) D( s )
D( s)
(k)
The transfer functions are obtained from Equations (i), (j), and (k) as C Ap ( s) G1 ( s) = Q( s)
[Ve
= G2 ( s) = C Bp ( s ) Q( s )
E /( RTs ) 2
D( s)
] EC RT
As 2 s
(l)
[Ve
= G3 ( s ) = = T p ( s) Q( s )
E /( RTs ) 2
D( s )
] EC RT
As 2 s
(m)
Ve E /( RTs ) Vs + q + Ve E /( RTs ) D( s)
(n)
6.5 Equations (k) and (l) of Example 4.25 are d & & o cori 2 L o + (2ro LU + co mo ) o 2roUL w = co mo oi (a) dt d & & w c w r22 ro2 L w 2ro LU o + (2ro LU + c w mw ) w = c w mw wi (b) dt The desired transfer functions are obtained by setting oi = 0 and taking the Laplace transforms of Equations (a) and (b) assuming all initial conditions are zero, & o c o ri 2 Ls o ( s ) + (2 ro LU + c o m o ) o ( s ) 2 ro UL w ( s ) = 0 (c)
(d)
201
Chapter 6 o (s) o cori 2 Ls + (2ro LU + co mo ) & 2roUL = 2 2 & w c w (r2 ro )Ls + (2ro LU + c w mw ) w ( s) 2roUL 0 (e) c m ( s ) w w & wi Application of Cramers rule to solve Equation (e) leads to 0 2roUL 2 2 & & cw mw wi (s) w cw (r2 ro )Ls + (2ro LU + cw mw ) o ( s) = & 2roUL o cori2 Ls + (2ro LU + co mo ) 2 2 & 2roUL wcw (r2 ro )Ls + (2ro LU + cw mw )
= = where & [ c r Ls + (2r LU + c m )][ c (r
2 o o i o o o w w
(f)
(g)
G2 ( s) = =
[ c r
o o
w ( s) wi ( s)
2 i
(j)
6.6 The transfer function derived in Problem 6.1 is (s) 1.08 x10 6 s 3 1.58 x10 8 s 2 5.02 x10 8 s 4.10 x1010 G2 ( s) = = Y ( s ) 3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 6.24 x10 8 s 2 + 1.11x10 9 s + 9.451010
(a)
The system is stable if all roots of the polynomial in the denominator have negative real parts. To this end consider
202
Chapter 6
3.75 x10 5 s 4 + 3.32 x10 6 s 3 + 6.24 x10 8 s 2 + 1.11x10 9 s + 9.4510 10 = 0 s 4 + 8.8533 s 3 + 3.17 x10 3 s 2 + 2.96 x10 3 s + 2.52 x10 5 = 0 Equation (6.29) is applied to Equation (b) leading to a Rouths matrix of the form 1 3.17 x10 3 2.52 x10 5 3 0 8.8533 2.96 x10 B = 2.84x10 3 2.52x10 5 0 3 0 0 2.17x10 2.52x10 5 0 0 Calculations in obtaining elements of B are b2,1b1, 2 b1,1b2, 2 b3,1 = b2,1
(8.8533)(3.17 x10 3 ) (1)(2.96 x10 3 ) 8.853 3 = 2.841x10 b2,1b1,3 b1,1b2,3 b3, 2 = b2,1 = (8.8533)(2.52 x10 5 ) (1)(0) 8.853 5 = 2.52 x10 b2,1b3, 2 =
(b)
(c)
(d)
(e)
b4,1 = =
b3,1b2, 2
b3,1
(2.84 x10 3 )(2.96 x10 3 ) (8.853)(2.52 x10 5 ) 2.84 x10 3 = 2.17 x10 3 b3,1b2,3 b2, 2 b3,3 b4, 2 = b3,1 b5,1 = =
(f)
=0 b4.1b3.2 b3,1b4, 2
b4,1
(g)
(2.17 x10 3 )(2.52 x10 5 ) (2.84 x10 3 )(0) 2.17 x10 3 = 2.52 x10 5 Since there are no sign changes in the first column of B the system is stable.
(h)
6.7 The transfer function of a third-order system is s+4 G(s) = 3 s + 8s 2 + 15s + 20 203
(a)
Chapter 6 Rouths method is applied in the solution of Example 6.9 to derive a stability criteria for a third-order system. If the denominator of the transfer function is of the form D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 (b) the system is stable if (c) a1 a 2 > a3 Equation (c) is checked for the transfer function of Equation (a) by (8)(15) > 20 ? 120 > 20 (d) Equation (d) shows that the system is stable. 6.8 The transfer function of a second-order system is K i (4 s + 25) G ( s) = 3 (a) 2 s + 5s + (10 + 4 K i ) s + 25K i (a) The stability of a third-order system is dependent upon the coefficients of the denominator of its transfer function. If D ( s ) = s 3 + a 2 s 2 + a1 s + a 0 the system is stable if all coefficients are positive and a1 a 2 > a 0 . Applying this criteria to Equation (a) leads to 5(10 + 4 K i ) > 25K i K i < 10 Thus the system is stable when K i < 10 and unstable if K i > 10 . (b) For K i = 5 , Equation (a) becomes
G(s) = 20 s + 125 s + 5s 2 + 30 s + 125
3
(b)
(c)
204
(d)
(e)
The impulsive response for the system whose transfer function is Equation (e) is
205
Chapter 6
The impulsive response shown above is clearly unstable. (c) The denominator of Equation (a) is written in the form of Equation (6.32) as D( s ) = Q( s) + K i R( s )
= s 3 + 5s 2 + 10s + K i (4s + 25) The root-locus diagram is generated by the MATLAB worksheet commands >> clear >> Q=[1 5 10 0]
Q= 1 5 10 0
(f)
Chapter 6
2s + 4 (a ) s + 4 s 2 + 5s + 3 (a) The stability of a third-order system is dependent upon the coefficients of the denominator of its transfer function. If D ( s ) = s 3 + a 2 s 2 + a1 s + a 0 the system is stable if
all coefficients are positive and a1 a 2 > a 0 . Applying this criteria to Equation (a), (5)( 4) > 3 ( b) Thus the system is stable. (b) The stability of the system with respect to s = 1 is tested by examining the stability of a system with D(s-1). To this end D ( s 1) = ( s 1) 3 + 4( s 1) 2 + 5( s 1) + 3
= s 3 3s 2 + 3s 1 + 4 s 2 8 s + 4 + 5 s 5 + 3 = s3 + s2 + 2 (c) Since the coefficient of the s term is zero in Equation (c) the system is not stable with respect to s=-1.
G ( s) =
(a)
(a) The denominator of Equation (a) is written in the form of Equation (6.32) as D( s ) = Q( s) + K i R( s )
(f)
Chapter 6 The root-locus diagram is generated by the MATLAB worksheet commands >> clear >> Q=[1 5 10 0] Q= 1 5 10 0
>> R=[4 25] R= 4 25 >> rlocus(R,Q) >> The MATLAB generated root locus diagram is
The above root locus curve shows that the system is stable for K i < 10.2 and unstable for larger values. 6.11 The transfer function for the fourth-order plant is s+2 G(s) = s ( 2 2)( s 2 + 6 s + 5) The closed loop transfer function is
(a)
208
Chapter 6 KG ( s) 1 + KG ( s) Substitution of Equation (a) into Equation (b) leads to s+2 K s( s 2)( s 2 + 6 s + 5) H ( s) = s+2 1+ K s( s 2)( s 2 + 6s + 5) K ( s + 2) = 2 s( s 2)( s + 6s + 5) + K ( s + 2) The denominator of Equation (c) is of the form of Equation (6.32) with Q( s) = 2( s 2)( s 2 + 6s + 5) R( s) = s + 2 The MATLAB commands used to generate the root locus diagram for the system are given below H (s) = >> R=[1 2] R= 1 2
(b)
209
Chapter 6
Since one of the poles of Q(s) is positive, s=2, one branch of the root locus diagram begins at s=2. A second branch originates at s=0. These branches coincide at s=1.16, which is a breakaway point. This corresponds to K=4.1. Two branches of the root locus for larger values of K are in the right-half plane. Thus the system is unstable for all values of K. 6.12 The transfer functions for the system of Example 5.37 are obtained from 5 18000s + 5 H 1 ( s ) F1 ( s) = 30000s + 9 H 2 ( s ) F2 ( s) 5 One transfer function is 5 1 5 30000 s + 9 G1,1 ( s ) = 5 18000 s + 5 5 30000 s + 9
= 30000 s + 9 5.4 x10 s 2 + 3.12 x10 5 s + 20
8
(a)
(b)
The closed loop transfer function is KG ( s ) 1 + KG ( s) Substitution of Equation (b) in Equation (c) leads to H (s) = (c)
210
Chapter 6 30000s + 9 5.4 x10 s 2 + 3.12 x10 5 s + 20 H (s) = 30000s + 9 1+ K 8 2 5.4 x10 s + 3.12 x10 5 s + 20 K (30000s + 9) = 8 2 5.4 x10 s + 3.12 x10 5 s + 20 + K (30000s + 9) Equation (d) is of the form of Equation (6.32) with Q( s ) = 5.4 x10 8 s 2 + 3.12 x10 5 s + 20 + K (30000s + 9) R( s) = 30000s + 9 The MATLAB commands to generate the root locus diagram for this system are K
8
211
Chapter 6 The root locus diagram shows that the system is stable for all values of K. Root locus diagrams for the other transfer functions for this system can be developed. The forms of H(s) resulting from using other transfer functions all have the same Q(s) as Equation (e), but with different forms of R(s). For G1, 2 ( s ) and G 2,1 ( s ) , R(s)=5 while for G 2, 2 ( s ) , R(s)=18000s+5. However root locus diagrams drawn for all cases show that all transfer functions are stable for all values of K. 6.13 Application of KVL to each loop of the circuit leads to t 1 50i1 (i1 i2 )dt + v(t ) = 0 0.2 x10 6 0
100i2 1 (i1 i2 )dt = 0 0.2 x10 6 0
t
(a ) (b)
Define I 1 ( s) = L{i1 (t )}, I 2 ( s) = L{i2 (t )} and V ( s) = L{v(t )} . Taking the Laplace transforms of Equations (a) and (b), using the properties of linearity and transform of integrals leads to 5 x10 6 50 I 1 + ( I 1 ( s) I 2 ( s )) = V ( s) (c) s 5 x10 6 (I1 (s) I 2 (s)) = 0 100 I 2 ( s ) (d) s Equations (c) and (d) are summarized in matrix form as 5 x10 6 5 x10 6 50 + I 1 ( s ) V ( s ) s s (e) = 6 6 5 x10 I 2 ( s ) 0 5 x10 100 + s s Cramers rule is used on Equation (e) leading to 5 x10 6 V ( s) s 5 x10 6 5 x10 6 V ( s ) 100 + 0 100 + s s I 1 ( s) = = 8 8 6 6 2.5 x10 5 x10 2.5 x1013 2.5 x1013 5 x10 5 x10 5 x10 3 + + + 50 + s s s2 s2 s s 6 6 5 x10 5 x10 100 + s s 6 (100s + 5x10 )V (s) (f) = 5 x10 3 s + 7.5 x10 8 The transfer function is determined from Equation (f) as I (s) 100s + 5 x10 6 (g) G ( s) = 1 = V ( s) 5 x10 3 s + 7.5 x10 8
Equation (g) is put in the standard form for the transfer function of a first-order system as
212
Chapter 6 0.02 s + 5 x10 3 s + 1.5 x10 5 (b) The circuits time constant is determined from Equation (h) as 1 T= 1.5 x10 5 = 6.67 x10 6 s (c) If F (t ) = 10u (t ) then 10 I 1 ( s) = G ( s) s 0.2 s + 5 x10 4 = s s + 1.5 x10 5 A partial fraction decomposition of Equation (j) leads to A B I 1 ( s) = + s s + 1.5 x10 5 0.33333 0.13333 = s s + 1.5 x10 5 Inversion of Equation (k) leads to 5 i1 (t ) = 0.3333u (t ) 0.1333e 1.5 x10 t G ( s) =
(h)
(i)
(j)
(k)
(l)
6.14 The 50 and 60 resistors are in parallel and have an equivalent resistance of 1 = 27.3 Req = 1 1 + 50 60 The electric potential in the capacitor before the switch is closed is Q 1.3x10 3 C V (0) = = = 433.3 V C 3x10 6 F Applying KVL around the resulting circuit, after the switch is closed leads to t 1 20i 27.3i idt 433.3 = 0 3 x10 6 0 Equation (c) is rearranged to
(a)
(b)
(c)
(d)
(e)
213
Chapter 6 Inversion of the transform in Equation (e) leads to 3 i(t ) = 9.16e 7.04 x10 t The voltage change across the capacitor is
(f)
(g)
Using Equation (f) in Equation (g) gives vc = 3.33x10 5 (9.16e 7.04 x10 t )dt + 433.3
3
+ 433.3
(h )
The voltage change across the capacitor when the charge is Q = 5 x10 C is Q 5 x10 6 C vc = = = 1.67 V . The time when this occurs is determined using Equation (h) C 3x10 6 F 3 1.67 V = 433.3e 7.04 x10 t V t= 1 1.67 ln 3 7.04 x10 433.3 = 7.90 x10 4 s
(i)
6.15 Currents are as defined below. Define v R as the voltage drop across the resistor in the direction of the current and v L as the voltage change across the inductor in the direction of the current.
iR + iL = i Application of KVL around the loop leads to vL vR = 0 Equation (b) implies that v = vR = vL
214
Chapter 6 The currents are related to the voltage changes across the circuit components by v iR = R t 1 i L = vdt L0 Substitution of Equations (d) and (e) in Equation (a) leads to t v 1 + vdt = i R L0 Taking the Laplace transform of Equation (f) leads to 1 1 V ( s) + V ( s) = I ( s) R Ls RsI ( s) V (s) = R s+ L The transfer function is determined as V ( s) G ( s) = I ( s) Rs = R s+ L The impulsive response is the inverse of the transfer function which is rewritten as R 1 L G(s) = R s+ R L (a) Equation (i) is inverted leading to R R t vi (t ) = R (t ) e L L (b) The current through the inductor is obtained using Equation (e) as t 1 i L = vdt L0
R t R R Lt = (t ) e dt L 0 L R R R Lt = u (t ) + e 1 L L Noting that for t>0, u(t)=1, Equation (k) can be rewritten as R R t i L = e L u (t ) L
(d) (e)
(f)
(g)
(h)
(i)
(j)
(k)
(l)
215
Chapter 6 The presence of the unit step function in Equation (l) is necessary to show that the current through the inductor is discontinuous at t=0. 6.16 It is given that
G(s) = 2 4 s + 15 0 .5 = s + 3.75
(a)
A partial fraction decomposition leads to 1 0.5 G ( s) = s s ( s + 3.75) 0.1333 0.1333 = s s + 3.75 Equations (c) and (d) lead to x s (t ) = 0.1333u (t ) 0.1333e 3.75t (c) The systems ramp response is 1 x s (t ) = L1 2 G ( s ) s A partial fraction decomposition leads to 1 0.5 G ( s) = 2 2 s s ( s + 3.75) 0.0356 0.1333 0.0356 = + + s s + 3.75 s2 Inversion of Equation (g) leads to x r (t ) = 0.0356u (t ) + 0.1333t + 0.0356e 3.75t (d) The time constant for the system is 1 T= 3.75 = 0.2667 s The 2% settling time for a first-order system is t s = 3.92T
(d)
(e)
(f)
(g) (h)
(i)
= 1.05 s (j) (e) The 10% rise time is the time required for the step response to reach 10% of its final value. Thus from Equation (e)
216
Chapter 6 (0.1)(0.1333) = (0.1333) (0.1333)e 3.75t10 1 ln(0.9) t10 = 3.75 = 0.0281 s The 90% rise time is similarly determined by 1 t 90 = ln(0.1) 3.75 = 0.614 s The 10%-90% rise time is t r = t 90 t10 = 0.586 s
(k)
(l)
(m)
6.17 If the drug is again infused 2 hr after being expended the rate of infusion as a function of time is given by I (t ) = 0.302[u (t ) u (t 12) + 0.302[u (t 14) u (t 260] (a) (a) Substitution of Equation (a) into Equation (a) of Example 6.35 leads to dC 9.52 + C = 0.0719[u (t ) u (t 12) + u (t 14) u (t 26)] (b) dt Taking the Laplace transform of Equation (b), assuming the initial condition is zero leads to 7.55 x10 3 1 e 12 s + e 14 s e 26 s C ( s) = (c) s( s + 0.105) Inversion of Equation (c) leads to C (t ) = 7.19 x10 2 1 e 0.105t (1 e 0.105( t 12 ) )u (t 12) + (1 e 0.105( t 14) )u (t 14)
(1 e 0.105( t 26) )u (t 26)] mg/L (b) The general form for the rate of infusion is I (t ) = 0.302[u (t 14(k 1)) u (t 12 14(k 1))]
k
(d)
(e)
Using the solution procedure of Example 6.35 the time dependent concentration of the drug in the plasma is C (t ) = 7.19 x10 2 {1 e 0.105( t 14 ( k 1)) u (t 14(k 1))
(f)
6.18 The differential equation for the perturbation in liquid level is dh 1 p A + h= gR dt R Assuming the flow is turbulent through the pipe
(a)
217
Chapter 6 R=2 hs qs
8m s = 17.78 2 (b) 3 m m 0.9 s 2 (2.1 m )2 = 3.44 m 2 . Substitution of given and D The area of the tank is A = = 4 4 calculated data into Equation (a) leads to N 2500 2 dh 1 m h= u (t ) 3.44 + kg m s dt 17.78 1000 3 9.81 2 17.78 2 m s m dh 3.44 + 0.0562h = 0.0143 u (t ) (c) dt Taking the Laplace transform of Equation (c) leads to 0.0143 3.44 sH ( s ) + 0.0562 H ( s ) = s 0.0143 H (s) = s (3.44 s + 0.0562) =2
4.16 x10 3 s ( s + 1.63 x10 2 ) (a) The final value of the liquid level perturbation is h f = lim sH ( s) =
s 0
(d)
4.16 x10 3 = 0.255 s 0 s + 1.63 x10 2 Thus the liquid level when the tank reaches steady state is 8.255 m. (b) The time dependent response of the system is obtained by inverting 1 4.16 x10 3 1 = 0.255 H ( s) = 2 2 s( s + 1.63x10 ) s s + 1.63 x10 Inversion of Equation (f) leads to 2 h(t ) = 0.255 1 e 1.63 x10 t = lim
(e)
(f)
(g)
(c) The 4% settling time is the time required for the response to be permanently within 4% of its final value. It is obtained from Equation (g) by 2 0.96(0.255) = 0.255 1 e 1.63 x10 ts (h) Equation (h) is solved leading to 2 .04 = e =1.63 x10 t s ln(0.04) ts = = 197.5 s (i) 1.63 x10 2
218
Chapter 6 6.19 The properties of the pipe are assumed to be lumped. The mathematical model is 1 dT 1 + T = T0 (a) mc p dt R R where R is the total resistance of the pipe. It is assumed that the air in the interior of the pipe is at a uniform temperature. The mass of air in the pipe is m = V
= ri2 L slugs = 2.38 x10 3 (1 ft )2 (20 ft ) 3 ft = 0.150 slugs Btu Btu = 1.158 mc p = (0.150 slugs) 7.72 (b) slugs F F The total resistance is obtained using Equation 4.109. The R value is a measure of thermal conductivity. From Equation (4.102) the resistance of a wall of area A, thickness w made of a material of thermal conductivity k is w (c) R= kA In Example 4.14 it is noted that the R value, call it R , is related to resistance through R (d) R= A Comparison of Equations (c) and (d) shows that the thermal conductivity is related to the R value by w k= (e) R Given that the insulation is 0.5 in thick its thermal conductivity is (0.5 in ) 1 ft 12 in k= hr ft 2 F 15 Btu Btu 2.78 x10 3 (f) ft hr F Btu Taking the thermal conductivity of the steel pipe as k s = 35.1 the total resistance ft hr F of the pipe, its insulation, and the convective resistance is obtained using Equation (4.109)
( )
219
Chapter 6
+ Btu 14 2 3.2 ft (20 ft) hr F ft 2 12 1 1 14.5 14 ln ln + Btu 14 2 35.1 Btu (20 ft) 12 2 2.78 x10 3 (20 ft) ft hr F ft hr F hr F (g) = 0.1672 Btu Substituting calculated values into Equation (a) leads to dT 1.158 + 5.98T = 5.98T0 (h) dt (a) The time constant for the system is T = mc p R (1 ft)(20 ft) = (1.158)(0.1672) = 0.194 hr (i) (b) Setting T0 = 2 and using the initial condition T(0)=0, taking the Laplace transform of Equation (i) leads to 5.98(2) 1.158s( s ) + 5.98( s ) = s 11.96 ( s ) = (1.158s + 5.98)s 10.33 = s ( s + 5.16)
1 1 = 2 s s + 5.16 (j)
R=
1 Btu 2 2.1 hr F ft 2
T (t ) = 2(1 e 5.16t ) (k) (c) The temperature in the interior of the pipe reaches 68.3 F when T=-1.7. To this end 1.7 = 2(1 e 5.16t ) (l) t = 0.367 hr
6.20 The mathematical model for the temperature distribution in the brake of Example 4.26 is dT 1 + hAT = hAT + T ( 0 Tt )1 u t 0 (a) gD 2 c p dt T 4 The angular velocity of the shaft due to a constant braking torque T is Tt (T ) = + 0 (b) I eq In order to bring the shaft to rest in 1 sec,
220
Chapter 6
T (1 s) rev 2 r 1 min + 350 2 min rev 60 s 1.25 slugs ft T = 45.8 ft lb 0=
(c)
(d)
Btu 1 hr 2 1 ft hA = 4.9 (2) (5 in) 2 hr ft F 60 s 12 in = 0.0284 Thus Equation (a) becomes dT (2 ) 0.022 + 0.0284T = 0.0284(25(1.8) + 32) + 45.8(350) 45.8t [1 u(t 1)] dt 60 dT 0.022 + 0.0284T = 1.68 + 1.68x103 2.10x103 t [1 u(t 1)] (e) dt Taking the Laplace transform of Equation (e), noting that the initial temperature is the same as the ambient, 1.68 1.68x103 2.10x103 4.2 x10 2 s 2.1x103 s e + 0.022( s 77) + 0.0248 = + + e (f) s s s s2 s2 1.69 1.68 x10 3 + 4.2 x10 2 e s 2.1x10 3 (1 e s ) (g) ( s ) = + 2 0.022s + 0.0248 s (0.022s + 0.0248) s (0.022s + 0.0248) Inversion of Equation (g) leads to (t ) = 77(1 e 1.13t ) + 6.77 x10 4 (1 e 1.13t ) + 1.5 x10 4 (1 e 1.13(t 1) )u (t 1)
(h)
6.21 The appropriate form of the differential equation is 1 dT gD 2 c p + hAT = hAT + T ( 0 Tt )[1 u (t 1) ] 4 dt + T ( 0 Tt )[u (t 3) u (t 4)] + T ( 0 Tt )[u (t 6) u (t 7)] + ...
(a)
6.22 The mathematical model for the change in pressure in a tank is derived in Example 4.10 as V dp p p s + = (a) Ra dt R R where p s is the difference in supply pressure from the initial pressure in the vessel and is the constant absolute temperature in the vessel. Taking the Laplace transform of both sides of Equation (a) leads to
221
Chapter 6 V 1 1 R sP( s ) + R P( s ) = R Ps ( s ) a 1 Ps ( s ) R (s) = P V 1 R s + R a Ra Ps ( s ) RV = R s+ a RV The transfer function is obtained from Equation (b) as P( s ) G ( s) = Ps ( s) (b)
Ra RV = R s+ a RV (a) The time constant is obtained from Equation (c) as RV T= Ra Substitution of given values in Equation (d) leads to 1 3 (20 m ) 16.0 ms T= J 287.0 (313 K ) kg K = 3.56 x10 3 s (b) Substitution of numerical values into Equation (c) leads to 280.7 G ( s) = s + 280.7 The supply pressure to the vessel has a sudden increase of 8000 Pa. Thus p s (t ) = 8000u (t ) Ps ( s ) = 8000 s
(c)
(d)
(e)
(f)
(g)
222
1 1 8000 s s + 280.7
p(t ) = 8000 u (t ) e 280.7 t Pa (i) (c) The pressure in the tank reaches 18 kPa when the change in pressure is 6 kPa. THe time required is determined using Equation (i) 6000 = 8000 1 e 280.7 t
0.75 = 1 e
280.7 t
(j)
6.23 The denominator of the transfer function of a second-order system has the general form 2 D ( s ) = s 2 + 2 n s + n (a) The transfer function may be written as 0.5s + 0.75 G(s) = 2 (b) s + 2.5s + 11.25 Comparison of Equation (b) with Equation (a) leads to n = 11.25
= 3.35 r s (c)
2 n = 2.5
2 .5 2(3.35) = 0.373
(d)
3s + 5 (a) s + 20 s + 500 is that of a second-order system with natural frequency n = 500 = 22.36 and damping 20 ratio = = 0.447 . Thus the system is underdamped. The systems damped 2(22.36) G ( s) =
2
223
Chapter 6 natural frequency is d = (22.36) 1 (0.447) 2 = 20.0 . It is also noted that n = (0.447)(22.36) = 10 (a)The impulsive response is obtained using linearity and Table 6.4 as 10 1 xi (t ) = 3e 10t cos(20t ) + sin(20t ) + 5 e 10t sin(20t ) 20 20 = e 10t [3 cos(20t ) + 1.75 sin(20t )] (b)
(b)The systems step response is obtained using linearity and Table 6.5 as 1 10 10t 10 t x s (t ) = 3 e 10t sin(20t ) + 5 1 e cos(20t ) + 20 e sin(20t ) 500
(c)
6.25 The differential equation for the system is & & 10 &+ 2000 x + 1x10 5 x = F (t ) (a ) x The systems transfer function is obtained from Equation (a) as 1 G ( s) = 2 10 s + 2000 s + 1x10 5 0 .1 (b) = 2 s + 200 s + 10000 r The systems natural frequency is n = 10000 = 100 and its damping ratio is s 200 = 1 . Thus the system is critically damped. = 2(100) & (a)The given initial conditions are x (0) = 0.002 and x(0) = 0 . With F(t)=0, Equation (a) can be rewritten as &+ 2 n x + n x = 0 & & 2 (c) x &+ 200 x + 10000 = 0 & & (d) x Taking the Laplace transform of Equation (d), using linearity, the property of transform of derivatives and applying the initial conditions leads to s 2 x ( s) s(.002) + 200 sx ( s) 200(.002) + 10000 x ( s ) = 0 0.002s + 4 x (s) = 2 s + 200 s + 10000 0.002 s + 4 = ( s + 100) 2 0.002( s + 100 100) + 4 = ( s + 100) 2 .002 2 = + (e) s + 100 ( s + 100) 2 Inversion of Equation (e) gives 224
Chapter 6 x(t ) = 0.002e 100t + 2te 100t (b) The systems impulsive response is obtained from Table 6.4 as xi (t ) = te 100t (c) The systems step response is obtained using Table 6.5 as 1 x s (t ) = 1 e 100t 100te 100 t 10000 (f)
(g)
(h)
6.26 The differential equation governing the angular displacement of the bar is derived in Example 2.14 resulting in Equation (d) repeated below 1 2& 1 2 & 4 2 mL & cL + kL = M (t ) + (a) 9 9 9 The free response is obtained by setting M(t)=0, leading to &+ c + 4k = 0 & & (b) m m The natural frequency is obtained from Equation (b) as 4k n = m = N 4 2.3x10 4 m 1.2 kg r s (c)
= 276.9
The damping ratio for the system is obtained from c 2 n = (d) m The system is overdamped when the damping ratio is greater than 1. The values of c for which the system is overdamped are obtained from Equation (d) as c > 2 m n
r = 2(1.2 kg ) 276.9 s N s 664.5 m
(e)
6.27 The differential equation governing the motion of the block of mass m1 in Example 2.17 is given in Equation (c) repeated below. I & & x (a) m1 r + 9m 2 r + &+ (c1 r + 9c 2 r )x + (k1 r + 9k 2 r )x = 0 r Substitution of given values leads to
225
Chapter 6
0.05 kg m 2 N s N s & & (1 kg)(0.1 m) + 9(0.2 kg)(0.1 m) + x (0.1 m) x (0.1 m) + 9 225 &+ 225 0.1 m m m N N + 1x10 5 (0.1 m ) + 91x10 5 (0.1 m ) x = 0 m m & & 0.780 &+ 225 x + 1x10 5 x = 0 x The systems natural frequency and damping ratio are obtained from Equation (b) as (b)
n =
= 358.1 2 n =
225 N s N s = 288.5 0.780 kg m kg m 288.5 = 2(358.1) = 0.403 (d) Since the damping ratio is less than one the system is underdamped. It is desired to find the free response of the system when the disk is rotated 1 clockwise from equilibrium and then released. The dependent variable is x, the downward displacement of the block of mass 1 kg, measured positive downward from the systems equilibrium position. The given rotation of the disk leads to the block moving upward such that 2 r x(0) = (0.1 m )(1) 360 = 1.75 mm (e) Since the system is released from rest & x ( 0) = 0 (f) The free-response of an underdamped second-order system is given by Equations (6.68)(6.71). Direct substitution into these equations leads to
d = n 1 2
r = 358.1 1 (0.403) 2 s r = 327.7 s
(g)
226
Chapter 6
& x + n x0 Ad = x + 0 d
2 0
( 1.75 x10
= 1.91 mm
(h) x 0 d & x 0 + n x0
d = tan 1
(i)
x(t ) = Ad e
n t
sin ( d t + d )
The door rotates about a fixed axis through its hinges. Summing moments about this axis leads to
227
Chapter 6
M
O
(a)
Substitution of given values into Equation (a) leads to &+ & 25.2& ct + 30 = 0
ct & + 1.19 = 0 25.2 The natural frequency of the door is determined from Equation (b) as n = 1.19
&+ & = 1.09 (b)
r (c) s The system is critically damped, thus = 1 and 2 n = (2)(1)(1.09) = 2.18 . Thus Equation (b) is written as &+ & 2.18& 1.19 = 0 + (d) The free response of a critically damped system is obtained using Equation (6.76)
(t ) = 0 e t + (& + 0 n t )e 0
n
nt
(e)
2 r 360 = 0.873 r
(f) (g)
Substituting Equations (c), (f) and (g) into Equation (e) leads to (t ) = 0.873e 1.09t + (0.873)(1.09)te 1.09t
= (0.873 + 0.952t )e 1.09t The time required to for the door to close to 5 is obtained by (5) 2 r = (0.873 + 0.952t )e 1.09t 360 0.0873 = (0.873 + 0.952t )e 1.09t Equation (i) is solved using a trial and error procedure leading to t = 3.57 s (b) Since the door is closed at t=0,
(h)
(i)
(j)
( 0) = 0
and Equation (e) becomes
(k)
(t ) = 1.09&te 1.09t (l) 0 The door will open to 70 = 1.22 r if the velocity of the door is zero when the door reaches 70 . To this end &t ) = 1.09&(1 1.09t )e 1.09t ( (m) 0
228
Chapter 6 Equation (m) leads to the conclusion that the velocity is zero when 1 t= = 0.917 s 1.09 Substituting Equation (n) into Equation (l) and requiring (0.917 ) = 1.22 r leads to & 1.22 = 1.09 (0.917 )e (1.09 )( 0.917 )
0
(n)
& = 0.448 0
r s
(o)
The door rotates about a fixed axis through its hinges. Summing moments about this axis leads to M O = I O & & 2k c = I & &+ & I O& ct + 2k t = 0 Substitution of given values into Equation (a) leads to
&+ & 25.2& ct + 30 = 0
t t O
(a)
ct & + 1.19 = 0 25.2 The natural frequency of the door is determined from Equation (b) as
&+ &
(b)
n = 1 .19
= 1 .09 r s (c)
The system has a damping ratio of = 1.15 and 2 n = (2)(1.15)(1.09) = 2.51 . Thus Equation (b) is written as &+ & 2.51& 1.19 = 0 + (d)
229
Chapter 6 The free response of an overdamped system is obtained using Equation (6.80)
(t ) =
where
(e)
) )
)
(f) (g)
(t ) = 1.55{[ 0.8275 0 0.917&]e 1.874 t + [1.473 0 + 0.917&]e 0.9029 t } 0 0 (a) The initial conditions are 2 r 0 = (0) = (50) 360 = 0.873 r & = (0) = 0
0
(h)
Substituting Equations (i) and (j) into Equation (h) leads to (t ) = 1.12e 1.874t + 1.99e 0.9029t The time required to for the door to close to 5 is obtained by (5) 2 r = 1.12e 1.874t + 1.99e 0.9029t 360 0.0873 = 1.12e 1.874t + 1.99e 0.9029t Equation (l) is solved using a trial and error procedure leading to t = 3.44 s (b) Since the door is closed at t=0,
(l)
(m)
( 0) = 0
and Equation (e) becomes
(n)
(o)
The door will open to 70 = 1.22 r if the velocity of the door is zero when the door reaches 70 . To this end &t ) = 1.42&(1.874e -1874t 0.9029 e 0.9029 t ) ( (p) 0 Equation (p) leads to the conclusion that the velocity is zero when
230
Chapter 6
1.874e -1874t 0.9029e 0.9029t = 0 0.9029 e 1.874t = 0.9029 t 1.874 e 0.9711t e = 0.482 1 t= ln (0.482) 0.9711 t = 0.752 s (q) Substituting Equation (q) into Equation (p) and requiring (0.917 ) = 1.22 r leads to & 1.22 = 1.42 (1.874e -1874(0.752) 0.9029e 0.9029( 0.752) )
0
r s
(s)
6.30 A free-body diagram of the vehicle drawn at an arbitrary instant is shown below
(a) Application of Newtons second law to the free-body diagram leads to & x F = m& & & & k ( y x) c( y x) = m& x & & m&+ cx + kx = cy + ky x & (a) (b) The systems transfer function is obtained as X (s) G ( s) = Y ( s) cs + k (b) = 2 ms + cs + k Substitution of given values into Equation (b) leads to 1x10 4 s + 3.2 x10 5 G ( s) = 500s 2 + 1x10 4 s + 3.2 x10 5 20s + 640 (c) = 2 s + 20s + 640 Equation (c) is the transfer function of a second-order system with a natural frequency of n = 640
= 25.3 r s (d)
231
2 n = 20
20 2 n
= 0.395 (e) Since the damping ratio is less than one the system is underdamped. The damped natural frequency is calculated as
d = n 1 2
= 23.2 r s (f)
The system response occurs due to an input of y (t ) = 0.0128u (t ) (g) Thus the response is x(t ) = 0.0128 x s (t ) (h) where x s (t ) is the step response of the system which is obtained using Table 6.6 and superposition as 640 n t x s (t ) = 20e nt sin( d t ) + cos( d t ) + n e nt sin( d t ) 1 e d d 10 t 10 t 10 t = 20e sin(23.2t ) + 27.5 1 e cos(23.2t ) + 0.430e sin(23.2t )
= 27.5 27.5e 10t cos(23.2t ) + 31.8e 10t sin(23.2t ) The system response is obtained using Equations (h) and (i) as x(t ) = 0.352 0.352e 10t cos(23.2t ) + 0.407e 10t sin(23.2t ) 6.31 The mathematical model for the system is & m&+ cx + kx = F (t ) x & 1 &+ 2 n x + n x = F (t ) & & 2 x m Where m=500 kg, the damping ratio is 0.28 and N 1.2 x10 6 m = 49.0 r/s n = 500 kg The input force is F (t ) = 500t[u (t ) u (t 2)] The convolution integral solution for the response is t 1 ( 0.28 )( 49.0 )( t ) x(t ) = sin[ d (t )] 500 [u( ) u( 2)]e m d 0 where the damped natural frequency is
(i) (j)
(a)
(b)
(c)
(d)
(e)
232
Chapter 6
t
(f)
The symbolic capabilities of MATLAB are used to evaluate the integral. For t<2 the commands and response are f= 213/10000*tau*exp(-137/10*t+137/10*tau)*sin(47*t-47*tau) >> y=int(f,tau,0,t) y= -1371507/287206147805+10011/23966900*t+1371507/287206147805*cos(47*t)*exp(137/10*t)-43053903/5744122956100*sin(47*t)*exp(-137/10*t) >> y1=vpa(y,3) y1 = -.478e-5+.418e-3*t+.478e-5*cos(47.*t)*exp(-13.7*t)-.750e-5*sin(47.*t)*exp(-13.7*t) For t>2 sec the appropriate commands for integration are >> z=int(f,tau,0,2) z= 2385611289/2872061478050*cos(47*t-94)*exp(137/10*t+137/5)+1802262651/7180153695125*sin(47*t-94)*exp(137/10*t+137/5)+1371507/287206147805*cos(47*t)*exp(-137/10*t)43053903/5744122956100*sin(47*t)*exp(-137/10*t) >> z1=vpa(z.3) >> z1=vpa(z,3) z1 = .831e-3*cos(47.*t-94.)*exp(-13.7*t+27.4)+.251e-3*sin(47.*t-94.)*exp(13.7*t+27.4)+.478e-5*cos(47.*t)*exp(-13.7*t)-.750e-5*sin(47.*t)*exp(-13.7*t) To summarize for t<2 sec x(t ) = 4.78 x10 06 + 4.18 x10 4 t + 4.78 x10 6 e 13.7 t cos(47t ) 7.5 x10 6 e 13.7 t sin(47t ) While for t>2 sec x(t ) = 8.31x10 4 e 13,7 t + 24 cos(47t 94) + 2.51x10 4 e 13, 7 t + 24 sin( 47t 94) (g)
(h)
233
Chapter 6
& The transmitted force is kx + cx . MATLAB is used to differentiate the response and plot. The MATLAB program to plot the response is % Problem 6.31 t=1.3 dt=4/100; t=0 k=1.2e6 m=500 wn=(k/m)^0.5; c=2*.28*m*wn; for i=1:101 tt(i)=t; if t<2 y1(i)=-.478e-5+.418e-3*t+.478e-5*cos(47.*t)*exp(-13.7*t)-.750e-5*sin(47.*t)*exp(13.7*t); dx=.418e-3-.121910e-3*sin(47.*t)*exp(-13.7*t)-.417986e-3*cos(47.*t)*exp(13.7*t); F(i)=k*y1(i)+c*dx; t=t+dt; else y1(i)=.831e-3*cos(47.*t-94.)*exp(-13.7*t+27.4)+.251e-3*sin(47.*t-94.)*exp(13.7*t+27.4)... +.478e-5*cos(47.*t)*exp(-13.7*t)-.750e-5*sin(47.*t)*exp(-13.7*t); dx=-.424957e-1*sin(47.*t-94.)*exp(-13.7*t+27.4)+.4123e-3*cos(47.*t-94.)*exp(13.7*t+27.4)... -.121910e-3*sin(47.*t)*exp(-13.7*t)-.417986e-3*cos(47.*t)*exp(-13.7*t); F(i)=k*y1(i)+c*dx; t=t+dt; end end figure plot(tt,y1) xlabel('t (s)') ylabel('x (m)') title('Displacement vs. time for Problem 6.31') figure plot(tt,F) xlabel('t (s)') ylabel('F (N)') title('Transmitted force vs. time for Problem 6.31')
234
The displacement is not discontinuous at t=2. When the force is removed the system response is that of a system with a free response. The large natural frequency leads to large derivatives.
235
Chapter 6
Application of KCL at node B leads to i1 i2 i3 i3 = i1 i2 (a) Application of KVL around loop ABEF in the clockwise direction and using Equation (a) gives t q 1 Ri1 + (i1 i2 )dt + 0 = 0 (b) C0 C Application of KVL around loop BCDE in the clockwise direction and using Equation (a) leads to t q di 1 L 2 + Ri2 (i1 i2 )dt 0 = 0 (c) dt C0 C Taking the Laplace transform of Equations (b) and (c) noting that i2 (0) = 0 leads to q 1 (I 1 ( s ) I 2 ( s ) ) + 0 = 0 RI 1 ( s ) + Cs Cs 1 (I 1 ( s ) I 2 ( s ) ) q 0 = 0 LsI 2 ( s ) + RI 2 ( s ) Cs Cs Equations (d) and (e) are summarized in matrix form as 1 1 q0 R + Cs I 1 ( s ) Cs Cs = 1 1 I 2 ( s ) q0 Ls + R + Cs Cs Cs Cramers rule is used to solve for I 2 ( s) leading to
(d) (e)
(f)
236
(g)
0.2s 2 + 1000s + 2.5 x10 6 (i) s s 2 + 7500s + 2.5 x10 7 The quadratic polynomial in the denominator of Equation (i) has complex roots. A partial fraction decomposition of Equation (i) leads to = 0.0388
237
Chapter 6 0.1s 250 0.1 I 2 ( s ) = 0.0388 + 2 7 s + 7500s + 2.5 x10 s 0.1 0.1( s + 3750) 625 = 0.0388 s + ( s + 3750) 2 + 1.094 x10 7 Inversion of Equation (j) leads to i2 (t ) = 0.0388 0.1u (t ) + 0.1e 3750t cos(3307t ) 0.189e 3750t sin(3307t )
(j)
(k)
(b) It is given that L=0.2 H, C=0.4 F, q 0 = 3.1 C , and R=2000 . Substitution into Equation (h) leads to 2 3 .1x10 6 0 .2 s 2 + 2000 s + 6 0 .4 x10 I 2 (s) = 2000 1 2 ( 0 .2)( 2000 )( 0.4 x10 6 ) s s 2 + 0.2 + ( 2000 )( 0 .4 x10 6 ) s + ( 0 .2)( 0.4 x10 6 )
(l)
The quadratic polynomial in the denominator of Equation (l) has real roots of 8.20 x10 3 and 3.05 x10 3 . A partial fraction decomposition leads to 0.0106 0.1106 0 .1 I 2 ( s ) = 0.0194 (m) + 3 s + 8.20 x10 s + 3.05 x10 3 s Inversion of Equation (m) leads to 3 3 i2 (t ) = 0.0194 0.1u (t ) 0.0106e 8.20 x10 t + 0.1106e 3.05 x10 t (n) (c) It is given that L=0.2 H, C=0.4 F, q 0 = 3.1 C , and R=3000 . Substitution into Equation (h) leads to 2 3.1x10 6 0.2 s 2 + 3000 s + 6 0.4 x10 I 2 (s) = 3000 1 2 (0.2)(3000 )(0.4 x10 6 ) s s 2 + 0.2 + (3000 )(0.4 x10 6 ) s + (0.2)(0.4 x10 6 ) 2 6 0.2 s + 3000 s + 2.5 x10 = 0.0129 2 (o) s s + 1.58 x10 4 s + 2.5 x10 7
The quadratic polynomial in the denominator of Equation (l) has real roots of 1.40 x10 4 and 1.78 x10 3 . A partial fraction decomposition leads to 0.0015 0.0105 0 .1 I 2 ( s ) = 0.0129 (p) + 4 s + 1.58 x10 s + 1.78 x10 3 s Inversion of Equation (m) leads to 4 3 i2 (t ) = 0.0129 0.1u (t ) 0.0015e 1.58 x10 t + 0.0105e 1.78 x10 t (q)
238
Chapter 6
6.33 The circuit diagram with the switch closed is illustrated below.
(a)
1 i2 dt + R (i2 i1 ) = 0 (b) C 0 Taking the Laplace transforms of Equations (a) and (b), assuming all initial conditions are zero leads to 1 (c) Ls + 2 R + I 1 ( s ) RI 2 ( s ) = V ( s ) Cs 1 RI 1 ( s ) + R + (d) I 2 ( s) = 0 Cs Equations (c) and (d) are summarized in matrix form as 1 R I ( s ) V ( s ) Ls + 2 R + Cs 1 (e) = 1 I 2 ( s ) 0 R R+ Cs Cramers rule is used with Equation (e) giving
239
Chapter 6
(f)
(g)
(h)
(i)
1 2 2 1 R LRCs s 3 + + s+ s + LC LRC 2 L RC (a) The system input is a step input of the form v1 = 12u (t )
(j)
12 (k) s Substituting given numerical values of L=0.25 H, R=3.6 k and C=0.15 F as well as Equation (k) in Equations (h) and (j) leads to V1 ( s ) =
240
Chapter 6
= s + 1.86 x10 s + 5.33x10 7 s + 4.94 x1010 Use of Equation (m) in Equation (l) leads to 3.20 x10 8 s 2 + 5.93x1011 V2 ( s ) = 2 3 s s + 1.86 x10 3 s 2 + 5.33x10 7 s + 4.94 x1010 Use of MATLABs residue command leads to 6.095 s + 161.0 0.0130 12 6.108 V2 ( s ) = + 2 + + 2 2 s s s + 9.41x10 s + 914.5s + 5.24 x10 7 Inversion of Equation (o) leads to V2 ( s) = 0.0130u (t ) + 12t 2 + 6.108e 941t 6.095e 557 t cos(7.22 x10 3 t )
3 3 2
(m) (n)
(o)
(p)
Application of KCL at node A gives i1 i2 i3 = 0 (a) The amplifier is assumed to be ideal, such that v A = 0 and i3 = 0 . The current through the 300 k resistor is v1 i1 = (b) 3x10 5 The current through the parallel combination of the resistor and capacitor is
241
Chapter 6
v2 dv 20 x10 6 2 5 dt 6 x10 Substitution of Equations (b) and (c) into Equation (a) leads to i2 =
3.33 x10 6 v1 + 1.67 x10 6 v 2 + 20 x10 6 12 dv 2 =0 dt
(c)
dv 2 + v 2 = 2v1 (e) dt Taking the Laplace transform of Equation (e), assuming initial conditions are zero leads to 2V ( s ) (f) V2 ( s ) = 1 12s + 1 If v1 = 20u (t ) then 20 V1 ( s ) = (g) s Which when substituted into Equation (f) leads to
V2 ( s ) = 40 s (12 s + 1) 10 3 = 1 s s + 12 1 1 = 40 s s+ 1 12
t 12 v 2 (t ) = 40 u (t ) e V
(h)
(a) (b)
242
9.7 m 6.8 m m3 4.0 s s = 1.45 2 m h2 s R2 = 2 q1s + q 2 s 6.8 m m3 m3 4.0 +0 s s s = 3.4 2 m =2 The perturbation in flow rates are
q p1 (t ) = 0 q p 2 (t ) = 0.2u (t )
m3 s
(c)
(d)
(e) (f)
Substitution of Equations (c)-(f) in Equations (a) and (b) leads to dh 1 1 450 1 + h1 h2 = 0 (g) dt 1.45 1.45 dh 1 1 1 h1 + (h) 790 2 + h2 = 0.2u (t ) dt 1.45 1.45 3.40 Equations (g) and (h) are rearranged as dh 450 1 + 0.690h1 0.690h2 = 0 (i) dt dh 790 2 0.690h1 + 0.984h2 = 0.2u (t ) (j) dt Taking the Laplace transforms of Equations (i) and (j), taking all initial conditions to be zero leads to 0.690 H 1 ( s ) 0 450s + 0.690 (k) = 0.2 0.690 790s + 0.984 H 2 ( s ) s Cramers rule is applied to Equation (k) leading to
243
Chapter 6
0 0.690 0 .2 790 s + 0.984 s H 1 ( s) = 450 s + 0.690 0.690 790 s + 0.984 0.690 s (450 s + 0.690)(790 s + 0.984) (0.690) 2 0.138 = 2 s s + 987.9 s + 0.2029 0.138 = s(s + 987.9)(s + 2.05x10 -4 ) =
(0.690) 0.2
4.93 1.02 x10 6 4.93 = 0.138 s + s + 987.9 s + 0.2029 Inversion of Equation (l) gives h1 (t ) = 0.680u (t ) + 1.41x10 7 e 987.9t 0.680e 0.2029t Application of Cramers rule to Equation (k) also leads to
(l)
(m)
0 0 .2 s 0.690
790 s + 0.984
s (450 s + 0.690)(790 s + 0.984) (0.690) 2 90 s + 0.138 = 2 s s + 987.9 s + 0.2029 90s + 0.138 = s(s + 987.9)(s + 2.05x10 -4 )
(n)
(o)
244
Chapter 6 6.37 Free-body diagrams of the blocks at an arbitrary instant are shown below.
Application of Newtons second law applied to free-body diagrams of each of the blocks at an arbitrary instant leads to & x1 F = m1 &
& 2 x10 5 x1 + 3x10 5 ( x 2 x1 ) = 2 & x1 & 2 & + 5 x10 5 x1 3x10 5 x 2 = 0 x1 & x F = m&2 & 3 x10 5 ( x 2 x1 ) 1x10 5 x 2 = 3& x2 & x2 3& 3 x10 5 x1 + 4 x10 5 x 2 = 0 Equations (a) and (b) are summarized in matrix form as & x1 3 x1 0 2 0 & 5 5 0 3 & + 10 3 4 x = 0 & x2 2 (b) (a)
(c)
Application of the Laplace transform to Equation (c) leads to 2s 2 + 5 3 X 1 ( s ) 0 (d) 3s 2 + 4 X 2 ( s) 0 3 The natural frequencies are obtained by setting the determinant of the coefficient matrix of Equation (d) to zero leading to 2s 2 + 5 x10 5 3 x10 5 =0 3x10 5 3s 2 + 4 x10 5 (2s 2 + 5 x10 5 )(3s 2 + 4 x10 5 ) (3 x10 5 )(3x10 5 ) = 0 6s 4 + 23x10 5 s 2 + 11x1010 = 0
2 2
(e)
The solutions of Equation (e) are s = 2.36 x10 j , 5.72 x10 j . Thus the natural frequencies are r 1 = 2.36 x10 2 (f) s r 2 = 5.72 x10 2 (g) s
245
Chapter 6 6.38 Free-body diagrams of the block and the junction between the spring and viscous damper are shown below
Application of Newtons law to the free-body diagram of the block leads to F = ma & F (t ) k1 x k 2 ( x z ) = m& x & m&+ (k1 + k 2 ) x k 2 z = F (t ) x Summing forces at the junction between the spring and viscous damper leads to & cz + k 2 ( x z ) = 0 & k 2 x cz k 2 z = 0 Taking the Laplace transform of Equations (a) and (b) leads to ms 2 X ( x) + (k1 + k 2 ) X ( s) k 2 Z ( s ) = F ( s ) k 2 X ( s) csZ ( s) k 2 Z ( s) = 0 Equations (c) and (d) are rearranged and written in matrix form as ms 2 + k1 + k 2 k 2 X ( s ) F ( s ) = k2 cs + k 2 Z ( s ) 0 (a) Cramers rule is used to solve Equation (e) for X(s) leading to F (s) k 2 cs + k 2 0 X ( s) = k2 ms 2 + k1 + k 2 k2 cs + k 2 = = (cs + k 2 ) F ( s) (ms + k1 + k 2 )(cs + k 2 ) k 22
2 3
(a)
(e)
(cs + k 2 ) F ( s) mcs + mk 2 s 2 + c(k1 + k 2 ) s + k1 k 2 The desired transfer function is obtained from Equation (f) as
(f)
246
cs + k 2 mcs + mk 2 s + c(k1 + k 2 ) s + k1k 2 (b) Cramers rule is again applied to determine Z(s) as ms 2 + k1 + k 2 F ( s) 0 k2 Z (s) = 2 ms + k1 + k 2 k2 cs + k 2 k2 = = k 2 F (s) (ms + k1 + k 2 )(cs + k 2 ) k 22
2 3 2
(g)
k 2 F ( s) (h) mcs + mk 2 s + c(k1 + k 2 ) s + k1k 2 The transmitted force is calculated as & FT (t ) = k1 x + cz (i) Taking the Laplace transform of Equation (i) leads to FT ( s) = k1 X ( s) + csZ ( s) (j) Substitution of Equations (f) and (h) into Equation (j) gives (cs + k 2 ) F ( s ) k 2 F ( s) FT ( s ) = k1 + cs 3 2 3 2 mcs + mk 2 s + c(k1 + k 2 ) s + k1 k 2 mcs + mk 2 s + c(k1 + k 2 ) s + k1 k 2
(k) mcs + mk 2 s 2 + c(k1 + k 2 ) s + k1 k 2 The desired transfer function is determined from Equation (k) as F ( s) G2 ( s) = T F ( s) c ( k1 + k 2 ) s + k1 k 2 = (l) 3 mcs + mk 2 s 2 + c(k1 + k 2 ) s + k1 k 2 (c) The impulsive responses are the inverses of the transfer functions. These cannot be determined in closed form without specific values for all parameters.
3
6.39 A free body diagram of the system at an arbitrary instant is shown below assuming small ,
247
Chapter 6
Summing moments about the pin support M O = I O L & L & F (t ) L kL (L ) c = I O& (a) 2 2 The moment of inertia of the bar about O is obtained using the parallel axis theorem as
1 L mL2 + m 12 2 1 = mL2 3 Substitution of Equation (b) in Equation (a) leads to 1 2 & L2 & 2 mL & c + kL = LF (t ) + 3 4 &+ 3c + 3k = 3 F (t ) & & 4m m mL Taking the Laplce transform of Equation (c) leads to the transfer function X (s) G ( s) = F ( s) 3 mL = 3c 3k s2 + s+ 4m m Substitution of given values leads to 3 (10.8)(2) G ( s) = 3(2500) 3(4 x10 5 ) s2 + s+ 4(10.8) 10.8 0.139 = 2 s + 173.6 s + 1.11x10 5 The transfer function of Equation (e) is that of a second-order system with IO =
2
(b)
(c)
(d)
(e)
n = 1.11x10 5
= 333.3 r s (f)
248
2 n = 173.6
= 0.2604
Since the damping ratio is less than one, the system is underdamped. (a) If F (t ) = 10 (t ) ,the response is obtained from Table 6.5 as 1 nt x(t ) = (10)(0.139) e sin( d t ) mm d where the damped natural frequency is
(g)
(h)
d = n 1 2
r s Substituting known values into Equation (h) leads to 1.39 ( 0.2604)(333.3)t x(t ) = e sin(321.8t ) 321.8 = 4.32e 86.8t sin(321.8t ) The MATLAB worksheet used to obtain a plot of Equation (j) is = 321.8 N=[1.39] N= 1.3900 >> D=[1 173.6 1.11e5] D= 1.0e+005 * 0.0000 0.0017 1.1100 >> impulse(N,D) (i)
(j)
249
Chapter 6
(b) The step response is obtained using Table 6.6 (100)(0.139) nt x(t ) = cos(d t ) + n e nt sin(d t ) 1 e d n 2
(0.2604)(333.3) ( 0.2604)(333.3) t sin(321.8t ) 1 e cos(321.8t ) + 321.8 86.8t [cos(321.8t ) + 0.270sin(321.8t )] mm = 0.125 1 e 13.9 (333.3) 2
(k)
250
Chapter 6
6.40 The transfer function for the series LRC circuit is s G ( s) = 2 0.2 s + 2.6 s + 1x10 7 0.2 s (a) = 2 s + 13s + 5 x10 7 The transfer function of Equation (a) is that of a second-order system of natural 13 frequency n = 5 x10 7 = 7.07 x10 3 and damping ratio = = 9.19 x10 4 . 2(7.07 x10 3 ) Thus the system is underdamped with a damped natural frequency of
d = (7.07 x10 3 ) 1 (9.19 x10 4 ) 2 = 7.07 x10 3 . It is noted that n = 6.5 . (a)The response for v(t ) = 10 (t ) is obtained from Table 6.4 as
6 .5 i (t ) = 10e 6.5t cos(7.07 x10 3 t ) + sin( 7.07 x10 3 t ) 3 7.07 x10 The response is plotted using MATLAB by the commands N=[2 0] D=[1 13 5E7] impulse(N,D)
(b)
251
Chapter 6
(b) The response when v (t ) = 10u (t ) is obtained using Table 6.5 as i (t ) = e 6.5t sin(7.07 x10 3 t ) (c) The command step(N,D) is used along with the previously defined N and D in MATLAB to generate the step response shown below
(d)
252
Chapter 6 Partial fraction decomposition of Equation (a) leads to 2 x10 9 2 x10 9 s + 2.6 x10 8 I ( s) = 2 s s + 13s + 5 x10 7 2 x10 9 2 x10 9 ( s + 6.5 6.5) + 2.6 x10 8 = s ( s + 6.5) 2 + (7.07 x10 3 ) 2
2 x10 9 2 x10 9 ( s + 6.5) 1.3 x10 8 s ( s + 6.5) 2 + (7.07 x10 3 ) 2 ( s + 6.5) 2 + (7.07 x10 3 ) 2 Inversion of Equation (e) leads to i (t ) = 2 x10 9 2 x10 9 e 6.5t cos(7.07 x10 3 t ) 1.3x10 8 e 6.5t sin(7.07 x10 3 t ) The response is plotted from MATLAB using the following commands N=[0.1 0] D=[1 13 5E7 0] step(N,D) = (e)
(f)
253
Chapter 6 6.41 The currents in the branches of the circuit are as defined below
Application of KVL to each of the loops leads to t di1 7 0.3 + 1x10 [i1 (t ) i2 (t )]dt = v(t ) dt 0 0.1
t di2 1x10 7 [i1 (t ) i2 (t )]dt + 2.6 x10 3 (i2 i3 ) = 0 dt 0
(a) (b)
di3 2.6 x10 3 (i2 i3 ) + 1.6 x10 3 i3 = 0 (c) dt Taking the Laplace transforms of Equations (a)-(c), assuming all initial conditions are zero leads to 1x10 7 [I1 ( s) I 2 ( s)] = V ( s) 0.3sI 1 ( s) + (d) s 1x10 7 [I1 ( s) I 2 ( s)] + 2.6 x10 3 [I 2 ( s) I 3 (s)] = 0 0.1sI 2 ( s) (e) s 0.2sI 3 ( s) 2.6 x10 3 [I 2 ( s ) I 3 ( s )] + 1.6 x10 3 I 3 ( s) = 0 (f) Equations (d)-(f) are summarized in matrix form as 1x10 7 1x10 7 0 0.3s + s s I 1 ( s ) V ( s) 7 1x10 7 3 3 1x10 2.6 x10 I 2 ( s ) = 0 0.1s + 2.6 x10 + (g) s s 2.6 x10 3 0 0.2s + 3.6 x10 3 I 3 ( s) 0 Cramers rule is applied to Equation (g) in order to solve for I 3 ( s ) leading to 0.2
254
Chapter 6
1x10 7 s 1x10 7 s 0 1x10 7 s
0.3s +
V ( s) 1x10 7 s 0 0
I 3 ( s) =
0.3s +
1x10 7 s 1x10 7 s 0
)
(h)
where
1x107 1x107 0.2s + 3.6 x103 2.6 x103 0.1s + 2.6 x103 + 0.3s + D( s) = s s
) (
1x107 0.3s + s
2.60x1013 s Use of Equation (i) in Equation (h) and simplifying leads to 4.33x1012V ( s ) I 3 ( s) = 4 s + 4.40 x104 s 3 + 2.63x108 s 2 + 3.27 x1012 s + 4.33x1015 The system output is the voltage change across the 1600 resistor 1600 I 3 ( s) G ( s) = V ( s) = 0.006s 3 + 264s 2 + 1.58x106 s + 1.96x1010 + =
(i)
(j)
6.93x1015 (k) s 4 + 4.40 x10 4 s 3 + 2.63x10 8 s 2 + 3.27 x1012 s + 4.33x1015 1 The step response is obtained by inverting G ( s ) . Thus s 6.93x1015 (l) V2 s ( s ) = s (s 4 + 4.40 x10 4 s 3 + 2.63 x10 8 s 2 + 3.27 x1012 s + 4.33x1015 ) The following MATLAB worksheet is used to help determine a partial fraction decomposition of the right-hand side of Equation (l) >> D=[1 4.40e4 2.63e8 3.27e12 4.33e15 0]
255
Chapter 6 D= 1.0e+015 * 0.0000 0.0000 0.0000 0.0033 4.3300 >> [r,p,k]=residue(N,D) r= 0.0031 0.0578 + 0.1283i 0.0578 - 0.1283i -1.7192 1.6005 p= 1.0e+004 * -3.9358 -0.1594 + 0.8551i -0.1594 - 0.8551i -0.1454 0 k= [] >> syms s >> f1=(s-p(2))*(s-p(3)) f1 = (s+7010686265853559/43980465111042350353352750307/274877906944*i)*(s+7010686265853559/4398046511104+235035 3352750307/274877906944*i) >> f2=vpa(simplify(f1),3) f2 = .517e-25*(.440e13*s+.701e16-.376e17*i)*(.440e13*s+.701e16+.376e17*i) 0
256
Chapter 6
>> g1=r(2)*(s-p(3))+r(3)*(s-p(2)) g1 = (8330162633398495/144115188075855872+4622007775436625/36028797018963968*i )*(s+7010686265853559/4398046511104+2350353352750307/274877906944*i)+(8330 162633398495/1441151880758558724622007775436625/36028797018963968*i)*(s+7010686265853559/43980465111042350353352750307/274877906944*i) >> g2=vpa(simplify(g1),3) g2 = .116*s-.201e4 The above shows that the partial fraction decomposition is 0.0031 1.72 0.116 s 2.01x10 4 V2 s ( s ) = + 2 s + 3.94 x10 4 s + 1.45 x10 3 s + 3.19 x10 3 s + 7.56 x10 7 0.0031 1.72 0.116 s 2.01x10 4 (m) = + s + 3.94 x10 4 s + 1.45 x10 3 ( s + 1.60 x10 3 ) 2 + (8.55 x10 3 ) 2 Inversion of Equation (m) leads to 4 3 3 v 2 s (t ) = 0.0031e 3.94 x10 t 1.72e 1.45 x10 t + 0.116e 1.60 x10 t cos(8.55 x10 3 t ) - 2.37e 1.60 x10 t sin(8.55 x10 3 t ) 6.42 Application of KCL at node A leads to i1 i 2 i3 = 0 where
3
(n)
(a)
257
Chapter 6
i1 = v D + v1 v A R1
(b)
d (v A v B ) (c) dt v vE i3 = A (d) R1 Substitution of Equations (b)-(d) in Equation (a) leads to v vE v D + v1 v A d C (v A v B ) A =0 (e) R1 R1 dt Since the amplifier is assumed to be ideal, there is no current in the amplifier. Thus application of KCL at node B leads to v vC d =0 (f) C (v A v B ) B dt R2 Also at node C v B vC v C v D =0 (g) R2 R2 Application of KCL at node E, noting the amplifier is ideal leads to v A vE d C (v E v D ) = 0 (h) dt R1 Since the operator is ideal, (i) v E = vC The desired system output is v2 = v D vB (j) Taking the Laplace transforms of Equations (e), (f), (g), and (h), using Equation (i) leads to VD ( s ) + V1 ( s ) (2 + R1Cs )V A ( s) + R1CsVB + VC ( s ) = 0 (k) i2 = C (R2 Cs + 1)VB ( s) + R2 CsV A ( s) + VC ( s) = 0 V A ( s ) (1 + R1Cs )VC ( s ) + R1CsVD ( s ) = 0 Equation (m) leads to
1 (V B ( s ) + VD ( s) ) (o) 2 Using Equation (o) in Equations (k) and (n) and rewriting the resulting system in a matrix form leads to 1.5 V A ( s ) V1 ( s ) 2 + R1Cs 0.5 R1Cs V ( s ) = 0 R Cs 0.5 + R2 Cs (p) 0.5 2 B 1 0.5(1 + R1Cs ) 0.5(1 R1Cs ) VD ( s) 0 VC ( s ) =
VB ( s ) 2VC ( s ) + VD ( s ) = 0
258
Chapter 6 2 + R1Cs V1 ( s ) 1.5 0 R2 Cs 0.5 0 0.5(1 R1Cs ) 1 2 + R1Cs 0.5 R1Cs 1.5 0.5 + R2 Cs R2 Cs 0.5 0.5(1 + R1Cs ) 0.5(1 R1Cs ) 1 D( s) (q)
VB ( s) =
VD ( s) =
2 + R1Cs 0.5 R1Cs V1 ( s) 0.5 + R2 Cs 0 R2 Cs 0.5(1 + R1Cs ) 0 1 2 + R1Cs 0.5 R1Cs 1.5 0.5 + R2 Cs R2 Cs 0.5 0.5(1 + R1Cs ) 0.5(1 R1Cs ) 1 D( s) V2 ( s ) = V D ( s ) V B ( s ) = (r)
= Hence
D( s) [0.5 + 0.5R2 Cs(1 R1Cs)]V1 ( s) D( s) R2 CsV1 ( s ) = D( s) Evaluation of the determinant and substitution of numerical values leads to 1.6 x10 4 s V1 ( s ) V2 ( s ) = 2.19 x10 7 s 2 + 6.98 x10 4 s + 2.25 Equation (t) is rewritten as 740.7 V1 ( s ) V2 ( s ) = 2 s + 3.23 x10 3 s + 1.04 x10 7 The system is of second-order with natural frequency r n = 1.04 x10 7 = 3.23 x10 3 s and damping ratio 3.23x10 3 = = 0.5 2(3.23x10 3 ) Thus the system is underdamped and has a damped natural frequency of
(s)
(t)
(u)
(v)
(w)
259
Chapter 6
r s The step response for this second-order system is determined using Table 6.6 as 3 x s (t ) = 740.7e 1.62 x10 t sin( 2.80 x10 3 t )
(x)
(y)
6.43 The mathematical model for the two-tank system is dC A1 VA + (q + k1V )C A1 k 2V A C B1 = qC Ai dt dC B1 VA k1V A C A1 + (q + k 2V A )C B1 = 0 dt dC A2 VB qC A1 (t ) + (q + k1V B )C A 2 k 2V B C B 2 = 0 dt dC B 2 VB qC B1 (t ) k1V B C B 2 + (q + k 2V B )C B 2 = 0 dt Substitution of given values into Equations (a)-(d) leads to dC A1 1.4 x10 3 + 4.6 x10 6 C A1 1.68 x10 6 C B1 = 1.8 x10 6 Cu (t ) dt dC B1 1.4 x10 3 2.8 x10 6 C A1 + 3.48 x10 6 C B1 = 0 dt dC A2 1.4 x10 3 1.8 x10 6 C A1 (t 2.5) + 4.6 x10 6 C A 2 1.68 x10 6 C B 2 = 0 dt dC B 2 1.4 x10 3 1.8 x10 6 C B1 (t 2.5) 2.8 x10 6 C A2 + 3.48 x10 6 C B 2 = 0 dt Dividing each equation by the reactor volume and taking their Laplace transforms, assuming all initial conditions are zero leads to 1.29 x10 3 C 3 3 sC A1 ( s) + 3.33x10 C A1 ( s) 1.2 x10 C B1 ( s ) = s 3 3 sC B1 ( s ) 2 x10 C A1 ( s ) + 2.5 x10 C B1 ( s ) = 0
(i) (j)
sC A2 ( s ) 1.29 x10 e
3
2.5 s 2.5 s
(k) (l)
sC B 2 ( s) 1.29 x10 e C B1 ( s ) 2 x10 C A 2 ( s) + 2.5 x10 C B 2 ( s) = 0 Equations (i) and (j) are summarized in matrix form as 3 s + 3.33x10 3 1.2 x10 3 C A1 ( s) 1.29 x10 C = s 2 x10 3 s + 2.5 x10 3 C B1 ( s) 0 Cramers rule is used to solve Equation (m) resulting in
(m)
260
C A1 ( s) =
1.29 x10 3 C ( s + 2.5 x10 3 ) = s ( s + 3.33x10 3 )( s + 2.5 x10 3 ) (1.2 x10 3 )(2 x10 3 ) (1.29 x10 3 s + 3.23x10 6 )C = s s 2 + 5.83x10 3 s + 5.92 x10 6 1.29 x10 3 C s + 3.33x10 3 s 2 x10 3 0 C B1 ( s ) = s + 3.33x10 3 1.2 x10 3
]
(n)
2 x10 3
s + 2.5 x10 3
1.29 x10 3 C (2 x10 3 ) = s ( s + 3.33x10 3 )( s + 2.5 x10 3 ) (1.2 x10 3 )(2 x10 3 ) 2.58 x10 6 C = s s 2 + 5.83x10 3 s + 5.92 x10 6 Equations (n) and (o) are inverted leading to 3 3 C A1 (t ) = C 0.5456 0.1792 e 4.5 x10 t 0.3664 e 1.3 x10 t 3 3 C (t ) = C 0.4358 + 0.1778 e 4.5 x10 t 0.6136 e 1.3 x10 t
]
(o) (p)
(q) Substituting Equations (n) and (o) in Equations (k) and (l)< rearranging and writing in a matrix form leads to
B1
[ [
] ]
s + 3.33x10 3 3 2 x10
1.29 x10 3 s + 3.23x10 6 C 2 s s + 5.83x10 3 s + 5.92 x10 6 = 1.29 x10 e 2.5 s 2.58x10 6 C 2 3 6 s s + 5.83x10 s + 5.92 x10
) )
(r)
261
Chapter 6
C A2 ( s) =
s s + 5.83x10 s + 5.92x10
2
1.29x103 Ce 2.5s
3
6 2
{(1.29x10
s + 3.23x106 (s + 2.5x103 )
2.5 s
6 2
(s)
C B 2 ( s) =
6 2
{(1.29 x10
(t)
Inversion of Equations (s) and (t) leads to C A 2 (t ) = Cq (t 2.5)u (t 2.5) C (t ) = Cg (t 2.5)u (t 2.5)
B2
(u) (v)
where q(t) and g(t) are obtained using MATLAB as q= .411+.205e-31*exp(-.292e-2*t)*((-.201e32-.341e29*t)*cosh(.160e-2*t)1.*(.152e32+.117e29*t)*sinh(.160e-2*t)) g= .554+.205e-31*exp(-.292e-2*t)*((-.273e29*t-.270e32)*cosh(.160e-2*t)1.*(.320e32+.497e29*t)*sinh(.160e-2*t)) 6.44 The system transfer function determined in the solution of Problem 6.33 is 1 Rs 2 + C G ( s) = 3 R 1 2 2 1 LRCs s + + s+ s + LC LRC 2 L RC The integro-differential equations governing the free response of the system are t di 1 L 1 + Ri1 + i1 dt + R (i1 i2 ) = 0 dt C0
1 i2 dt + R (i2 i1 ) = 0 C 0
t
(a)
(b) (c)
Assuming an initial condition of the form i (0) = i0 , application of the Laplace transform to Equations (a) and (b) leads to
262
(d)
R 0 L 2R 1 LRs 3 + R 2 + s 2 + s+ 2 C C C RLi0 = L 2R 1 LRs 3 + R 2 + s 2 + s+ 2 C C C i0 = (e) 1 2 2 1 R 3 s + + s+ s + LC LRC 2 L RC The transfer function has either three real poles or one real pole and two poles which are complex conjugates. Consider the former case. Then Equation (e) can be written as i0 I 2 (s) = ( s s1 )( s s 2 )( s s3 ) s 2 s3 s3 s1 s 2 s1 i0 + + ( s1 s 2 )( s 2 s3 )( s3 s1 ) s s1 s s2 s s3 The free response obtained from Equation (f) is i0 i2 (t ) = ( s 2 s3 )e s1t + ( s3 s1 )e s2t + ( s 2 s1 )e s3t ( s1 s 2 )( s 2 s3 )(s3 s1 ) In the case of only one real pole, Equation (e) can be written as i0 I 2 ( s) = ( s s1 )( s 2 + as + b) where if the complex poles are written as s r js j = (f )
(g)
(g)
a = 2s r b=s +s A partial fraction decomposition of Equation (g) leads to 1 i s + a + s1 I 2 ( s) = 2 0 2 s1 + as1 + b s s1 s + as + b Inversion of Equation (j) leads to a + s1 a t a s1t t i0 2 i2 (t ) = 2 e 2 sin( bt ) e e 2 cos( bt ) + s1 + as1 + b b
2 r 2 j
(k)
263
Chapter 6
A MATLAB program which satisfies all requirements is % Program for Problem 6.44 % Input data disp('Please eneter resistance in ohms ') R=input('>>'); disp('Please enter capacitance in farads ') C=input('>>'); disp('Please enter inductance in henrys ') L=input('>>'); disp('Plese enter initial current in amperes ') i0=input('>>'); % Define transfer function N=1/(L*R*C)*[R 0 1/C] D=[1 R/L+1/(R*C) 2/(L*C) 1/(L*R*C^2) 0]; % Determines poles and resiudes of transfer fucntion [r,p,k]=residue(N,D) % Determines free response p=[1 R/L+1/(R*C) 2/(L*C) 1/(L*R*C^2)]; s=roots(p); if imag(s(1))==0 & imag(s(2))==0 % Poles are all real tt=min(abs(s)); dt=tt/500; c=1/(s(1)-s(2))/(s(2)-s(3))/(s(3)-s(1)); for j=1:201 t(j)=(j-1)*dt/200; i2(j)=-i0*c*((s(2)-s(3))*exp(s(1)*t(j))+(s(1)-s(2))*exp(s(2)*t(j))... +(s(3)-s(1))*exp(s(3)*t(j))); end else tt=min(abs(real(s))); dt=tt/500; if real(s(1))<=1.e-5 s1=s(1); a=-2*real(s(2)); b=abs(s(2))^2; else s1=s(3); a=-2*real(s(2)); b=abs(s(2))^2; end for j=1:201 t(j)=(j-1)*dt/200; i2(j)=i0/(s1^2+a*s1+b)*(exp(s1*t(j))-exp(-a/2*t(j))*(cos(b^0.5*t(j))...
264
Chapter 6 +(a/2+s1)/b^0.5*sin(b^0.5*t(j)))); end end plot(t,i2) xlabel('t (s)') ylabel('i_2 (A)') title('Free Response for System of Problem 6.44') Two examples of the programs execution follow. The first corresponds to the former case with three real poles >> Please eneter resistance in ohms >>10000 Please enter capacitance in farads >>1e-6 Please enter inductance in henrys >>1 Plese enter initial current in amperes >>10 N= 1000000 r= 1.0e+004 * -0.0103 1.0202 -1.0100 0.0001 p= 1.0e+003 * -9.8990 -0.1010 -0.1000 0 0 100000000
265
Chapter 6 k= [] >>
The next example corresponds to a case with one real pole and two complex conjugate poles >> Please eneter resistance in ohms >>100 Please enter capacitance in farads >>1e-6 Please enter inductance in henrys >>1 Plese enter initial current in amperes >>10 N= 1.0e+010 * 0.0001 r= 1.0e+002 * 0 1.0000
266
Chapter 6
-1.0001 0.4950 - 0.0243i 0.4950 + 0.0243i 0.0100 p= 1.0e+004 * -1.0000 -0.0050 + 0.0999i -0.0050 - 0.0999i 0 k= [] >>
267
Chapter 6 6.45 The differential equations governing the motion of the system of Figure P6.45 are & & x1 0 & c 0 0 x1 3k 2k 0 x1 F1 (t ) m 0 0 2m 0 & + 0 c c x + 2k 3k k x = F (t ) & & (a) x2 2 2 2 0 0 3m & 0 c c x3 0 & k 3k x3 F3 (t ) x3 & Application of the Laplace transform to Equation (a) assuming all initial conditions are zero leads to ms 2 + cs + 3k X 1 ( s ) F1 ( s ) 2k 0 2 2k cs k X 2 ( s ) = F2 ( s ) 2ms + cs + 3k (b) 2 cs k 0 3ms + cs + 3k X 3 ( s ) F3 ( s ) The matrix of transfer functions is obtained by
2 cs k 2ms + cs + 3k (c) 2 cs k 3ms + cs + 3k A MATLAB program which specifies the values of m,c, and k and determines the matrix of transfer functions symbolically follows % Program for Problem 6.45 m=1 c=2 k=10 syms s t A(1,1)=m*s^2+c*s+3*k; A(1,2)=-2*k; A(1,3)=0; A(2,1)=A(1,2); A(2,2)=2*m*s^2+c*s+3*k; A(2,3)=-c*s-k; A(3,1)=A(1,3); A(3,2)=A(2,3); A(3,3)=3*m*s^2+c*s+3*k; A G=A^-1 G1=vpa(G,3) 2k 0 ms 2 + cs + 3k G ( s) = 2k 0
1
6.46 If the door of Problem 6.29 is subject to a moment the appropriate differential equation is &+ & 25.2& ct + 30 = M (t ) The system has a damping ratio of 1.15. Thus ct = 2(1.15) (25.2)(30)
(a)
= 63.2
Equation (a) becomes
(b)
268
Chapter 6 &+ & 25.2& 63.2 + 30 = M (t ) The systems transfer function is determined from Equation (c) as 1 ( s ) = 2 25.2 s + 63.2 s + 30 The MATLAB commands to determine the impulsive response of the door >> clear >> N=[1]; >> D=[25.2 63.2 30]; >> impulse(N,D) >> The impulsive response is (c)
(d)
6.47 The step response for the perturbations in temperature of the oil and water in the heat exchanger of Example 6.41 are o (t ) = 1.62 + 0.222e 2.63t 1.87e 0.314 t (a)
(b)
Chapter 6 thw(i)=8.14-0.0104*exp(-2.63*t(i))-8.12*exp(-0.314*t(i)); end plot(t,tho,'-',t,thw,'.') xlabel('t (s)') ylabel('T (C)') title('Temperature perturbations in heat exachanger of Problem 6.47') legend('\theta_o','\theta_w') The resulting plot is
270
Chapter 7
7. Frequency Response
7.1 The transfer function for the LR circuit is
1 Ls + R 1 = 0.3s + 100 (a) The sinusoidal transfer function at the given frequency is 1 G (800 j ) = 0.3(800 j ) + 100 1 = 100 + 240 j 100 240 j = (100) 2 + (240) 2 G(s) =
(a )
= 1.48 x10 3 3.55 x10 3 j (b)The steady-state amplitude is The steady-state phase is
I = 120 1.48 x10 3
(b) (c)
= tan 1
Thus the steady-state current is
(d)
(e)
271
Chapter 7 0.1s + 13 V ( s ) 0.1s 0 0.1sV ( s ) = I 2 ( s) = 0.1s + 13 0.1s (0.1s + 13)(0.1s + 18) (0.1s) 2 0.1s 0.1s + 18 0.1sV ( s ) 3.1s + 234 The transfer function is determined as
=
(d)
I 2 (s) 0.1s (e) = V ( s) 3.1s + 234 (a) A frequency of 2.1 kHz is converted into r/s as cycle rad r = 1.32 x10 4 . Thus the appropriate sinusoidal transfer = 2100 2 s cycle s function is 0.1(1.32 x10 4 j ) G (1.32 x10 4 j ) = 3.1(1.32 x10 4 j ) + 234 G ( s) =
1.32 x10 3 j 234 4.09 x10 4 = 4.09 x10 4 j + 234 234 4.09 x10 4 5.40 x10 7 + 3.09 x10 5 j = 1.67 x10 9 = .0323 + 1.85 x10 4 j (b) The steady-state amplitude is
2 2
j j
(f) (g)
1.85 x10 4 = .0057 rad .0323 Thus the steady-state current through the 10 Ohm resistor is i (t ) = 2.58 sin(1.32 x10 4 t + .0057)
= tan 1
(h)
(i)
7.3 A free-body diagram of the junction between the spring and viscous damper at an arbitrary instant is illustrated below
Since the system is assumed to be massless application of Newtons second law to the free-body diagram leads to
272
Chapter 7
F = 0
& kx cx + + F (t ) = 0 & cx + kx = F (t ) (a) Taking the Laplace transform of Equation (a), assuming the initial condition is zero leads to csX ( s) + kX ( s) = F ( s) F ( s) X ( s) = (b) cs + k The systems transfer function is determined using Equation (b) X (s) G(s) = F (s) 1 = (c) cs + k Substitution of given values leads to 1 G(s) = 2000s + 1000 5 x10 4 (d) = s + 0.5 The input is sinusoidal with a frequency of 40 r/s and amplitude of 50 N. The sinusoidal transfer function at this frequency is
(e)
) + (1.25x10 )
2
5 2
(f)
Im(G (40 j ) Re(G (40 j ) (g)
= tan 1
273
Chapter 7 The systems steady-state response is x(t ) = X sin(t + ) = 0.625 sin( 40t 1.558) mm 7.4 The transfer function for the system is
1 (a) 20 s + 300 s + 20000 The system input has an amplitude of 20 N at a frequency of 80 r/s. The sinusoidal transfer function for this frequency is 1 G (80 j ) = 2 20(80 j ) + 300(80 j ) + 20000 G ( s) =
2
(h)
1 = 1.08 x10 5 + 2.4 x10 4 1.08 x10 5 2.4 x10 4 j = 1.22 x1010 = 8.82 x10 6 1.96 x10 6 The steady-state amplitude of the block is
2
j j
(b) (c)
X = (20) ( 8.82 x10 6 ) + (1.96 x10 6 ) 2 = 1.81x10 4 m The steady-state phase is 1.96 x10 6 = tan 1 8.82 x10 6 = 3.36 r The steady-state displacement is x(t ) = 1.81x10 4 sin(80t + 3.36)
(d)
(e)
7.5 The differential equation governing the motion of the system of Figure P7.5 is & & & m&+ cx + kx = cy + ky x The transfer function for the system is cs + k G ( s) = 2 ms + cs + k 2 2 n s + n = 2 2 s + 2 n + n The given values are used to calculate the natural frequency and damping ratio as k n = m
= 2 x10 4 r s N m 20 kg
(a)
(b)
= 31.6
(c)
274
Chapter 7
c 2m n 300
N s m = r 2(20 kg) 31.6 s = 0.237 Substituting Equations (c) and (d) into Equation (b) leads to 15s + 1000 G ( s) = 2 s + 15s + 1000 The input is sinusoidal with a frequency of 80 r/s and amplitude of 0.05 m. The sinusoidal transfer function at this frequency is 15(80 j ) + 1000 G (80 j ) = (80 j ) 2 + 15(80 j ) + 1000 1000 + 1200 j = 6400 + 1200 j + 1000 1000 + 1200 j = 5400 + 1200 j
(d)
(e)
1000 + 1200 j 5400 1200 j = 5400 + 1200 j 5400 1200 j (1000)(5400) + (1200)(1200) + j[(1000)(1200) + (1200)(5400)] = (5400) 2 + (1200) 2 3.96 x10 6 7.68 x10 6 j = 3.06 x10 7 = 0.129 0.251 j
(f)
The amplitude of the steady-state response is X = Y G (80 j ) = 0.05 (-0.129) 2 + (0.251) 2 = 1.41 cm The phase of the steady-state response is (g) Im(G (80 j ) Re(G (80 j )
= tan 1
0.251 = tan 1 0.129 = 2.05 r The steady-state response for the absolute displacement is
(g)
275
Chapter 7
x(t ) = X sin(t + ) = 1.41sin(80t 2.05) cm (h)
7.6 The differential equation governing the motion of the system of Example 2.17 is given by Equation (c) in the solution of the example as I & & x (a) m1 r + 9m 2 r + &+ (c1 r + 9c 2 r )x + (k1 r + 9k 2 r )x = 0 r where x = r . If the disk is subject to a moment Equation (a) can be rewritten using as the dependent variable as &+ (m1r 2 + m2 r 2 + I )& (c1r 2 + 9c2 r 2 )&+ (k1r 2 + 9kr 2 ) = M (t ) (b) Substitution of given values into Equation (b) leads to &+ & 5.90& 1000 + 4000 = M (t ) (c) The transfer function is obtained from Equation (c) as ( s ) G ( s) = M ( s) 1 = (d) 2 5.9 s + 1000s + 4000 The sinusoidal input is given as M (t ) = 20.5 sin( 200t ) (e) The sinusoidal transfer function evaluated at the input frequency is 1 G (200 j ) = 2 5.9(200 j ) + 1000(200 j ) + 4000 1 = 2.32 x10 5 + 2 x10 5 j
2.32 x10 5 2 x10 5 j = 9.38 x1010 The magnitude of the sinusoidal transfer function is 1 G (200 j )] = (2.32 x10 5 ) 2 + (2 x10 5 ) 2 (f)
(g)
276
Chapter 7 7.7 Application of KVL to each loop in the circuit leads to t di 1 (i1 i2 )dt + v(t ) = 0 0.25 1 dt .2 x10 6 0
1 (i1 i2 )dt = 0 .2 x10 6 0 Application of the Laplace transform to Equations (a) and (b) leads to 5 x10 6 5 x10 6 0.25s + I 1 ( s ) V ( s ) s s = 6 6 5 x10 I 2 ( s ) 0 5 x10 20 + s s Cramers rule is used to solve Equation (c) leading to 5 x10 6 V ( s) s 5 x10 6 5 x10 6 V ( s ) 20 + 0 20 + s s = I 1 ( s) = 6 6 1x10 8 5 x10 5 x10 4 s + 1.25 x10 6 + 0.25s + s s s 6 6 5 x10 5 x10 20 + s s 6 20s + 5 x10 V ( s ) = 2 4 s + 1.25 x10 6 s + 1x10 8 The transfer function is I (s) 20s + 5 x10 6 V ( s ) G(s) = 1 = 2 V ( s) 4s + 1.25 x10 6 s + 1x10 8 The appropriate sinusoidal transfer function is 20(300 j ) + 5 x10 6 G (300 j ) = 4(300 j ) 2 + 1.25 x10 6 (300 j ) + 1x10 8 20i2 +
t
(a) (b)
(c)
(d)
(e)
9.96 x10 7 3.75 x10 8 5 x10 6 + 6000 j = 9.96 x10 7 + 3.75 x10 8 j 9.96 x10 7 3.75 x10 8 14 15 5.00 x10 1.87 x10 j = 1.51x1017 = 3.32 x10 3 0.0125 j The steady-state amplitude is I = 200 (3.32 x10 3 ) + (0.0125) 2 = 2.58
2
j j
(f) (g)
(h)
= tan 1
The steady-state response is
277
Chapter 7
i (t ) = 2.58 sin(300t + 1.83) A (i)
7.8 The mathematical model for the perturbations in liquid levels in the tanks is dh 1 1 A1 1 + h1 h2 = 0 dt R1 R1
p p (t ) 1 dh2 1 1 h1 + + R R h2 = gR dt R1 2 2 1 The resistances are obtained by considering the steady state 2h R = 1 q1 A2 2(8 m - 6 m ) m3 0.8 s s =5 2 m =
(a) (b)
(c)
R2 =
2h2 q2
N 2.5 x10 3 2 6 m m 2 kg m 1000 3 9.81 2 m s = 3 m 0.8 s s = 14.36 2 m Substitution of all values into Equations (a) and (b) leads to dh 10.2 1 + 0.2h1 0.2h2 = 0 dt dh 14.1 2 0.2h1 + 0.270h2 = 4.26 x10 3 sin(50t ) dt Dividing Equations (e) and (f) by the areas leads to dh1 + 0.0196h1 0.0196h2 = 0 dt dh2 0.0142h1 + 0.0191h2 = 3.02 x10 4 sin(50t ) dt 4 Defining F (t ) = 3.20 x10 sin(50t ) , taking the Laplace transforms of Equations (g) (h) assuming all initial conditions are zero leads to s + 0.0196 0.0196 H 1 ( s ) 0 0.0124 s + 0.0191 H ( s) = F ( s) 2
(d)
(e) (f)
(g) (h)
and
(i)
278
Chapter 7 The solutions to Equation (i) are obtained using Cramers rule as 0.0196 0 F ( s ) s + 0.0191 H 1 (s) = s + 0.0196 0.0196 0.0124 s + 0.0191
0.0196 F ( s ) ( s + 0.0196)( s + 0.0191) (0.0196)(0.0124) 0.0196 F ( s ) = 2 s + 0.0387 s + 6.17 x10 4 s + 0.0196 0 0.0124 F ( s ) H 2 (s) = s + 0.0196 0.0196 0.0124 s + 0.0191
=
2
(j)
( s + 0.0196) F ( s ) s + 00387 s + 6.17 x10 4 The systems transfer functions are determined from Equations (j) and (k) as H (s) G1 ( s ) = 1 F (s) 0.0196 = 2 s + 0.0387 s + 6.17 x10 4 H (s) G2 ( s) = 2 F ( s) s + 0.0196 = 2 s + 0.0387 s + 6.17 x10 4 The sinusoidal transfer functions corresponding to an input frequency of 50 r/s are 0.0196 G1 (50 j ) = 2 (50 j ) + 0.0387(50 j ) + 6.17 x10 4
(k)
(l)
(m)
2500 1.935 j 0.0196 = 2500 + 1.935 j 2500 1.935 j 0.0195(2500 1.935 j ) = 6.25 x10 6 50 j + 0.0196 2500 1.935 j G2 (50 j ) = 2500 + 1.935 j 2500 1.935 j =
(n)
47.75 1.25 x10 5 j (o) 6.25 x10 6 The steady-state amplitudes of the perturbations are obtained using the sinusoidal transfer functions. The amplitude of the perturbation in the first tank is
279
Chapter 7
H 1 = 3.02 x10 4 G1 (50 j ) = 3.02 x10
)(0.0196)
= 6.04 x10 6 m
7.9 The differential equation governing the motion of the system of Figure P7.9 is & & & m&+ cx + kx = cy + ky x The transfer function for the system is cs + k G ( s) = 2 ms + cs + k 2 2 n s + n = 2 2 s + 2 n + n The given values are used to calculate the natural frequency and damping ratio as k n = m = 1x10 5 N m 80 kg
(a)
(b)
r s c = 2m n = 35.4
300
(c)
N s m = r 2(80 kg) 35.4 s = 0.0530 (d) Substituting Equations (c) and (d) into Equation (b) leads to 3.75s + 1250 G ( s) = 2 (e) s + 3.75s + 1250 Equation (e) is the transfer function for the displacement of the block. The acceleration of the block is & A=& x (f) Taking the transform of Equation (f) leads to
280
Chapter 7
A( s) = s 2 X ( s ) = s 2 G ( s)Y ( s) (g) Equation (g) is rewritten to define the transfer function for the acceleration of the block as A( s ) Ga ( s ) = Y ( s)
= s 2 G ( s) = s 2 (3.75s + 1250) (h) s 2 + 3.75s + 1250 The input is sinusoidal with a frequency of 45 r/s and amplitude of 0.022 m. The sinusoidal transfer function at this frequency is ( 45 j ) 2 [3.75( 45 j ) + 1250] G a ( 45 j ) = (45 j ) 2 + 3.75( 45 j ) + 1250 = = = 2.534 x10 6 3.42 x10 5 j 2025 + 168.75 j + 1250 2.534 x10 6 3.42 x10 5 j 775 + 168.75 j j 775 168.75 j 775 168.75 j j (i )
2.534 x10 6 3.42 x10 5 = 775 + 168.75 j 2.09 x10 9 6.92 x10 8 = 6.29 x10 5 = 3.21x10 3 1.10 x10 3 j
) + ( 1.10 x10 j )
2 3
(j)
Im(G (80 j ) Re(G (80 j )
= tan 1
(k)
281
7.10 Free-body diagrams of the blocks at an arbitrary instant are illustrated below
Application of Newtons Second Law to each of the free-body diagrams leads to & x1 F = m1 &
& 6 x10 5 x1 + 8 x10 5 ( x 2 x1 ) = 4 & x1 & 4 & + 1.4 x10 6 x1 8 x10 5 x 2 = 0 x1 & x F = m2 &2 & 8 x10 5 ( x 2 x1 ) + F (t ) = 6 & x2 & 6 & 8 x10 5 x1 + 8 x10 5 x 2 = F (t ) x2 Taking the Laplace transforms of both sides of Equations (a) and (b) leads to 4s 2 X 1 ( s ) + 1.4 x10 6 X 1 ( s) 8 x10 5 X 2 ( s) = 0 (b) (a)
6s 2 X 2 ( s) 8 x10 5 X 1 ( s ) + 8 x10 5 X 2 ( s) = F ( s) Equations (c) and (d) are summarized in matrix form as 4s 2 + 1.4 x10 6 8 x10 5 X 1 ( s) 0 = 5 6s 2 + 8 x10 5 X 2 ( s) F ( s ) 8 x10 Application of Cramers rule to Equation (e) leads to 8 x10 5 0 F ( s) 6 s 2 + 8 x10 5 X 1 ( s) = 8 x10 5 4s 2 + 1.4 x10 6 8 x10 5 6s 2 + 8 x10 5
= 8x10 5 F ( s ) (4s 2 + 1.4 x10 6 )(6s 2 + 8x105 ) (8x10 5 ) 2
8x10 5 F ( s ) = 10s 4 + 1.16 x10 7 s 2 + 4.80 x1011 8x10 4 F ( s ) = 4 s + 1.16 x10 6 s 2 + 4.80 x1010 The transfer function is obtained from Equation (f) as
(f)
282
Chapter 7 G1 ( s ) = = X 1 (s) F ( s)
8x10 4 (g) s 4 + 1.16 x10 6 s 2 + 4.80 x1010 The input is sinusoidal at a frequency of 150 r/s and an amplitude of 200 N. The sinusoidal transfer function at this frequency is 8x10 4 G (150 j ) = (150 j ) 4 + 1.16 x10 6 (150 j ) 2 + 4.80 x1010
8x10 4 2.12 x1011 = 3.76 x10 7 The sinusoidal transfer function is real, but is negative, thus (G (150 j ) = 3.76 x10 7 = X = 200(3.76 x10 7 ) = 7.42 x10 5 m = The blocks steady-state response is x(t ) = 7.42 x10 5 sin(150t + ) 7.11 The transfer function for a series LRC circuit with = 1.2 and n = 500 form
s s + 1200 s + 2.5 x10 4 The Bode diagram is obtained using the following MATLAB work session G(s) =
2
>> N=[1 0] N= 1 0
283
Chapter 7
The function L( ) the series LRC circuit is given in Equation (7.39). Substitution of given parameters leads to L( ) = 20 log 2.5 x10 4 2 2 + [2(1.2)(500) ]2
= 20 log 2.5 x10 4 2 2 + [1200 ]2 Properties of logarithms are used to rewrite Equation (b) as
(b)
L( ) = 20 log( ) 10 log (2.5 x10 4 2 ) + [1200 ] (c) The low-frequency asymptote is obtained by taking the limit of Equation (c) as 0 , leading to lim L( ) = 20 log( ) 20 log(2.5 x10 4 )
2
= 20 log( ) 87.96 (d) The high-frequency asymptote is obtained by taking the limit of Equation (c) as leading to lim L( ) = 20 log( ) 40 log( )
= 20 log( )
(e)
284
(a)
285
Chapter 7 The sinusoidal transfer function is obtained from Equation (a) as 0 .05 G ( j ) = 2 ( j ) + 15 ( j ) + 1000
0 .05 1000 2 + 15 j
1000 2 15 j 0 .05 = 1000 2 + 15 j 1000 2 15 j 1 2 0 .75 j 5 20 = (b) (1000 2 ) 2 + (15 ) 2 The real and imaginary parts of the sinusoidal transfer function are obtained from Equation (b) as 1 2 5 20 (c) Re [G ( j ) ] = (1000 2 ) 2 + (15 ) 2
Im[G(j )] =
(d)
(e)
(f)
= 26.02 10 log (1000 2 ) 2 + (15 ) 2 (g) The low-frequency asymptote is obtained by taking the limit of Equation (g) as 0 , leading to lim L( ) = 26.02 20 log(1000)
0
= 86.02 dB
(h)
286
Chapter 7 The high-frequency asymptote is obtained by taking the limit of Equation (g) as leading to lim L( ) = 26.02 40 log( ) (i)
7.13 The transfer function for an undamped mass-spring system of natural frequency r 100 is s 1 (a) G ( s) = 2 s + 1x10 4 The MATLAB work session to draw the Bode diagram is >> clear >> N=[1] N= 1 >> D=[1 0 10000] D= 1 0 10000 >> bode(N,D) >> The resulting Bode diagram is
287
Chapter 7 7.14 The differential equation of Problem 6.30 is & & & m&+ cx + kx = cy + ky x (a) Dividing Equation (a) by m leads to &+ 2 n x + n x = 2 n y + n y & & 2 & 2 x (b) where the natural frequency and damping ratio are determined using the parameter values of Problem 6.30 k n = m = 3.x10 5 r s N m 500 kg (c)
= 25.3
c 2m n 10000
(d)
(e)
Substitution into Equation (b) leads to &+ 20 x + 640 x = 20 y + 640 y & & & x The system transfer function is obtained from Equation (e) as X (s) G ( s) = Y ( s) 20s + 640 = 2 s + 20s + 640 As the vehicle traverses the road it is subject to a sinusoidal input of the form y (t ) = Y sin(t ) where Y=0.002 m and 2v = 1.2 = 5.24v where v is the horizontal speed of the vehicle. The Bode diagram is determined for the transfer function of Equation (f) using MATLAB. The MATLAB work session to generate the Bode diagram is >> N=[20 640] >> D=[1 20 640] >> bode(N,D) 288
(f)
(g)
(h)
s+3 s + 10 s + 50
2
(a)
G ( j ) =
( j ) + 3 ( j ) + 10( j ) + 50 3 + j = 50 2 + 10 j
2
+ 100 2 The real and imaginary parts of the sinusoidal transfer function are
150 + 7 + j 20 2
2
(50 )
]
(b)
2 2
289
Chapter 7
150 + 7 2
2 2 2
Re[G ( j )] =
(c) (d)
G ( j ) =
2 2
(e)
(150 + 7 ) + (20 ) = 20 log [(50 ) + 100 ] (150 - 7 ) + (20 ) = 10log (50 ) + 100 = 10log[(150 - 7 ) + (20 ) ] 20 log[(50 )
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
2 2
+ 100 2
(f)
(g)
( )
( )
= 60 log 80 log = 20 log (h) (b) The Nyquist diagram is a plot of Re[G(j0] on the horizontal scale with Im[G(j)] on the vertical scale. Repeating Equation (c) and (d) 150 + 7 2 Re[G ( j )] = (c) 2 50 2 + 100 2
Im[G(j )] =
(50 )
(20 2 )
2 2
+ 100 2
(d)
290
Chapter 7 (ii) The Nyquist diagram intercepts the horizontal axis when Im[G(j)]=0. From Equation (d) the intercepts occur for = 0, 20 . The value of the intercept for =0 is 0.06, The intercept values for = 20 are obtained from Equation (c) as 150 + 7(20) Re G ( 20 j ) = (50 20) 2 + 100(20) = 0.01 (i) The Nyquist diagram intercepts the vertical axis when Re[G(j)]=0. From Equation (c) the only intercept occurs for =0 for which Im[G (0)] = 0 (iii) Since Re[G ( j )] is always positive the Nyquist diagram exists only in the first and fourth quadrants.
(a) The MATLAB work session to develop the Bode and Nyquist diagrams is presented below >> clear >> N=[1 3] N= 1 3
>> D=[1 10 50] D= 1 10 50 >> bode(N,D) >> figure >> Nyquist(N,D) >> The resulting Bode diagram is
291
Chapter 7
292
Chapter 7 7.16 Consider the transfer function s 2 + 2s + 4 (a) s 3 + 5s 2 + 6 s + 10 The sinusoidal transfer function corresponding to the transfer function of Equation (a) is ( j ) 2 + 2( j ) + 4 G( j ) = ( j ) 3 + 5( j ) 2 + 6( j ) + 10 G(s) = = 4 2 + 2j 10 5 2 + j (6 3 )
(b) (10 5 ) + (6 ) The real and imaginary parts of the sinusoidal transfer function are obtained from Equation (b) as 40 18 2 + 18 4 Re[G ( j )] = (c) 2 (10 5 2 ) 2 + 2 (6 2 )
2 2 2 2 2
40 18 2 + 18 4 j 4 + 4
Im[G ( j )] =
4 +4
(d)
G ( j ) = =
(40 18
(10 5 2 ) 2 + 2 (6 2 )
+ 18 4 ) + ( 4 + 4
2 2
])
(e)
(40 18
2
+ 18 4
2 2 4 2
2 2
2 2
) +2 6 2
) ] (f)
2
(g)
293
Chapter 7
lim L( ) = 10 log 9 20 log 6 = 90 log 120 log = 30 log (h) (b) The Nyquist diagram is drawn from the real and imaginary parts of the sinusoidal transfer function. Equations (c) and (d) are repeated below 40 18 2 + 18 4 Re[G ( j )] = (c) 2 (10 5 2 ) 2 + 2 6 2
( )
( )
(d) 2 (10 5 ) + 6 2 For = 0 , Equation (c) shows that =0.4 and Equation (d) shows that (i) Im[G ( j )] =0. As , Equations (c) and (d) show that Re[G ( j )] 0 and Im[G ( j )] 0 . (ii) Equation (d) shows that the only horizontal intercept corresponds to =0 while Equation shows that a vertical intercept may occur for any real such that 40 18 2 + 18 4 =0. The quadratic equation is applied to show that the values of which satisfy this equation are complex. Thus there are no Imintercepts. (iii) Since there are no y-intercepts the Nyquist diagram does not exist in the second and third quadrants. (c) The MATLAB work session necessary to draw the Bode and Nyquist diagrams is shown below >> clear >> N=[1 2 4]
2 2 2
Im[G ( j )] =
4 +
N= 1 2 4
294
Chapter 7
295
(a) s( s + 6)( s + 3)( s + 2) The amplitude part of the Bode diagram is unaffected by the time delay. The following MATLAB commands leads to the correct amplitude part of the Bode diagram >> N=[2 3]
N= 2 3
G ( s) =
(2s + 3)e 3s
>> D=[1 11 36 36 0] D= 1 11 36 36 0
296
Chapter 7 The correct phase part of the Bode diagram can be determined by a superposition 3 1 1 1 G ( s ) = 2 s + e 3 s (b) 2 s + 2 s + 3 s + 6 The phase of a constant is 0, the phase of a first-order lead is given by Equation (7.51); the phase of a first-order lag is given by Equation (7.52); the phase of a time delay is given by Equation (7.55). Thus 2 1 1 1 ( ) = tan 1 tan 1 tan 1 tan 1 3 (c) 3 2 3 6 The MATLAB M file used to generate the frequency part of the Bode diagram is given below % Program to generate phase portion of Bode iagram for Problem 7.17 % Defining the range of frequencies clear omega=logspace(-2,2,200); phi=atan(2/3.*omega)-atan(1/2.*omega)-atan(1/3.*omega)-atan(1/6.*omega)3.*omega semilogx(omega,phi) xlabel(\omega) ylabel(\phi) title(Phase part of Bode diagram for Problem 7.17) The resulting phase portion of the Bode diagram is
297
Chapter 7
7.18 The transfer function for the system is s 2 + 3s + 4 G ( s) = s ( s 2 + 5s + 12) The MATLAB commands to draw the Bode diagram are >> N=[1 3 4]; >> D=[1 5 12 0]; >> bode(N,D) >> The resulting Bode diagram is (a)
G ( s) =
The sinusoidal transfer function is
(2s + 3)e 3s
s( s + 6)( s + 3)( s + 2)
(a)
298
Chapter 7
G ( j ) =
[2( j ) + 3]e 3 j
( 36 ) + (36 11 ) [3 cos(3 ) + 2 sin(3 )]( 36 ) + [2 cos(3 ) + 3 sin(3 )][36 11 ] = ( 36 ) + (36 11 ) [2 cos(3 ) + 3 sin(3 )][ 36 ] [3 cos(3 ) + 2 sin(3 )][36 11 ] + j ( 36 ) + (36 11 )
4 2 2 2 3 2 4 3 4 2 2 2 3 2 4 3 4 2 2 3 2
( j ) 4 + 11( j ) 3 + 36( j ) 2 + 36( j ) (3 + j 2 )[cos(3 ) j sin(3 )] = 4 36 2 + j 36 11 3 3 cos(3 ) + 2 sin(3 ) + j[2 cos(3 ) + 3 sin(3 )] = 4 36 2 + j 36 11 3
j 36 11 3 }
(b)
The real part of the sinusoidal transfer function is obtained from Equation (b) as [3 cos(3 ) + 2 sin(3 )] 4 36 2 + [2 cos(3 ) + 3 sin(3 )] 36 11 3 Re[G ( j )] = 2 4 36 2 + (36 11 3 ) 2
(c)
The imaginary part of the sinusoidal transfer is obtained from Equation (b) as
Im[G ( j )] =
36 2
+ (36 11 3 ) 2
(d)
(
2
36 2
+ (36 11 3 ) 2
= lim
0
2 2 ( 2 36) + (36 11 2 ) 2
(e)
299
Chapter 7
Im[0] = lim
0
= lim
0
( 36) + (36 11 2 ) 2
2 2 2
36 2
+ (36 11 3 ) 2
]
(f)
=0 Thus the point corresponding to = 0 is(-1/36,0). Also 2 5 sin(3) 22 4 cos(3 ) lim Re[G( j )] = lim 2 2 2 ( 2 36) + (36 11 2 ) 2 =0
]
(g)
2 2 ( 2 36) + (36 11 2 ) 2
2
2 5 cos(3) + 22 4 sin(3 )
=0 (h) The real axis intercepts correspond to values of such that Im[G ( j )] = 0 which from Equation (d) occurs for values of such that [2 cos(3) + 3sin(3)] 4 36 2 [3 cos(3) + 2 sin(3)] 36 11 3 = 0
2 5 39 3 108 (i) 19 4 + 180 2 The imaginary axis intercepts correspond to the to values of such that Re[G ( j )] = 0 which from Equation (c) occurs for values of such that [3 cos(3 ) + 2 sin(3 )] 4 36 2 + [2 cos(3 ) + 3 sin(3 )] 36 11 3 = 0 (j) tan(3 ) =
(a)
300
G ( j ) =
= =
4 2 + j (3 ) ( j )[12 2 + j (5 )]
4 2 + j (3 ) 5 2 + j (12 2 )
]
(b)
(c)
4 15 3 + 16 2 48 (d) Im[G ( j )] = 2 4 + 2 + 144 The Nyquist diagram is a polar plot of Equations (c) and (d) with as a parameter. The following are determined from Equations (c) and (d) (i) For =0, Re(0)=1/9, Im(0)=- . Thus the Nyquist diagram has an asymptote at Re=1/9. (ii) As , Re[G ( j )] 0 and Im[G ( j )] 0 (iii) As , Re[G ( j )] 0 and Im[G ( j )] 0 (iv) There are no real values of such that Re[G ( j )] =0. Thus the only imaginary axis intercept is the origin. 3 (v) Im[G ( j )] = 0 when 4 15 3 + 16 2 48 = 0 which has no real roots. Thus the only real axis intercept is at the origin. (vi) Since the only intercepts are the origin and Re(0)=1/9, the Nyquist diagram exists only in the first and fourth quadrants.
301
Chapter 7
the coefficient of s k in the numerator of G(s). (b) The MATLAB workspace used to create the Bode diagram for the transfer function is >> clear >> N=[1 0 2 6] N= 1 0 2 6
>> D=[1 10 20 100] D= 1 10 20 100 >> bode(N,D) >> The Bode diagram for this transfer function is 302
Chapter 7
While its the transfer function satisfies necessary conditions, the circuit is not an effective band-reject filter. The numerator must be modified such that the low frequency asymptote is approximately the same as the high frequency asymptote. 7.22 (a) The natural frequency of the undamped system is k n = m = 5 x10 6 r s N m 500 kg (a)
n
110
(b)
Chapter 7 The amplitude of the steady-state response of an undamped second-order mechanical system is F 1 X = 02 m n 1 r 2 1 2 r 1 (1.1) (500 kg )100 s 4 = 1.43x10 m (c) Since the input frequency is greater than the natural frequency the phase angle is = r . Thus the steady-state response is x(t ) = 1.43x10 4 sin(100t + ) m (d) (a) When the absorber is tuned to the operating frequency the steady-state amplitude of the absorber is F (e) X2 = 0 k2 In order to limit the steady-state amplitude of the absorber to 1 cm 150 N 0.01 m > k2 = 150 N
2
k2 >
N m = 2 r 100 s = 1.5 kg (g) (b) If a 10-kg machine is removed from the machine its mass is 490 kg. The resulting steady-state amplitude of the machine is calculated using Equation (7.92) F0 (k 2 m2 2 ) (f) X1 = m1 m2 4 [(k1 + k 2 )m2 + k 2 m1 ] 2 + k1 k 2 However, when mass is removed from the machine the absorber is still tuned to the input frequency, k 2 = m2 2 . Thus the steady-state amplitude of the machine is still zero. 1.5 x10 4
304
Chapter 7 7.23 The differential equation governing the motion of the machine is & & m&+ cx + kx = m0 e 2 sin(t ) x Equation (a) is of the form of a second-order system with F (t ) = m0 e 2 sin(t ) The transfer function for the system of Equation (a) is X (s) G(s) = F (s) 1 = 2 ms + cs + k 1/ m = 2 2 s + 2 n s + n The sinusoidal transfer function is 1/ m G ( j ) = 2 n 2 + 2 n j
2 1 n 2 2 n j 2 m n 2 2 + (2 n )2 The steady-state response of the system is x(t ) = X sin(t + ) where X = F0 G ( j )
(a) (b)
(c)
(d)
(e)
2 2 2 1 ( n ) + (2 n ) = (m0 e ) 2 m ( n 2 )2 + (2 n ) 2 2 2
m0 e 2 m
2 n
2 2
+ (2 n ) 1
(f)
2
2 n
2 2 r 2 n ) + (2r n )
m0 e r2 m (1 r 2 ) 2 + (2r ) 2
(g)
305
= tan 1
2 n = tan 1 2 2 n
1
2 2r n 2 = tan 2 2 n r n 2r = tan 1 2 1 r
306
Chapter 8
(d)
(e) (f)
8.2 The differential equations for the perturbations of the concentrations of the reactants in a CSTR due to the perturbation in inlet flow rate are dC 0.2 A + 1.2C A 0.4C B = 0.2q (a) dt dC 0.2 B 0.4C A + 0.9C B = 0.1q (b) dt Application of the Laplace transform to Equations (a) and (b) leads to 0.2sC A ( s ) + 1.2C A ( s) 0.4C B ( s) = 0.2Q( s) (c)
0.2 sC B ( s) 0.4C A ( s ) + 0.9C B ( s) = 0.1Q( s ) Equations (c) and (d) are rearranged as 0.2Q( s ) + 0.4C B ( s ) C A ( s) = 0.2s + 1.2 0.1Q( s ) + 0.4C A ( s ) C B ( s) = 0.2s + 0.9 The block diagram model for the system is developed from Equations (e) and (f)
(d)
(e) (f)
307
Chapter 8
8.3 The differential Equations for the temperature of the oil and water in the heat exchanger of Example 6.41 are given in Equations (e) and (f) of the solution of the Example as d 5.73 x10 3 o + 1.48 x10 4 o 2.99 x10 3 w = 0 (a) dt d 4.04x10 4 w 2.99 x10 3 o + 1.35 x10 4 w = F (t ) (b) dt where the right-hand side of Equation (b) has been replaced by an arbitrary F(t). Application of the Laplace transform to Equations (a) and (b) and rearrangement leads to 2.99 x10 3 w ( s ) (c) o (s) = 5.73x10 3 s + 1.48 x10 4 F ( s ) + 2.99 x10 3 o ( s ) (d) w ( s) = 4.04 x10 4 s + 1.35 x10 4 The block diagram model for the system is developed from Equations (c) and (d)
8.4 The differential equations which model the concentrations of the drug using the twocompartment model are given by Equations (g) and (h) of the solution of Example 4.30 as dC P Vp + (k e + k1 )V p C p k 2Vt C t = I (t ) (a) dt dC t Vt k1V p C p + k 2Vt C t = 0 (b) dt Application of the Laplace transform to Equations (a) and (b) and rearrangement leads to
308
Chapter 8
I ( s ) + k 2Vt C t ( s ) V p s + (k e + k1 )V p k1V p C p ( s )
C P ( s) = Ct ( s) =
(c) (d)
Vt s + k 2Vt The block diagram model for this system is developed using Equations (c) and (d)
8.5 The feedback loop involving G2 ( s) is in series with G1 leading to the reduction of Figure (a).
309
Chapter 8
Elimination of the unity feedback loop leads to the reduction of Figure (b) with G1G2 1 + G2 H 1 ( s) = GG 1+ 1 2 1 + G2
= G1G2 1 + G2 + G1G2 (a)
Noting that H 1 is in series with G3 and elimination of the remaining feedback loop leads to Figure (c) with H 1G3 H ( s) = 1 + G 4 G3 H 1 G1G2 1 + G2 + G1G 2 = G1G2 1 + G 4 G3 1 + G2 + G1G2 G3 = G1G2 G3 1 + G2 + G1G 2 + G1G2 G3G4 (b)
8.6 The transfer functions G 2 and G 3 are in parallel, but both have negative signs. Their equivalent is in series with G1 as illustrated in Figure (a). The feedback loop is eliminated leading to a transfer function of G1 (G2 + G3 ) (a) H 1 ( s) = 1 G1G5 (G2 + G3 ) The resulting reduction has a second feedback loop which is eliminated resulting in a system transfer function of H1 H ( s) = 1 G4 H 1
G1 (G2 + G3 ) 1 G1G5 (G2 + G3 ) = G1 (G2 + G3 ) 1 G4 1 G1G5 (G2 + G3 ) = G1 (G2 + G3 ) 1 G1G5 (G2 + G3 ) + G4 G1 (G2 + G3 ) (b)
310
Chapter 8
311
Chapter 8 Equation (b) is solved for B(s) as ( s + 4)( s 2 + 16) + 2( s + 2) B ( s ) = ( s + 2)[A( s ) C ( s )] ( s + 2)[ A( s ) C ( s )] B( s) = 3 s + 4 s 2 + 18s + 68 Substitution of Equation (c) into Equation (a) leads to ( s + 2)[ A( s ) C ( s )] 2 C (s) = 3 + C ( s) s + 4s 2 + 18s + 68 s s s 3 + 4 s 2 + 18s + 68 C ( s ) = s ( s + 2)[ A( s ) C ( s )] + 2 s 3 + 4 s 2 + 18s + 68 C ( s )
(c)
[(s 2)(s
+ 4 s + 18s + 68 + s ( s + 2) C ( s) = s( s + 2) A( s )
2 3
s ( s + 2) A( s ) s + 2 s + 11s 2 + 34s 136 The closed-loop transfer function is obtained from Equation (d) as C (s) H ( s) = A( s ) s( s + 2) = 4 3 s + 2s + 11s 2 + 34s 136 C (s) =
4
(d)
(e)
that C ( s) 0.2 A( s) s + 2 C ( s) = 2 s + 6s + 1 C ( s) ( s 2 + 6s + 1)C ( s) = 0.2 A( s) s + 2 1 5( s 2 + 6s + 1)C ( s) + C ( s) = A( s ) s+2 5( s + 2)( s 2 + 6s + 1) + 1 C ( s ) = A( s ) s+2 ( s + 2) A( s ) C ( s) = 3 5s + 40s 2 + 65s + 11 Thus the closed-loop transfer function is
(a)
312
Chapter 8 H ( s) = = C (s) A( s )
3
(b)
it is clear that
0.2[ A( s ) 0.3C ( s )] s ( s + 4) C (s) = s+2 s ( s + 4) A( s) + 0.2[A( s) 0.3C ( s)] ( s + 2)C ( s ) = s( s + 4) s ( s + 2)( s + 4)C ( s ) = [s( s + 4) + 0.2]A( s ) 0.6C ( s ) [s( s + 2)(s + 4) + 0.6]C ( s) = [ s( s + 4) + 0.2] A( s) A( s ) + C (s) = (a)
s 2 + 4 s + 0.2 A( s ) s 3 + 6s 2 + 8s + 0.6 The closed-loop transfer function is obtained from Equation (a) as C (s) H (s) = A( s ) = 8.10 From the annotated block diagram s 2 + 4 s + 0.2 s 3 + 6s 2 + 8s + 0.6
(b)
it is clear that
313
Chapter 8 B( s) s +1 2 A( s) B( s) 3C ( s) B( s) = s+5 s + 3 s + 2 C (s) = Equation (b) is solved for B(s) as 3 ( s + 2)[A( s ) B( s )] 3( s + 5)C ( s ) B( s) = ( s + 2)( s + 5) s+2 [3( s + 5) + ( s + 2)]B( s) = ( s + 2) A( s) 3( s + 5)C ( s) ( s + 2) A( s) 3( s + 5)C ( s ) B( s) = 4s + 17 Substitution of Equation (c) in Equation (a) leads to ( s + 2) A( s) 3( s + 5)C ( s ) C (s) = (4s + 17 )( s + 1) (4s + 17)( s + 1)C ( s) = ( s + 2) A( s ) 3( s + 5)C ( s) [(4s + 17)(s + 1) + 3( s + 5)]C ( s) = ( s + 2) A( s) s+2 C (s) = 2 A( s) 4s + 24s + 32 The closed-loop transfer function is obtained from Equation (d) as H ( s) = = C (s) A( s ) s+2 4s + 24s + 32
2
(a) (b)
(c)
(d)
(e)
8.11 The differential equation for the system of Example 4.15 is d 1.68 x10 5 (a) + 2.095 x10 3 = 500u (t ) dt Equation (a) is rewritten in the standard form of the differential equation for a first-order system as d 80.2 + = 0.2387u (t ) (b) dt The time constant for the system is T=80.2 s. The settling time for a first-order system is 2.93T. Thus in order to reduce the settling time by 30% the time constant must be reduced by 30%. The required time constant when placed in a feedback control system is T = 0.7T = (0.7)(80.2)
= 56.1 s (c) The transfer function for the system is obtained by taking the Laplace transform of Equation (a)
314
Chapter 8
(d)
The time constant of the system when placed in a feedback loop with a proportional controller is given by Equation (8.38) T (e) T= 1 + K p KT Using Equation (c) in Equation (e) leads to 80.2 56.1 = 1 + K p (0.003)(80.2)
K p = 1.80
(f)
The closed loop transfer function when this proportional controller is used is given by Equation (8.37) which when applied to this problem leads to (1.80)(0.003) H (s) = s + 0.0125 + (1.80)(0.003) 0.0054 = (g) s + 0.0179 Application of the Final Value Theorem to Equation (g) leads to a final value of 0.0054 Tf = 0.0179 = 0.302 (h)
8.12 The differential equation for the system of Example 4.15 is d 1.68 x10 5 + 2.095 x10 3 = 500u (t ) (a) dt Equation (a) is rewritten in the standard form of the differential equation for a first-order system as d 80.2 + = 0.2387u (t ) (b) dt The transfer function for the system is obtained by taking the Laplace transform of Equation (a)
315
Chapter 8 80.2s + = 0.2387 F ( s) 0.2387 ( s ) = 80.2s + 1 0.003 F (s) = s + 0.0125 0.003 G ( s) = (c) s + 0.0125 The damping ratio of the second-order system that results when a first-order plant is used in a feedback control loop with an integral controller is given by Equation (8.46) 1 = (d) 2T K i K Requiring the damping ratio of the closed-loop system to be 2.5 leads to 1 2.5 = 2(80.2) K i (0.003) K i = 2.07 x10 3 8.13 The transfer function of a first-order plant is
0 .2 (a) s+5 When used in a feedback loop with a PI controller the closed loop transfer function is ( K i + sK p )G ( s ) H (s) = (b) s + (K i + sK p )G ( s ) G ( s) =
(e)
(a) Using K i = 5.2, K p = 10 and Equation (a) in Equation (b) leads to 0.2 s+5 H (s) = 0.2 s + (5.2 + 10s ) s+5 1.04 + 2s = s( s + 5) + 1.04 + 2s 1.04 + 2s = 2 (c) s + 7 s + 1.04 (b) Equation (b) is the transfer function of a second-order system of natural frequency n = 1.04 (5.2 + 10s )
= 1.020
and damping ratio
(d)
316
Chapter 8
=
=
7 2 n
7 2(1.020) (e) = 3.43 (c) Since the damping ratio is greater than one, the system is undamped. Its impulsive response is obtained using Table 6.5 where
s1 = 1.020 3.43 + (3.43) 2 1 = 6.85 s2 (3.43) 2 Use of superposition leads to 0.152t 1 xi (t ) = 1.04 e 6.85t ) 0.152 (6.85) (e
( = 1.020(3.4 +
(
) 1 ) = 0.152
)
(f) (g)
1 6.85 t + 6.85e 0.152t + 2 0.152 (6.85) 0.152e 0.152 t 6.85 t = 3.88e 0.354e (h) (d) The step response is obtained using Table 6.6. Use of superposition leads to 1.04 1 6.85t x s (t ) = (6.85)e 0.152t (0.152)(6.85) 0.152 (6.85) 6.85 (0.152) + (0.152)e
)
(i)
0.833 (a) s + 0.134 (a) When used in a feedback loop with a proportional controller with a gain of 0.2 the closed-loop transfer function becomes
317
Chapter 8
0.833 s + 0.134 H ( s) = 0.833 1 + 0 .2 s + 0.134 0.167 = s + 0.301 0 .2
(b)
1 0.301 = 3.321 s The systems impulsive response is determined from Equation (6.41) as xi (t ) = (0.833)(0.301)e 0.301t T=
(c)
0.250e 0.301t (d) The offset when subject to a unit step input is given by Equation (8.42) as T = T 3.321 = 7.44 (e) = 0.446 (b) The closed-loop transfer function when the plant is in a feedback loop with an integral controller of gain 1.5 is 0.833 1 .5 s + 0.134 H ( s) = 0.833 s + 1 .5 s + 0.134 1.25 = 2 (f) s + 0.134 s + 1.25 Equation (f) is that of a second-order system with a natural frequency of n = 1.25 = 1.118 r/s
and a damping ratio of
0.134 2(1.118) = 0.0601 Since the system is underdamped its damped natural frequency is
(g)
(h)
d = 1.118 1 0.(0601) 2
= 1.116 r/s impulsive response is determined using Table 6.5 as (i)
318
Chapter 8 1.25 ( 0.0601)(1.118)t e sin(1.116t ) 1.116 = 1.120e 0.0674t sin(1.116t ) (j) Since an integral controller is used the offset is zero. (c) The transfer function when the plant is placed in a feedback loop with a PI controller with K p = 0.2 and K i = 1.5 is
xi (t ) =
0.833 s + 0.134 H (s) = 0.833 s + (0.2 s + 1.5) s + 0.134 0.167 s + 1.25 = 2 s + 0.301s + 1.25 Equation (k) is that of a second-order system with a natural frequency of n = 1.25 (0.2 s + 1.5)
(k)
= 1.118 r/s
and a damping ratio of
0.301 2(1.118) = 0.135 Since the system is underdamped its damped natural frequency is
(l)
(m)
d = 1.118 1 0.(0601) 2
= 1.116 r/s impulsive response is determined using Table 6.5 as (0.135)(1.118) xi (t ) = (0.167)e ( 0.135)(1.118)t cos(1.118t ) + sin(1.118t ) + 1.116 1.25 ( 0.135)(1.118) t e sin(1.116t ) 1.116 = 0.167e ( 0.135)(1.118)t [cos(1.118t ) + 0.135 sin(1.118t )]1.120e 0.151t sin(1.116t ) (n)
(o)
Since a PI controller is used the offset is zero. (d) The closed loop transfer function for the system when the plant is placed in a feedback loop with a PD controller with K p = 0.2 and K d = 1.5 is
319
Chapter 8 0.833 s + 0.134 H (s) = 0.833 1 + (0.2 + 1.5s ) s + 0.134 1.25s + 0.167 = 2.25s + 0.301 0.556 s + 0.0742 = s + 0.134 Equation (p) is that of the transfer function of a first-order system of time constant 1 T= 0.134 7.46 s Equation (p) can be rewritten as H ( s ) = 0.556 Thus the systems impulsive response is xi (t ) = 0.556 (t ) (e) The transfer function when the plant is placed in a feedback loop with a controller with K p = 0.2 , K d = 1.5 and K i = 1.5 is (0.2 + 1.5s) 0.833 s + 0.134 H ( s) = 0.833 2 s + (1.5s + 0.2 s + 1.5) s + 0.134 2 1.25s + 0.167 s + 1.25 = 2.25s 2 + 0.301s + 1.25 0.556s 2 + 0.0742s + 0.556 = s 2 + 0.134s + 0.556 Equation (t) is that of a second order system with natural frequency n = 0.556 (1.5s 2 + 0.2 s + 1.5)
(p)
(q)
(r)
(s) PID
(t)
= 0.745 r/s
and damping ratio
(u)
=
Equation (t) can be rewritten as
(v) (w)
(x)
0.247 s + 0.134 s + 0.556 The impulsive response is obtained by inverting Equation (w) leading to xi (t ) = 0.556 (t ) + 0.333e 0.167 t sin( 0.742t ) Since a PID controller is used the offset is zero.
H ( s ) = 0.556 +
2
320
Chapter 8 8.15 (a) The differential equations governing the perturbations in liquid level are dh 1 1 A 1 + h1 h2 = q dt R R dh 1 2 A 2 h1 + h2 = 0 dt R R Substitution of given values in Equations (a) and (b) leads to dh 1.2 1 + 0.161h1 0.161h2 = q dt dh 1.2 2 0.161h1 + 0.322hh = 0 dt
(a) (b)
(c) (d)
Taking the Laplace transforms of Equations (c) and (d) assuming all initial conditions are zero leads to 0.161 H 1 ( s) Q( s ) 1.2s + 0.161 = (e) 0.161 1.2s + 0.322 H 2 ( s) 0 The transfer functions are obtained from Equation (e) as 1.2 s + 0.322 (f) G1 ( s ) = 2 1.44 s + 0.577 s + 0.0259 0.161 G2 ( s) = (g) 2 1.44 s + 0.577 s + 0.0259 The closed-loop transfer function for this arrangement is obtained using the annotated block diagram
C (s) =
(h) (i)
Q( s)Ga ( s) + 0.161C ( s ) 1.2s + 0.161 + Ga ( s) Substituting Equation (j) into Equation (h) leads to B( s) =
(j)
321
Chapter 8
Q( s )Ga ( s ) + 0.161C ( s ) 1.2s + 0.322 C ( s) = 0.161 1.2s + 0.161 + Ga ( s) Equation (k) is solved for C(s) as 0.161Q( s)Ga ( s) C (s) = 2 1.44s + 0.577 s + 0.0259 + (1.2s + 0.322)Ga ( s) The closed-loop transfer function for this system is C ( s) H 2 ( s) = Q( s ) 0.161G a ( s ) = 2 1.44 s + 0.577 s + 0.0259 + (1.2 s + 0.322)G a ( s ) Substitution of Equation (l) into Equation (h) leads to 0.161Q ( s )Ga ( s ) 1.2 s + 0.322 B( s) = 0.161 1.44 s 2 + 0.577 s + 0.0259 + (1.2 s + 0.322)Ga ( s )
H 1 ( s) = = B( s) Q( s )
2
(k)
(l)
(m)
(n)
Note that in the remainder of the solution of this problem the numerators for both H 1 ( s) and H 2 ( s) should be multiplied by 0.161, as should the values on all figures. (b) If G a ( s ) = 0.2 , Equation (m) becomes 0 .2 H 2 ( s) = 2 1.44 s + 0.817 s + 0.0903 0.139 = 2 (o) s + 0.567 s + 0.0627 and Equation (n) becomes 1.03s + 0.278 (p) H 1 (s) = 2 s + 0.567 s + 0.0627 The poles of the transfer function are the roots of the denominator of the transfer functions which are s1 = 0.151, s 2 = 0.416 (q) Since both poles are real and negative the system is stable. The impulsive and step responses for the first tank obtained using MATLAB are given below N=[1.03 0.278]; >> D=[1 0.567 0.0627]; >> impulse(N,D) >>
322
Chapter 8
323
Chapter 8 (c) If G a ( s ) =
1 .5 Equation (m) becomes s
1.5 (1.44s + 0.577s + 0.0259)s + (1.2s + 0.322)1.5 1.5 = 3 1.44 s + 0.577 s 2 + 1.826 s + 0.483 While Equation (n) becomes H 2 ( s) =
2
(r)
(7.45s + 2)1.5 (1.44s + 0.577s + 0.0259)s + (1.2s + 0.322)1.5 11.18s + 3 (s) = 3 1.44 s + 0.577 s 2 + 1.826 s + 0.483 The poles, impulsive, and step response corresponding to H 1 ( s) are obtained using the MATLAB commands H 1 ( s) =
2
>> N=[11.18 3]; >> D=[1.44 0.577 1.826 0.483]; >> p=roots(D) p= -0.0643 + 1.1086i -0.0643 - 1.1086i -0.2720 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles , -0.0643 + 1.1086i, -0.0643 - 1.1086i, and -0.2720 all have negative real parts the system is stable. The impulsive and step responses are
324
Chapter 8
325
(t)
H 1 ( s) = =
1.49s 2 + 11.58s + 3 (u) 1.44s 3 + 0.817 s 2 + 1.89s + 0.483 The poles, impulsive, and step response corresponding to H 1 ( s) are obtained using the MATLAB commands >> N=[1.49 11.58 3] N= 1.4900 11.5800 3.0000 >> D=[1.44 0.817 1.89 0.483]; >> p=roots(D) p= -0.1476 + 1.1002i -0.1476 - 1.1002i -0.2722 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles -0.1476 + 1.1002i, -0.1476 - 1.1002i, and -0.2722 all have negative real parts the system is stable. The impulsive and step responses are given below
326
Chapter 8
327
Chapter 8 (e) If G a ( s) = 1.5s + 0.2 the transfer functions of Equations (m) and (n) become H 2 (s) = 1.5s + 0.2 1.44s + 0.577 s + 0.0259 + (1.2s + 0.322)(1.5s + 0.2) 1.5s + 0.2 = 2 3.24s + 1.30s + 0.0644
2
(v)
H 1 (s) =
11.18s 2 + 4.49s + 0.4 = (w) 3.24s 2 + 1.30s + 0.0644 The poles, impulsive, and step response corresponding to H 1 ( s) are obtained using the MATLAB commands >> N=[11.18 4.49 0.4]; >> D=[3.24 1.30 0.0644]; >> p=roots(D) p= -0.3433 -0.0579 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles, -0.3433 and -0.0579 both have negative real parts the system is stable. The impulsive and step responses are given below.
328
Chapter 8
329
Chapter 8
1 .5 the transfer functions of Equations (m) and (n) become s 1 .5 s 2 + 0 .2 s + 1 .5 H 2 (s) = 1.44 s 2 + 0.577 s + 0.0259 s + (1.2 s + 0.322) 1.5s 2 + 0.2 s + 1.5
(x)
H 1 (s) = =
(7.45s + 2)(1.5s 2 + 0.2 s + 1.5) 1.44 s 2 + 0.577 s + 0.0259 s + (1.2 s + 0.322) 1.5s 2 + 0.2 s + 1.5
11.18s 3 + 4.49 s 2 + 11.58s + 3 (y) 3.24 s 3 + 1.3s 2 + 1.89 s + 0.483 The poles, impulsive, and step response corresponding to H 1 ( s) are obtained using the MATLAB commands >> N=[11.18 4.49 11.58 3]; >> D=[3.24 1.3 1.89 0.483]; >> p=roots(D)
p= -0.0646 + 0.7376i -0.0646 - 0.7376i -0.2719 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles -0.0646 + 0.7376i, -0.0646 - 0.7376i, and -0.2719 all have negative real parts the system is stable. The impulsive and step responses follow
330
Chapter 8
331
Chapter 8 8.16 (a) The differential equations governing the perturbations in liquid level are dh 1 1 A 1 + h1 h2 = q (a) dt R R dh 1 2 A 2 h1 + h2 = 0 (b) dt R R Substitution of given values in Equations (a) and (b) leads to dh 1.2 1 + 0.161h1 0.161h2 = q (c) dt dh 1.2 2 0.161h1 + 0.322hh = 0 (d) dt Taking the Laplace transforms of Equations (c) and (d) assuming all initial conditions are zero leads to 0.161 H 1 ( s) Q( s ) 1.2s + 0.161 = (e) 0.161 1.2s + 0.322 H 2 ( s) 0 The transfer functions are obtained from Equation (e) as 1.2 s + 0.322 (f) G1 ( s ) = 2 1.44 s + 0.577 s + 0.0259 0.161 (g) G2 ( s) = 2 1.44 s + 0.577 s + 0.0259 The closed-loop transfer function for this arrangement is obtained using the annotated block diagram 0.161 (h) C ( s) = B( s) 1.2 s + 0.322 1 {[Q( s) C ( s)]Ga ( s) + 0.161C ( s)} (i) B(s) = 1.2 s + 0.161 Use of Equation (i) in Equation (h) leads to
1 1.2 s + 0.322 {[Q ( s ) C ( s ) ]G a ( s ) + 0.161C ( s )} C (s) = 1.2 s + 0 .161 0.161 0 .161G a ( s )Q ( s ) C (s) = 2 1.44 s + 0.577 s + 0.0259 + 0.161G a ( s ) H 2 (s) = 0.161G a ( s ) 1 .44 s 2 + 0.577 s + 0.0259 + 0.161G a ( s ) (j)
Substituting Equation (j) in Equation (h) leads to 0.161Ga ( s )Q( s) 1.2s + 0.322 B( s) = 2 0.161 1.44s + 0.577 s + 0.0259 + 0.161Ga ( s) (1.2s + 0.322)Ga ( s) H 1 ( s) = 2 1.44s + 0.577 s + 0.0259 + 0.161Ga ( s ) (b) If G a ( s ) = 0.2 , Equations (j) and (k) become
(k)
332
Chapter 8 0.161(0.2) 1.44s + 0.577 s + 0.0259 + 0.161(0.2) 0.0322 = (l) 2 1.44s + 0.577 s + 0.0581 (1.2s + 0.322)(0.2) H 1 (s) = 2 1.44s + 0.577 s + 0.0259 + 0.161(0.2) 0.24s + 0.0644 = (m) 1.44s 2 + 0.577 s + 0.0581 The following MATLAB commands are used to determine the poles, impulsive response, and step response for the second tank H 2 (s) =
2
>> N=[0.0322]; >> D=[1.22 0.577 0.0581]; >> p=roots(D) p= -0.3276 -0.1454 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles, -0,3276 and -0.1454 both have negative real parts, the system is stable. The impulsive and step responses follow.
333
Chapter 8
334
Chapter 8 (c) If Ga ( s ) =
1 .5 , Equations (j) and (k) become s
0.161(1.5) (1.44s + 0.577s + 0.0259)s + 0.161(1.5) 0.242 (n) = 3 1.44 s + 0.577 s 2 + 0.0259s + 0.24 (1.2s + 0.322)(1.5) H 1 ( s) = 2 (1.44s + 0.577s + 0.0259)s + 0.161(1.5) 1.80s + 0.483 (o) = 3 1.44 s + 0.577 s 2 + 0.0259s + 0.24 The following MATLAB commands are used to determine the poles, impulsive response, and step response for the second tank >> N=[0.242]; >> D=[1.44 0.577 0.0259 0.242]; >> p=roots(D) H 2 ( s) =
2
p= -0.7093 0.1543 + 0.4616i 0.1543 - 0.4616i >> impulse(N,D) >> figure >> step(N,D) >> Since the complex poles have positive real parts the system is unstable. The impulsive and step responses follow
335
Chapter 8
336
(p)
0.24s + 1.864s + 0.483 (q) 1.44s + 0.577 s 2 + 0.0581s + 0.242 The following MATLAB commands are used to determine the poles, impulsive response, and step response for the second tank >> N=[0.0322 0.242]; >> D=[1.44 0.577 0.0581 0.242]; >> p=roots(D) =
2 3
p= -0.6927 0.1460 + 0.4704i 0.1460 - 0.4704i >> impulse(N,D) >> figure >> step(N,D) >> Since the complex poles have positive real parts the system is unstable. The impulsive and step responses follow
337
Chapter 8
338
Chapter 8 (e) If G a ( s) = 1.5s + 0.2 , Equations (j) and (k) become 0.161(0.2 + 1.5s) H 2 ( s) = 2 (1.44s + 0.577s + 0.0259) + 0.161(0.2 + 1.5s) 0.0322 + 0.242s = 1.44s 2 + 0.8185s + 0.0581 (1.2s + 0.322)(0.2 + 1.5s) H 1 (s) = 2 1.44s + 0.577 s + 0.0259 + 0.161(0.2 + 1.5s )
(s)
1.80 s + 0.723s + 0.0644 (t) 1.44s 2 + 0.8185s + 0.0581 The following MATLAB commands are used to determine the poles, impulsive response, and step response for the second tank =
2
>> N=[0.0322 0.242]; >> D=[1.44 0.8185 0.0581]; >> p=roots(D) p= -0.4853 -0.0831 >> impulse(N,D) >> figure >> step(N,D) >> Since the poles have negative real parts, the system is stable. The impulsive and step responses follow.
339
Chapter 8
340
Chapter 8
1 .5 , Equations (j) and (k) become s 0.161(1.5s 2 + 0.2 s + 1.5) H 2 ( s) = 1.44 s 2 + 0.577 s + 0.0259 s + 0.161(1.5s 2 + 0.2 s + 1.5)
0.483s 2 + 0.0322 s + 0.242 1.44 s 3 + 0.8185s 2 + 0.0581s + 0.242 (1.2s + 0.322)(1.5s 2 + 0.2s + 1.5) H 1 ( s) = 1.44 s 2 + 0.577 s + 0.0259 s + 0.161(1.5s 2 + 0.2 s + 1.5)
(u)
1.80 s + 0.723s + 1.864 s + 0.483 (v) 1.44 s 3 + 0.8185s 2 + 0.0581s + 0.242 The following MATLAB commands are used to determine the poles, impulsive response, and step response for the second tank =
3 2
>> N=[0.483 0.0322 0.242]; >> D=[1.44 0.8185 0.0581 0.242]; >> p=roots(D) p= -0.7879 0.1098 + 0.4486i 0.1098 - 0.4486i >> impulse(N,D) >> figure >> step(N,D) >> Since the complex poles have positive real parts the system is unstable. The impulsive and step responses follow.
341
Chapter 8
342
s + 0.2 s + 0.16 + 0.2 K p The natural frequency of the second-order system is n = 0.16 + 0.2 K p
2
0.2 K p
(b)
(c)
2 n = 0.2
=
Requiring the damping ratio to be 0.8 leads to
(d)
0.8 =
(e)
However, from Equation (d) it is clear that the largest possible value of the damping ration is 0.25. Thus it is not possible to design a proportional controller such that the system has a damping ratio of 0.8.
343
Chapter 8 0.2 ( s + 0.4) 2 H ( s) = 0.2 s + Ki ( s + 0.4) 2 0.2 K i = s( s + 0.4) 2 + 0.2 K i Ki = 0.2 K i s + 0.8s + 0.16s + 0.2 K i
3 2
(b)
Rouths criteria shows that a third-order system with D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 is stable if a1 a 2 > a3 . Application of this criterion to the transfer function of Equation (b) leads to (0.8)(0.16) > 0.2 K i K i < 0.64 8.19 The transfer function of the plant is
0 .2 (a) ( s + 0 .4 ) 2 The closed-loop transfer function of the system when the plant is used in a feedback control loop with a PI controller is 0 .2 ( K i + sK p ) ( s + 0 .4 ) 2 H (s) = 0 .2 s + ( K i + sK p ) ( s + 0 .4 ) 2 0.2(K i + sK p ) = (b) s ( s + 0.4) 2 + 0.2( K i + sK p ) C A2 ( s) =
(c)
H ( s) =
(c)
Rouths criteria shows that a third-order system with D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 is stable if a1 a 2 > a3 . Application of this criterion to the transfer ( 0 .8)( 0 .16 + 0 .2 K p ) > 0 .2 function of Equation (c) leads K p > 0 .45 (d) to
344
Chapter 8 The closed-loop transfer function of the system when the plant is used in a feedback control loop with a PI controller is 0 .2 ( K i + sK p ) ( s + 0 .4 ) 2 H (s) = 0 .2 s + ( K i + sK p ) ( s + 0 .4 ) 2 0.2(K i + sK p ) = (b) s ( s + 0.4) 2 + 0.2( K i + sK p ) Substituting K p = 2 in Equation (b) leads to
H ( s) =
(c)
Rouths criteria shows that a third-order system with D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 is stable if a1 a 2 > a3 . Application of this criterion to the transfer function of Equation (c) leads to (0.8)(0.56) > 0.2 K i K i < 2.24 8.21 The transfer function of the plant is
0 .2 (a) ( s + 0 .4 ) 2 The closed-loop transfer function of the system when the plant is used in a feedback control loop with a PID controller is 0 .2 (K d s 2 + K p s + K i ) ( s + 0 .4 ) 2 H ( s) = 0 .2 s + (K d s 2 + K p s + K i ) ( s + 0 .4 ) 2 C A2 ( s) =
(d)
0 .2 ( K d s 2 + K p s + K i ) s ( s + 0 .4 ) 2 + 0 .2 ( K d s 2 + K p s + K i )
(b)
Substitution of K i = 1 and K p = 2 in Equation (b) leads to H ( s) = 0.2( K d s 2 + 2s + 1) s 3 + (0.8 + 0.2 K d ) s 2 + 0.56 s + 0.2 (c)
Rouths criteria shows that a third-order system with D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 is stable if a1 a 2 > a3 . Application of this criterion to the transfer function of Equation (c) leads to (0.8 + 0.2 K d )(0.56) > 0.2 (d) Equation (d) is satisfied for all positive values of K d . If K d is allowed to be negative, Equation (d) leads to K d > 2.21 (e)
345
Chapter 8
(b)
(b) Rouths criteria shows that a third-order system with D ( s ) = s 3 + a1 s 2 + a 2 s + a 3 is stable if a1 a 2 > a3 . Application of this criterion to the transfer function of Equation (b) leads to (0.6)(0.12) > 0.008 + 0.1K p
K p < 0.64
(c) Substitution of K p = 1 in Equation (b) leads to
H (s) =
(c)
0 .1 (d) s + 0.6 s + 0.12 s + 0.108 The MATLAB commands to determine the impulsive response for the system whose transfer function is Equation (b) are
3 2
>> N=[0.1]; >> D=[1 0.6 0.12 0.108]; >> impulse(N,D) >>
346
Chapter 8
0.1K i (b) s + 0.6s + 0.12s 2 + 0.008s + 0.1K i The root-locus method can be used to determine the values of K i for which the system is stable. The denominator of Equation (c) is written in the form of Equation (6.32) with Q ( s) = s 4 + 0.6s 3 + 0.12s 2 + 0.008s (c) R( s) = 0.1 (d) The MATLAB commands to draw the root-locus plot are =
4 3
The root locus plot shows that the system is stable for K i < 0.0138 (c) Substituting K i = 0.5 in Equation (b) leads to
0.05 (e) s + 0.6 s + 0.12 s 2 + 0.008s + 0.05 The MATLAB commands to determine the step response of the system whose transfer function is that of Equation (e) are H ( s) =
4 3
>> N=[0.05]; >> D=[1 0.6 0.12 0.008 0.05]; >> step(N,D) >>
348
Chapter 8
(b)
(b) Substituting K i = 0.5 and K p = 1.5 in Equation (b) gives 0.1K d s 2 + 0.1s + 0.05 (c) s 4 + 0.6s 3 + (0.12 + 0.1K d )s 2 + 0.108s + 0.05 The root-locus method can be used to determine the values of K i for which the system is stable. The denominator of Equation (c) is written in the form of Equation (6.32) with Q( s) = s 4 + 0.6s 3 + 0.12s 2 + 0.108s + 0.05 (d) H ( s) =
(e)
349
Chapter 8 >> Q=[1 0.6 0.12 0.108 0.05]; >> R=[0.1 0 0]; >> rlocus(R,Q) >> The resulting root-locus plot is
The root locus plot shows that the system is stable for K d > 3.38 . (c) Substituting K d = 0.5 in Equation (c) leads to 0.05s 2 + 0.1s + 0.05 (f) s 4 + 0.6s 3 + 0.17 s 2 + 0.108s + 0.05 The MATLAB commands to determine the step response of the system whose transfer function is Equation (f) are H (s) = >> N=[0.05 0.1 0.05]; >> D=[1 0.6 0.17 0.108 0.05]; >> step(N,D) >> The resulting step response is
350
Chapter 8
K p (28s + 400)
(c)
(d)
351
Chapter 8 1 = 0.7 1 + K p K p = 1.04 (f) (b) The total mass of the golf cart can be written as m = 500 + (g) where is the additional mass of the golfers and clubs. When empty the natural frequency is 20 r/s. Thus with the added mass k n = m = = The damping ratio of the golf cart with added mass is c = 2mn = 2(0.7)(500)(20) 500 2(500 + )20 500 + 0.7 1+ 500(20) 2 500 + 20 1+
500
(h)
(i)
500 The closed loop transfer function of the system with proportional damping is 2 K p (2 n s + n ) H ( s) = 2 2 s + 2 n (1 + K p ) s + n (1 + K p ) Defining =
(j)
500
28 400 Kp 1+ s + 1+ H (s) = 400 28 s2 + (1 + K p ) (1 + K p ) s + 1+ 1+ The natural frequency of the closed loop system is 1+ K p n = 20 1+ The damping ratio of the closed loop system is
(k)
(l)
352
Chapter 8
1+ K p 1+ 1+ K p 1+ (m)
28
=
2(20) = 0 .7
1+ K p
1+ The system is critically damped when its damping ratio is one, thus 1+ K p 1 = 0.7 1+
K p = 1.041 + 2.041 500 = 1.041 + 0.0408 1 = 0.7 1 + K p K p = 1.04 (f) = 1.041 + 2.041
(n)
8.26 The total mass of the golf cart can be written as m = 500 + (g) where is the additional mass of the golfers and clubs. When empty the natural frequency is 20 r/s. Thus with the added mass k n = m = = The damping ratio of the golf cart with added mass is 500(20) 2 500 + 20 1+
500
(a)
353
Chapter 8
=
=
0.7 1+
500 Defining = / 500 m, the transfer function of the plant is 28 400 s+ 1+ 1+ (c) G(s) = 28 400 2 s + s+ 1+ 1+ (a) The closed-loop transfer function when the plant is placed in a feedback loop with a proportional controller is 28 400 Kp 1+ s + 1+ H (s) = (d) 28 400 2 s + (1 + K p ) s + (1 + K p ) 1+ 1+ The impulsive response is the inverse of the transfer function. The closed-loop system when a proportional controller is used is a second-order system. The increase in the mass leads to a decrease in the damping ratio. Thus as the mass increases the damping ratio of the system decreases, leading to larger overshoot for the impulsive response. (b) The closed-loop transfer function of the system when an integral controller is used is 28 400 Ki 1+ s + 1+ H ( s) = 28 28 400 400 + Ki s s 2 + s+ 1+ 1+ s + 1+ 1 + 28 400 Ki 1+ s + 1+ = 400 K i 28 2 400 + 28K i s3 + s + s+ 1+ 1+ 1+ Application of Rouths stability criterion for a third-order system requires 28 400 + 28 K i 400 K i > 1 + 1 + 1+ (28)(400 + 28K i ) > (1 + )400 K i
(e)
(f)
354
Chapter 8 The range of is 0<<0.5. Equation (f) shows that the system is stable for all values of in this range. The mass does not affect the stability, but does affect relative stability. The larger the value of , the closer the real part of the dominant pole is to zero. (c) The closed-loop transfer function when the plant is placed in a feedback loop with a PD controller is 28 400 ( K d s + K p ) 1+ s + 1+ H ( s) = 28 28 400 400 s2 + 1 + s + 1 + + ( K d s + K p ) 1 + s + 1 + 28 400 ( K d s + K p ) 1+ s + 1+ = 400 K p 28 K d 2 28 + 400 K d + 28 K p s + 1 + s+ 1+ 1+ 1+ (K d s + K p )(28s + 400) = (g) (1 + + 28K d )s 2 + (28 + 400 K d + 28K p ) s + 400 K p The transfer function of Equation (g) is that of a second-order system. After simplification the increase in mass appears only in the denominator multiplying the s 2 term. For a given controller, this has the effect of decreasing both the natural frequency and the damping ratio. 8.27 The transfer function for the system is
2 2 n s + n (a) 2 s 2 + 2 n s + n When the golf cart is empty the natural frequency is 20 r/s and has a damping ratio of 0.7. Thus the transfer function for the empty gold cart is 28s + 400 G ( s) = 2 (b) s + 28s + 400 If the plant is placed in a feedback loop with an integral controller of gain 1.3 the resulting closed-loop transfer function is 1.3( 28s + 400) H ( s) = 2 s( s + 28s + 400) + ( 28s + 400)(1.3) 36.4 s + 520 = 3 (c) s + 28s 2 + 436.4 s + 520 The step response is obtained by inverting 36.4 s + 520 (d) X s (s) = 3 s (s + 28s 2 + 436.4 s + 520 ) The MATLAB commands to obtain the poles and residues for Equation (d) are >> N=[36.4 520]; >> D=[1 28 436.4 520 0]; >> [r,p,k]=residue(N,D)
G ( s) =
355
Chapter 8 r= -0.0048 + 0.0471i -0.0048 - 0.0471i -0.9905 1.0000 p= -13.3530 +14.9512i -13.3530 -14.9512i -1.2940 0 k= [] The above leads to a partial fraction decomposition of 1 0.9905 0.00953s + 1.54 2 (e) X s (s) = s s + 1.294 s + 26.7 s + 401.84 Inversion of Equation (e) leads to x s (t ) = 1 0.9905e 1.294t e 13.35t (0.00953 cos( 20.05t ) + 0.0705 sin( 20.05t ) ) (f) (b) The 20% settling time is the time required for the step response to be permanently within 2% of its final value. To this end 0.98 = 1 0.9905e 1.294 t s e 13.35t s (0.00953 cos( 20.05t s ) + 0.0705 sin( 20.05t s ) ) (g) Trial and error leads to t s = 3.02 s (c) The 10%-90% rise time is determined as the time required for the response to grow from 10% of its final value to 90%of its final value. It is determined as 1.65 s. 8.28 The transfer function for the system is
2 2 n s + n (a) 2 s 2 + 2 n s + n When the golf cart is empty the natural frequency is 20 r/s and has a damping ratio of 0.7. Thus the transfer function for the empty gold cart is 28s + 400 G ( s) = 2 (b) s + 28s + 400 When the plant is placed in a feedback loop with a differential controller of gain 1.2 the closed loop transfer function is
G ( s) =
356
Chapter 8 1.2s (28s + 400) s + 28s + 400 + 1.2(28s + 400) 33.6s + 480 = 2 (c) s + 51.6s + 880 The step response of the system is obtained using MATLAB by >> N=[33.6 480]; >> D=[1 51.6 880]; >> step(N,D) >> The final value of the step response is obtained using the final value theorem as 480/880=0.5455. (b) The 2% settling time is the time it takes for the step response to be permanently within 2% of its final value: (0.02)(0.5455)=0.0109, or for the response to be permanently less than 0.5564 and greater than 0.5346. The plot of the step response shows that the settling time is 0.18 s H ( s) =
2
(c) The 10-90 percent rise time is the time it takes for the response to grow from (0.1)(0.5545) to (0.9)(0.5545) or from 0.05545 to 0.4990. The plot of the step response shows that the rise time is 0.0219 s- 0.00172 s=0.2173 s
357
Chapter 8
8.29 The SIMULINK model for the system of Example 8.2 follows the block diagram for the system presented in the solution of Example 8.2. The SIMULINK model is shown below
Customized plots from the data sent to the workspace are developed using the MATLAB commands >> t1=CA(1).time; >> CA1=CA(1).signals.values; >> plot(t1,CA1) >> xlabel('t') >> ylabel('C_A') >> title('Simulation of C_A for Problem 8.29')
358
Chapter 8 >> t2=CB(1).time; >> CB1=CB(1).signals.values; >> plot(t2,CB1) >> xlabel(t) >> ylabel(C_B) >> title (Simulation of C_B for Problem 8.29) >> The customized plots are shown below.
359
Chapter 8 8.30 The transfer function for the system of Example 8.25 is 7.49 s + 250 G ( s) = 2 (a) s + 7.49 s + 250 The SIMULINK model for the system when this plant is placed in a feedback loop with a PID controller is
A customized plot is developed from the MATLAB worksheet through the commands >> plot(t1,x1) >> xlabel('t') >> ylabel('x') >> title('Simulation of Problem 8.30') >>
360
Customized plots for the concentrations in each reactor are developed from the time and concentration values sent to the worksheet. The commands to draw and annotate the plot for the concentration in the first reactor are >> t1=CA1(1).time; >> CA11=CA1(1).signals.values; >> plot(t1,CA11) >> xlabel('t') >> ylabel('C_A_1') >> title('Concentration of reactant A in first reactor')
361
Chapter 8
Note, that as predicted in Problem 8.24, the system is unstable. 8.32 The closed-loop transfer function is
K s + 20 s + 80 + K The denominator of Equation (a) is of the form of Equation (6.32) with H ( s) =
3
(a)
362
Chapter 8 Q( s ) = s 3 + 20s + 80 (b) R( s) = 1 (c) The poles of the diagram, the roots of Q(s) are -2.85, 1.423 5.107 j . There are no zeroes of the diagram and all branches end at infinity. The poles are located in Figure (a).
Since Q(s) can be factored into ( s + 2.85)( s 2 + 2.846s + 28.1) and it is shown in Appendix D that quadratic factors do not affect the portion of the real axis which may be on the a branch of the root locus diagram application of the angle criterion becomes s + 2.85 = (2n + 1) (d) If s is a real value less than -2.85 then s + 2.85 = and the angle criterion is satisfied. Now if s is a real value greater than -2.85, then s+2.85>0 and s + 2.85 = 0 . Thus the angle criterion cannot be satisfied. Thus the only part of the real axis which is on a branch of a root locus is s<-2.85, as illustrated in Figure (b). The breakaway points are obtained by solving dQ dR R Q =0 ds ds (1)(3s 2 + 20) ( s 3 + 20s + 90)(0) = 0 3s 2 + 20 = 0 (e)
363
Chapter 8 Since there are no real solutions of Equation (e), there are no breakaway points for this root-locus diagram. There are three poles and no zeroes for the diagram. Thus the angles at which the asymptotes to the branches approach infinity are determined by ( 2n + 1) a = = (2n + 1) (f) 03 3 For n=0, Equation (f) leads to a = , For n=1, Equation (f) gives a = , and for 3 n=-1, Equation (a) gives a = . The asymptote corresponding to a = corresponds 3 to a branch along the negative real axis. The values where the asymptotes meet on the real axis is determined using Equation (D.13) as 1 s = [ 2.85 + 1.423 + j 5.107 + 1.423 j5.107] 3 =0 (g) The asymptotes are sketched in Figure (c). The crossover frequencies and the gain at cross-over are obtained from ( j ) 3 + 20( j ) + 80 + K = 0
(h) Setting the real part of Equation (h) to zero leads to a gain at cross-over of K=-80. Thus there are no positive values for which cross-over occurs. This implies that the branches originating in the right half plane remain in the right-half plane.
The angle of departure at a complex pole is obtained using the angle criterion 5.107 p = tan 1 2 2.85 + 1.423 = 0.6967 rad = 39.9 The root-locus diagram is sketched in Figure (e).
(20 2 ) j + 80 + K = 0
(i)
The MATLAB generated root-locus diagram, shown below, verifies the sketch of Figure (f).
364
Chapter 8
8.33 The dominant poles of the root-locus diagram of Problem 8.32 have positive real parts for all values of K. Thus this problem has no answer. 8.34 The transfer function for a second-order plant is
1 (a) s + 20 s + 80 When the plant is placed in a feedback loop with a PI controller with K p = 30 the closed-
G(s) =
(c) (d) R( s) = 1 The poles of the diagram are the roots of Q(s) which are s = 0,10 3.16 j . Since R(s)=1, there are no zeroes of the diagrams. Each branch of the root-locus diagram begins at a pole. Since there are no zeroes, each branch ends at infinity. It is noted that Q( s ) = s ( s 2 + 20s + 110) . Thus since a quadratic factor has an angle of zero when evaluated along the real axis, the points on the real axis which are on a branch of the diagram are determined by s . Since s = for all negative values of s and s = 0 for all positive values of s, the angle criterion implies that all points on the
30 s + K i s + 20 s 2 + 110 s + K i The denominator of Equation (b) is of the form of Equation (6.32) with Q( s ) = s 3 + 20s 2 + 100s
(b)
365
Chapter 8 negative real axis are on a branch of the root locus while no points on the positive real axis lie on a root locus. The breakaway points are obtained by finding the real values of s, on the negative real axis, such that dQ dR R Q =0 ds ds (1)(3s 2 + 40s + 110) = 0 s = 9.46, 3.88 (e) It is known that (a) the poles are at s=0 and s = 10 3.16 j (b) all points on the negative real axis are on a branch of the root locus and (c) the breakaway or break-in points are at s=-9.46 and s=-3.88. This knowledge, coupled with the knowledge that two branches may not intersect over a finite region, leads to the following conclusions: (a) the branch which begins at s=0 has a breakaway point at s=-3.88, (b) The branches that begin at the complex poles have a break-in point at s=-9.46. One branch then moves along the negative real axis toward s = as K increases while the other branch moves along the negative real axis toward s=0, but has a breakaway point at s=-3.88. These deductions are illustrated in Figure (a).
There are three poles and no zeroes for the diagram. Thus application of Equation (D.12) to determine the angles of the asymptotes leads to ( 2n + 1) a = = (2n + 1) (f) 03 3 Evaluation of Equation (f) for n=0,1, and -1 leads to the angles of the asymptotes as a = , , (g) 3 3 The branch with the asymptote of is the branch which travels along the negative real axis to s = . The points of intersection of the asymptotes with the real axis are obtained using Equation (D.13),
366
Chapter 8 1 [0 10 3.16 j 10 + 3.16 j ] 3 = 6.67 The asymptotes are illustrated in Figure (b) . s=
(h)
The crossover frequencies and the corresponding gains are obtained from ( j ) 3 + 20( j ) 2 + 110( j ) + K = 0
(110 2 ) j + K 20 2 = 0
Setting the imaginary part of Equation (i) to zero leads to a crossover frequency of c = 110 = 10.49 r/s Setting the real part of Equation (i) to zero leads to a crossover gain of K c = 20 c2 = 2200 The above information is used to sketch the root-locus diagram of Figure (c).
(i)
(j)
(k)
The MATLAB generated root locus diagram confirms the sketch 367
Chapter 8
G(s) =
30s + K i (c) s + 20s 2 + 110s + K i The root-locus diagram for the transfer function of Equation (c) is obtained by defining Q( s ) = s 3 + 20s 2 + 110s (d) R( s) = 1 (e) The following MATLAB commands are used to draw the root-locus diagram H (s) =
3
368
The root-locus diagram shows that the largest value of the integral gain such that the dominant pole is less than -2 is K i = 714 . The corresponding damping ratio is 0.304 with a frequency of 6.69 r/s. 8.36 The transfer function for a plant is
1 (a) s + 20 s + 80 The closed-loop transfer function for the system when it is placed in a feedback loop with a PID controller with K p = 30 and K i = 50 is
G(s) =
369
Chapter 8
K d s 2 + 30s + 50 H ( s) = s( s 2 + 20s + 80) + K d s 2 + 30s + 50 K d s 2 + 30s + 50 s 3 + 20s 2 + 110s + 50 + K d s 2 The denominator of Equation (b) is of the form of Equation (6.32) with Q( s) = s 3 + 20s 2 + 110s + 50 =
2
(b)
(c)
R( s) = s (d) The poles of the diagram are the roots of Q(s) which are s = 0.4986, - 9.751 2.28 j , Each branch originates at a pole. The zeroes of the diagram are the roots of R(s) which has a double root at s=0. Thus the branches end either at infinity or at s=0.
The angle criterion is used to determine the parts of the real axis which lie on a branch of the root locus. Since Q( s) = ( s + 0.4896)(s 2 + 19.5s + 100.3) and R ( s) = s 2 and a quadratic factor has an angle of zero along the real axis the angle criterion reduces to (a) s + 0.4896 = (2n + 1) If s<-0.4896, s+0.4896 is negative and s + 0.4896 = . If s>-0.4896 then s+0.4896 is positive and s + 0.4896 = 0 . Thus, the portion of the real axis such that s<-0.4896 is on some branch of a root locus. Note however that the zero s=0 is a special case in that 0 can be any value. Thus s=0 could be on a branch of the root locus. The breakaway and break-in points are the real values of s which lie on a branch and satisfy dQ dR R Q =0 ds ds ( s 2 )(3s 2 + 40s + 110) ( s 3 + 20s 2 + 110 s + 50)(2s ) = 0 2s 4 110s 2 100s = 0 s (2s 3 110s 100) = 0 s = 0,10,0.9161,10.9161 (b) The following is known about the root locus diagram (a) The origins of the branches are s = 0.4986, - 9.751 2.28 j . (b) The branches terminate either at s=0 or at infinity (c) Every point of the portion of the real axis , s<-0.4896 is on a branch of the root locus (d) The breakaway points and break-in points are s=0,-10, -0.9161. The above knowledge along with the knowledge that two branches cannot intersect over a finite region leads to the following conclusions (a) The branch of the root locus which originates at s=-0.4896 travels along the negative real axis for increasing K until it reaches s=-0.9161 where its branch breaks away from the axis. (b) The branches originating at s = - 9.751 2.28 j have break-in points at s=10. One branch travels along the negative real axis for increasing K, terminating at infinity. The other branch, for increasing K, travels along the negative real axis toward s=0 until it reaches a breakaway point at
370
Chapter 8 S = -0.9161. The branches that have breakaway points have break-in points at s=0 where they terminate. A sketch of the root locus curves drawn form this knowledge is shown in Figure (a). Angles of departure and arrival may be calculated as well as asymptotes and the intersection of the asymptotes with the real axis. It can also be shown that the crossover frequency is zero which corresponds to a K of infinity. The root locus drawn using MATLAB confirms the above observations
8.37 The transfer function for a second-order plant is 2 G ( s) = (a) ( s + 1.2)( s + 1.8) The closed-loop transfer function of the system when the plant is used in a feedback control loop with a PI controller is 2 ( K i + sK p ) ( s + 1.2)( s + 1.8) H ( s) = 2 s + ( K i + sK p ) ( s + 1.2)( s + 1.8) 2(K i + sK p ) = s ( s + 1.2)( s + 1.8) + 2( K i + sK p )
= 2K p s + 2K i s 3 + 3s 2 + (2.16 + 2 K p ) s + 2 K i H (s) = 12 p s + 2 K i s 3 + 3s 2 + 14.16 s + 2 K i (b)
371
Chapter 8 Equation (c) is in the form of Equation (6.32) with Q( s ) = s 3 + 3s 2 + 14.16s R( s) = 2 (a)The MATLAB commands to draw the root-locus diagram for this system are >> Q=[1 3 14.16 0]; >> R=[2]; >> rlocus(R,Q) >>
(d) (e)
(b) The minimum overshoot of 25.5 percent corresponds to a gain of zero. (c) The minimum overshoot corresponds to a frequency of 3.76 r/s and a damping ratio of 0.399. 8.38 The transfer function for a second-order plant is 2 (a) G ( s) = ( s + 1.2)( s + 1.8) The closed-loop transfer function of the system when the plant is used in a feedback control loop with a PI controller is
372
Chapter 8 2 ( s + 1.2)( s + 1.8) H ( s) = 2 s + ( K i + sK p ) ( s + 1.2)( s + 1.8) 2(K i + sK p ) = s ( s + 1.2)( s + 1.8) + 2( K i + sK p ) ( K i + sK p ) (b) s + 3s + (2.16 + 2 K p ) s + 2 K i It is desired to determine the minimum overshoot for several values of the proportional gain. Equation (b) can be written in the form of Equation (6.32) with Q( s) = s 3 + 3s 2 + (2.16 + 2 K p ) s (c)
3 2
2K p s + 2K i
R( s) = 2 The MATLAB commands to generate the root locus diagram for K p = 0.2 are >> Kp=0.2; >> Q=[1 3 2.16+Kp 0]; >> R=[2]; >> rlocus(R,Q) >> The corresponding root locus diagram indicating the minimum overshoot is
(d)
373
Chapter 8
The root locus diagram can be drawn in this fashion for any value of the proportional gain. The following table gives, for several values of proportional gain, the minimum overshoot, corresponding gain, damping ratio, and frequency.
Kp
min
0 0 0.0396 1.42 3.24 11.9 22.4 38.9
0.469 to 1.18 0.557 to 1.47 1.08 1.46 1.85 2.68 3.49 5.21
374
s+2 (a) s + 3s 2 + 5s + 10 (a) When placed in a feedback loop with a proportional controller the closed loop transfer function is s+2 Kp 3 s + 3s 2 + 5s + 10 H (s) = s+2 1+ K p 3 s + 3s 2 + 5s + 10 K p ( s + 2) = 3 (b) 2 s + 3s + 5s + 10 + K p ( s + 2) The root locus diagram is drawn for this system using Q( s ) = s 3 + 3s 2 + 5s + 10 (c)
R( s) = s + 2 The appropriate MATLAB commands to draw the root locus diagram are >> Q=[1 3 5 10]; >> R=[1 2]; >> rlocus(R,Q) >> The resulting root locus diagram is
(d)
(b) The closed-loop transfer function when the plant is placed in a feedback loop with a PI controller is
375
Chapter 8
s+2 s + 3s 2 + 5s + 10 H (s) = s+2 s + (K p s + K i ) 3 s + 3s 2 + 5s + 10 ( K p s + K i )( s + 2) = 4 3 s + 3s + 5s 2 + 10 s + ( K p s + K i )( s + 2) (K p s + K i )
3
(e)
(25s + K i )( s + 2) (f) s + 3s + 30s 2 + (60 + K i ) s + 2 K i A root-locus diagram can be drawn for the system whose transfer function is Equation (f) with Q( s ) = s 4 + 3s 3 + 30s 2 + 60s (g) R( s ) = s + 2 (h) The MATALB commands used to generate the root-locus plot are H ( s) =
4 3
>> Q=[1 3 30 60 0] >> R=[1 2] >> rlocus(R,Q) >> The resulting root-locus diagram is
The root-locus diagram shows that the system is stable for K i < 21.2 . (c) The closed-loop transfer function for the system when the plant is placed in a feedback loop with a PID controller is
376
H ( s) =
s 4 + 3s 3 + 5s 2 + 10 s + ( K d s 2 + K p s + K i )( s + 2)
(i)
Substitution of K p = 25 and K i = 100 in Equation (i) leads to ( K d s 2 + 25s + 100) s 4 + (3 + K d ) s 3 + (30 + 2 K d ) s 2 + 160s + 200 A root-locus diagram is constructed for this system using Q( s ) = s 4 + 3s 3 + 30s 3 + 160s + 200 H (s) = (j)
(k) (l)
>> Q=[1 3 30 160 200]; >> R=[1 3 0 0]; >> rlocus(R,Q) >> The resulting root-locus diagram is
The root-locus diagram shows that the system is stable for K d > 2.29 (d) The system with the proportional controller is stable for all values of the proportional gain, but has offset. The use of integral control eliminates the offset, but is destabilizing 377
Chapter 8 and increases overshoot. The addition of derivative control is stabilizing, but relative stability can be a problem. s +1 (a) ( s + 2)( s + 3)( s + 4) The closed-loop transfer function when this plant is used in a feedback loop with a PID controller is s +1 (K d s 2 + K p s + K i ) ( s + 2)( s + 3)( s + 4) H (s) = s +1 s + (K d s 2 + K p s + K i ) ( s + 2)( s + 3)( s + 4) G ( s) = = = ( s + 1)( K d s 2 + K p s + K i ) s ( s + 2)( s + 3)( s + 4) + ( s + 1)(( K d s 2 + K p s + K i ) K d s 3` + ( K d + K p ) s 2 + ( K p + K i ) s + K i s 4 + (9 + K d ) s 3 + (+26 + K d + K p ) s 2 + (24 + K p + K i ) s + K i (b) 8.40 The transfer function of a third-order plant is
A MATLAB M file written to plot the root-locus curve, plot the impulsive response, and to plot the step response for input values of the controller gains is listed below % Program for Problem 8.40 % Design of a PID controller % Input controller gains disp('Please enter proportional gain') Kp=input('>>'); disp('Please enter integral gain ') Ki=input('>>'); disp('Please enter differential gain ') Kd=input('>>'); disp('Which parameter do you want to vary for root-locus curve?') disp('Enter 1 for proportional gain') disp('Enter 2 for integral gain') disp('Enter 3 for differential gain') i=input('>>') while i~=1&i~=2&i~=3 disp('Invalid enry, please reenter') disp('Enter 1 for proportional gain') disp('Enter 2 for integral gain') disp('Enter 3 for differential gain') i=input('>>'); end if i==1 Q=[1 9+Kd 26+Kd 24+Ki Ki]; R=[1 1 0]; elseif i==2
378
Chapter 8 Q=[1 9+Kd 26+Kd+Kp 24+Kp 0]; R=[1 1]; else Q=[1 9 26+Kp 24+Kp+Ki Ki]; R=[1 1 0 0]; end rlocus(R,Q) % Impulsive reponse and step response for input values N=[Kd Kd+Kp Kp+Ki Ki]; D=[1 9+Kd 26+Kd+Kp 24+Kp+Ki Ki]; figure impulse(N,D) figure step(N,D) % End of program Some output from the program follows. First consider K p = 100, K d = 10, K i = 10 , but with the root-locus diagram drawn for the integral gain. The root locus plot is
Root Locus 20
15
10
5 Imaginary Axis
System: sys System: sys Gain: 2.15e+003 Gain: 237 Pole: 0.0275 + 10.9i Pole: -7.08 - 3.7i Damping: -0.00251 Damping: 0.887 Overshoot (%): 101 Overshoot (%): Frequency (rad/sec): 10.9 0.243 Frequency (rad/sec): 7.99
-5
-10
-15
-20 -25
-20
-15
-5
379
10
20
30 Time (sec)
40
50
60
The figures show that the minimum overshoot occurs for a gain of 210 and that the settling time is almost 25 seconds for this design. Thus try a controller with a larger integral gain and a small differential gain. The next set is for Kp=100, Ki=100 and Kd=25 with the root-locus diagram drawn for varying Ki. 8.41 The transfer function of a third-order plant is 1 (a) s( s + 1)( s + 2) The closed-loop transfer function when the plant is placed in a feedback loop with a proportional controller is 1 H ( s) = s ( s + 1)( s + 2) + K 1 = 3 (b) 2 s + 3s + 2s + K The denominator of the transfer function of Equation (b) is of the form of Equation (6.32) with Q( s ) = s 3 + 3s 2 + 2s (c) R( s) = 1 (d) The MATLAB commands to draw the root-locus diagram for this system are >> Q=[1 3 2 0]; >> R=[1]; G ( s) = 380
Chapter 8 >> rlocus(R,Q) >> rlocus(R,Q) >> The resulting root-locus diagram is
(b) The transfer function of the plant when placed in a feedback loop with a compensator is s+a 1 K s + b s ( s + 1)( s + 2) H ( s) = s+a 1 1+ K s + b s ( s + 1)( s + 2) K (s + a) = (e) s ( s + 1)( s + 2)( s + b) + K ( s + a ) A maximum overshoot of 10 percent corresponds to a damping ratio of 0.591. Thus it is desired to design he compensator such that the dominant pole of the system has a damping ratio of 0.591 and a frequency of 1.5 r/s. From Equation (9.80) this implies the dominant pole is
s = (0.591)(1.5) j (1.5) 1 (0.591) 2
= 0.887 1.21 j (f) In order for this point to be on a branch of the root-locus diagram the angle criterion must be satisfied s + a s s +1 s + 2 s +b = (2n + 1) (g) where
381
Chapter 8
1.21 s = 0.887 + j (1.21) s = tan = 2.203 0.887 1.21 s + 1 = 0.113 + j (1.21) s +1 = tan 1 = 1.478 0.113 1.21 s + 2 = 1.113 + j (1.21) s + 2 = tan 1 = 0.827 1.113
(h)
(i)
Equation (h) becomes For n=-1, Equation (j) leads to Equation (k) requires
(j) (k)
(l)
Thus choosing a=0.319, b=2.5, and K=4.52, Equation (e) becomes 4 . 52 ( s + 0 . 319 ) H (s) = s ( s + 1)( s + 2 )( s + 2 . 5 ) + 4 . 52 ( s + 0 . 319 ) 4 . 52 s + 1 . 442 = 4 3 s + 5 . 50 s + 9 . 50 s 2 + 9 . 52 s + 1 . 44
(n)
The step responses without the compensator is determined using the MATLAB commands >> N=[1]; >> D=[1 3 2 0]; >> step(N,D) >> 382
Chapter 8
The step response of the system with the compensator is obtained using MATLAB by >> N=[4.52 1.44]; >> D=[1 5.5 9.5 9.52 1.44]; >> step(N,D) >> The resulting step response is
383
Chapter 8
8.42 The transfer function of the plant is 2 (a) s ( s + 5s + 20) Use of the Ziegler-Nichols tuning rules require knowledge of the crossover frequency when a proportional controller is used. The closed-loop transfer function when the plant of Equation (a) is placed in a feedback loop with a proportional controller is 2 K 2 s ( s + 5s + 20) H ( s) = 2 1+ K 2 s ( s + 5s + 20) 2K (b) = 2 s ( s + 5s + 20) + 2 K The root-locus diagram for the transfer function of Equation (b) is obtained by defining Q( s ) = s 3 + 5s 2 + 20 s (c) R( s) = 2 (d) The MATLAB commands used to generate the root-locus diagram are G ( s) =
2
>> Q=[1 5 20 0]; >> R=[2]; >> rlocus(R,Q) >> The resulting diagram is
384
Chapter 8 The root-locus diagram shows that the gain at the crossover frequency is K c = 50.7 and the crossover frequency is 4.5 r/s. (a) The Ziegler-Nichols tuning rule for a proportional controller is to choose K p = 0.5 K c = 0.5(50.7) = 25.4 (b) Application of the Ziegler-Nichols tuning rule for a PI controller gives K p = 0.45 K c (c)
= 22.8 Ki = = 4 3 c
(d)
4 3(4.5) = 0.931 (c) Application of the Ziegler-Nichols tuning rules for a PID controller leads to K p = 0 .6 K c
= 30.4 Ki =
(e)
(f)
c
(g)
= 0.698 Kd =
4 c
(h)
= 0.175
385
Chapter 8
386
Chapter 9
9. State-Space Methods
9.1 Consider the differential equation & + 3&+ 16 x + 25 x = F (t ) & & & x& x Define y1 = x & y2 = x & y3 = & x Use of Equations (b)-(d) in Equation (a) gives & y 3 + 3 y 3 + 16 y 2 + 25 y1 = F (t ) & y 3 = 25 y1 16 y 2 3 y 3 + F (t ) Equations (c), (d), and (e) are summarized in matrix form as & 1 0 y1 0 y1 0 y = 0 & 0 1 y2 + 0 2 y 3 25 16 3 y 3 F (t ) & Assuming the system output is x(t ) = y1 the output is given as
(a )
( e)
(f)
y1 x = [1 0 0] y 2 + [0][1]F (t ) y3 The state and input matrices are determined from Equation (f) as 1 0 0 0 A= 0 1 25 16 3
0 B = 0 1 The output matrix and the transmission matrix are determined from Equation (g) as C = [1 0 0] D = [0]
(g )
(h)
(i)
(j) (k)
9.2 Consider a system governed by the differential equations dy 0.2 1 + 2 y1 y 2 = F1 (t ) (a) dt dy 0.25 2 y1 + 3 y 2 = F2 (t ) (b) dt The system inputs are F1 (t ) and F2 (t ) . The system outputs are y1 (t ) and y 2 (t ) . Equations (a) and (b) are each divided by the coefficient of their derivative term leading to
387
Chapter 9
dy1 = 10 y1 + 5 y 2 + 5 F1 (t ) (c) dt dy 2 = 4 y1 12 y 2 + 4 F2 (t ) (d) dt Using y1 and y 2 as state variables, Equations (c) and (d) can be summarized in a matrix form as & y1 10 5 y1 5 0 F1 (t ) (e) y = 4 12 y + 0 4 F (t ) 2 2 2 &
The output relations are written as y1 1 0 y1 0 0 F1 (t ) y = 0 1 y + 0 0 F (t ) 2 2 2 The state matrix and input matrix are determined from Equation (e) as 10 5 A= 4 12
5 0 B= 0 4 The output matrix and transmission matrix are determined from Equation (f) as 1 0 C= 0 1
(f)
(g) (h)
(i) (j)
0 0 D= 0 0
9.3 Consider a system governed by the differential equations dy dy 2 1 + 3 2 + 4 y1 2 y 2 = F (t ) (a) dt dt dy dy 3 1 + 2 2 2 y1 + 4 y 2 = 0 (b) dt dt The system has one input F (t ) and two outputs, y1 (t ) and y 2 (t ) . Let y1 (t ) and y 2 (t ) be the state variables. Equations (a) and (b) are summarized in matrix form as & 2 3 y1 4 2 y1 (t ) 1 (c) 3 2 y + 2 4 y (t ) = 0 F (t ) 2 2 & Multiplying Equation (c) by the inverse of the coefficient matrix multiplying the vector of derivatives leads to
388
Chapter 9
2 3 2 3 & 5 5 4 2 y1 5 5 1 y1 F (t ) + y = 3 2 2 4 y 2 3 2 0 2 & 5 5 5 5 & y1 2.8 3.2 y1 0.4 y = 3.2 2.8 y + 0.6 F (t ) 2 2 & The relation between the input and output is y1 1 0 y1 1 y = 0 1 y + [0 0 ] 0 F(t) 2 2
(d)
(f)
The state matrix and input matrix are determined from Equation (d) as 2.8 3.2 A= 3.2 2.8 0.4 B= 0.6 The output matrix and transmission matrix are determined from Equation (f) as 1 0 C= 0 1
D = [0 0]
(h) (i)
(j) (k)
9.4 The mathematical model is given in Equations (b), (d), and (f) in the solution to Example 3.27 t t di1 di2 1 1 L +M + Ri1 + i1 dt i2 dt = v(t ) (a) dt dt C0 C0
M M
(b) (c)
The system has one input , v (t ) , and three outputs i1 (t ), i2 (t ) and i3 (t ) . Define the charges through the circuit branches by dq i1 = 1 (d) dt dq i2 = 2 (e) dt dq i3 = 3 (f) dt The state variables are the charges and the currents,
389
Chapter 9
y1 = q1 y2 = q2 y3 = q3
& y 4 = i1 = y1 & y 5 = i2 = y 2
( g1) ( g 2) ( g 3) ( g 4) ( g 5)
& ( g 6) y 6 = i3 = y 3 Equations (a)-(c) are rewritten using the state variables as 1 1 & & Ly 4 + My 5 + Ry 4 + y1 y 2 = v(t ) (h) C C 1 1 & & & My 4 + Ly 5 + My 6 + Ry5 Ry 6 y1 + y 2 = 0 (i) C C 1 & & My 5 + Ly 6 Ry 5 + Ry 6 + y 3 = 0 (j) C Equations (g4), (g5), (g6), (h), (i), and (j) are summarized in a matrix form as 0 0 1 0 0 0 & 1 0 0 0 0 0 y1 0 0 0 0 1 0 y1 0 0 1 0 0 0 0 y & 0 0 0 0 1 y 2 0 2 0 1 & 0 0 1 0 0 0 y 3 1 0 0 y 3 = 0 v (t ) (k ) R 0 = C & C y 4 1 0 0 0 L M 0 y 4 1 1 0 0 R R y 5 0 0 0 0 M L M y 5 & C C & 1 y 0 0 0 0 0 M L y 6 0 0 0 R R 6 C The inverse of the coefficient matrix on the left-hand side of Equation (k ) is determined using the symbolic capabilities of MATLAB as
1
1 0 0 0 0 0
0 1 0 0 0 0
0 0 0 0 1 0 0 L 0 M 0 0
0 0 0 M L M
0 0 0 0 M L
1 0 0 = 0 0 0
0 0 1 0 0 1 0 0 0 0 0 0
0 0 0 0 0 0 2 2 L M M 2 2 2 L(L 2 M ) L 2M 2 M L 2 2 2 L 2M L 2M 2 2 M M 2 2 2 L(L 2 M ) L 2M 2
0 0 0 2 M 2 2 L(L 2 M ) (l) M 2 L 2M 2 M 2 L2 2 L 2M 2
Pre-multiplying Equation (k) by the matrix on the right-hand side of Equation (l) leads to
390
Chapter 9
0 & y1 0 y & 0 2 y 3 + ML & = LC & y4 M + L y5 & C & y 6 M ( M + L) LC 0 0 0 0 0 0 M2 LC M C 1 0 0 0 1 0 MR( L + M ) L R( L + M ) R ( + ML) L y 0 1 y2 1 MR( L + M ) y 3 L y R( L + M ) 4 y5 R ( + ML) y 6 L 0
M +L C M ( M + L) LC LC 0 0 0 + v(t ) M M2 L
+ ML LC
R L
MR M 2R L
(m)
where
= L2 2 M 2 = L2 M 2
The state matrix and input matrix are determined from Equation (m) as
0 0 0 + ML A = LC M +L C M ( M + L) LC 0 0 0 0 0 M2 LC M C 1 0 0 1 0 MR ( L + M ) L R( L + M ) R( + ML) L 1 MR ( L + M ) L R( L + M ) R ( + ML) L 0 0
(n) (o)
M +L C M ( M + L) LC
+ ML LC
R L
MR M 2R L
(p)
LC
0 0 0 B= M 2 M L
(q)
391
(r)
(s)
(t)
(a)
v1 R1 v dv i2 = A C1 A R2 dt i1 =
(b)
(c)
392
Chapter 9 Substitution of Equations (b) and (c) into Equation (a) leads to v1 v A dv + + C1 A = 0 R1 R2 dt
dv A v v = A 1 dt R2 C R1C (d)
Application of KCL at node Q2 gives Assuming an ideal amplifier, i6 = 0 and vQ2 i 4 i5 i 6 = 0 = v 2 . Thus (e)
(f)
(g)
dv B v v dv = A + 2 + 2 (h) dt R3 C 2 R3 C 2 dt Equations (d) and (h) provide a mathematical model for the system. However note that Equation (h) has a term proportional to the derivative of an input. Define state variables as (i) y1 = v A
y 2 = v B + v 2 Use of these state variables in Equations (d) and (h) lead to 1 1 & y1 = y1 v1 R2 C1 R1C1
& & y 2 v 2 = 1 1 & y1 + v2 + v2 R3 C 2 R3 C 2
(j)
(k) (l)
0 y1 + 0 y 2 The output from the system is v B . Thus the output obtained from
1 R1C1 0
0 v 1 1 v 2 R3 C 2
(n)
393
Chapter 9
[v B ] = [0 1]
Substitution of numerical
C1 = C 2 = 30 F leads to
8.33 A= 5.56 16.67 B= 0
C = [0 1] D = [0 1]
9.6 The differential equations governing a mechanical system are & & x1 2 0 0 & 2 2 0 x1 3 2 0 x1 0 0 4 0 & + 2 2 0 x + 2 5 3 x = 0 & & x 2 2 2 & 0 & & 0 3 3 x 3 F ( t ) 0 0 x 3 0 0 3 x3 The state variables are defined by & & &T y = [x1 x 2 x3 x1 x 2 x3 ] The state matrix is as presented in Equation (9.25) where 1 1 0 2 0 0 3 2 0 1.5 1 0 4 0 2 5 3 = 0.5 1.25 0.75 M K= 1 1 0 0 3 0 3 3 0
1 0 2 0 0 2 2 0 1 1 0 4 0 2 2 0 = 0.5 0.5 0 M Cd = 0 0 0 0 0 0 0 3 0 Thus 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1 0 A= 1 1 0 1 0 1.5 0.5 1.25 0.75 0.5 0.5 0 1 1 0 0 0 0 The input matrix is given by Equation (9.26) where 1 2 0 0 0 0 M 1Q = 0 4 0 0 = 0 0 0 3 1 0.333
1
(a)
(b)
(c)
(d)
(e)
(f)
394
Chapter 9 Thus the input vector is 0 0 0 B= 0 0 0.333 The output matrix is 1 0 0 0 0 0 C = 0 1 0 0 0 0 0 0 1 0 0 0 and the transition matrix is (h)
(g)
D = [0]
(i)
9.7 The integro-differential equations derived in Example 3.14 are t di1 1 L1 + R1i1 + R2 (i1 i2 ) + (i1 i2 )dt = v(t ) dt C0
t di2 1 + R3 i2 + R2 (i2 i1 ) + (i2 i1 )dt = 0 dt C0 Substitution of given values into Equations (a) and (b) leads to t t di 0.2 i + 2500i1 + 4 x10 6 i1 dt 1500i2 4 x10 6 i2 dt = v(t ) dt 0 0
(a) (b)
L2
(c) (d)
The system has one input, v (t ) , and two outputs i1 (t ) and i2 (t ) . The state-space variables are the charges and currents (e1) y1 = q1
y2 = q2
& y 3 = i1 = y1 & y 4 = i2 = y 2 Use of Equations (e) in Equations (c) and (d) leads to & 0.2 y3 + 2500 y3 + 4 x106 y1 1500 y 4 4 x106 y 2 = v (t )
& 1500 y3 4 x106 y1 + 0.5 y 4 + 3500 y 4 + 4 x106 y 2 = 0 Equations (f) and (g) can be rewritten as & y 3 = 2 x10 7 y1 + 2 x10 7 y 2 1.25 x10 4 y 3 + 7.5 x10 3 y 4 + 5v(t ) & y 4 = 8 x10 6 y1 8 x10 6 y 2 + 3x10 3 y 3 7 x10 3 y 4 Equations (e3), (e4), (h) and (i) are summarized in matrix form as
395
Chapter 9 0 0 0 1 0 + v (t ) (j) 2 x10 7 1.25 x10 4 7.5 x103 5 8 x10 6 7 x10 3 0 3x10 3 The state matrix and input matrix are obtained for this state-space model using Equation (j) resulting in 0 1 0 0 0 0 0 1 (k) A= 2 x107 2 x107 1.25 x10 4 7.5 x103 6 3x10 3 8 x106 7 x10 3 8 x10 0 0 (l) B= 5 0 The relation between the state variables and the output variables is y1 i1 0 0 1 0 y 2 0 (m) = i 0 0 0 1 y + 0 v (t ) 3 2 y4 The output matrix and transmission matrix are determined using Equation (m) as 0 0 1 0 (n) C= 0 0 0 1 0 1 0 0 D= 0 9.8 The differential equations obtained in Example 2.28 are (o) & y1 0 y 0 & 2 = & y 3 2 x107 & 6 y 4 8 x10
& & c2 2 c11 x k1 + k 2 x m 0 & c1 + c2 + 0 I & c c c 2 + c 2 + k k & & 2 2 2 2 1 1 1 1 2 2 1 1 & & k1 y1 + k 2 y 2 + c1 y1 + c2 y 2 = & & k11 y1 + k 2 2 y 2 c11 y1 + c2 2 y 2
396
Chapter 9
k1 m k 1 1 I
k2 m y1 k 22 y2 I (b)
where
c 2 2 c11 c1,1 c1, 2 c1 + c 2 c = 2 2 2,1 c 2, 2 c 2 2 c11 c11 + c 2 2 k 2 2 k11 k1,1 k1, 2 k1 + k 2 k = 2 2 2,1 k 2, 2 k 2 2 k11 k11 + k 2 2 State variable are defined in the form of z1 = x z2 = & z 3 = x + 1 y1 + 1 y 2 & z = + y + y
4 2 1 2
(c) (d)
Use of Equations (e)-(h) in Equation (b) leads to c1,1 c1, 2 k1,1 z 3 1 1 y1 & & z1 1 1 y1 m m m z y = c c 2, 2 z 4 2 2 y 2 k 2,1 & & 2 2 ,1 2 2 2 I I I k2 c2 c1 k1 & m y1 m m m y1 k k + c c & 2 2 y2 2 2 y2 1 1 1 1 I I I I
The terms proportional to the derivatives of the input functions are eliminated from Equation (i) by choosing c2 c1 1 1 m m (j) = c c 2 2 2 2 1 1 I I
Chapter 9
c1,1 & z1 m z = c & 2 2,1 I c1, 2 k1,1 z3 m m c 2, 2 z 4 k 2,1 I I k1, 2 m z1 + k 2, 2 z 2 I
k1 k 2 c1,1 c1, 2 c1 c2 m m m m m m y1 k11 k 2 2 c 2,1 c 2, 2 c11 c 2 2 y 2 I I I I I I c1, 2 k1,1 k1, 2 z3 m m z1 + 1,1 1, 2 y1 m z k c 2, 2 4 2,1 k 2, 2 z 2 2,1 2, 2 y 2 I I I
(k)
k1
k 2 c1,1 m m c k 2 2 2,1 I I
c1, 2 c1 m m c 2, 2 c11 I I
c2 m c 2 2 I
k1 c1,1c1 c1, 2 c11 m m2 mI c c k c c 1, 2 = 2 1,1 2 2 + 1, 2 2 2 m m mI k c c c c 2,1 = 1 1 2,1 1 2, 2 21 1 mI I I c c k c c 2, 2 = 2 2 2,1 2 + 2, 2 32 2 mI I I The state-space formulation of the equations is 0 1 0 0 0 & z1 z 0 0 0 1 1 z 0 z 0 0 y1 & 2 = k1,1 k1, 2 c1,1 c1, 2 2 + z3 m & m m m z 3 1,1 1, 2 y 2 k 2,1 k 2, 2 c 2,1 c 2, 2 z & z 4 4 2,1 2, 2 I I I I T Assuming the output is the vector [x ] , the output is determined from
1,1 =
(l)
(m)
z1 x 1 0 0 0 z 2 0 0 y1 + = 0 1 0 0 z 3 0 0 y 2 z4
(n)
398
Chapter 9 9.9 The differential equations governing the motion of the system of Figure 9.9 are & & & x1 m 0 & c 0 x1 2k k x1 cy + ky (a) 0 m & + 0 0 x + k k x = 0 & & x2 2 2 Since the right-hand side of Equation (a) involves the derivative of the input, a statespace formulation similar to that of Example 9.6 is required. To this end define state variables as z1 = x1 & z 2 = x1 + y z3 = x2 & z 4 = x 2 + y Substitution into the differential equations of Equation (a) leads to & & & m 0 z 2 y c 0 z 2 y 2k k z1 cy + ky + + = 0 m z y 0 0 z y k k z & 3 0 4 4 & & m 0 z 2 c 0 z 2 2k 0 m z + 0 0 z + k & 4 4
2 k z1 k c y = m k z 3 0
(b)
(c)
(d)
Equations (b) and (d) can be summarized as & 1 0 0 0 z1 0 1 0 0 z1 c 2 0 m 0 0 z 2 k c k 0 z k c & y (t ) 2 + 2 = 0m 0 0 1 0 z 3 0 0 0 1 z 3 & & 0 0 0 m z 4 k 0 k 0 z 4 0
& z1 1 0 0 0 z & 2 = 0 m 0 0 z3 0 0 1 0 & & 0 0 0 m z4 0 & z1 2k z & 2 = m 0 z3 & k & z4 m 1 c m 0 0
1
0 2k 0 k
c k 0 0 0 k
0 z1 1 0 0 0 0 z 2 0 m 0 0 + 1 z 3 0 0 1 0 0 z 4 0 0 0 m
c c2 k y (t ) 0m 0
0 k m 0 k m
c 0 z 1 m 0 z k c 2 2 + y (t ) 1 z 3 m m 2 0 z 4 0 0
(e)
399
Chapter 9 The state matrix and input matrix are determined from Equation (e) as 0 0 1 0 c k 2k 0 m m m A= 0 0 0 1 k k 0 0 m m c m k c2 B= 2 m m 0 0 The output and transition matrices for the model are 1 0 0 0 C= 0 1 0 0 0 D= 0 9.10 The state matrix derived in the solution of Problem 9.1 is 1 0 0 0 A= 0 1 25 16 3 The state transition matrix is To this end
(t ) = L1 (sI A )
(f)
(g)
(h) (i)
(a)
(b)
0 s 1 0 sI A = s 1 (c) 25 16 s + 3 The symbolic capabilities of MATLAB are used to invert the matrix in Equation (b). The MATLAB workspace to accomplish the inversion is
>> syms s >> A=[s -1 0;0 s -1;25 16 s+3] A= [ s, -1, 0] [ 0, s, -1] [ 25, 16, s+3] >> A1=A^-1 400
Chapter 9 A1 = [ (s^2+3*s+16)/(s^3+3*s^2+16*s+25), (s+3)/(s^3+3*s^2+16*s+25), 1/(s^3+3*s^2+16*s+25)] [ -25/(s^3+3*s^2+16*s+25), s*(s+3)/(s^3+3*s^2+16*s+25), s/(s^3+3*s^2+16*s+25)] [ -25*s/(s^3+3*s^2+16*s+25), -(16*s+25)/(s^3+3*s^2+16*s+25), s^2/(s^3+3*s^2+16*s+25)] >> Thus
s 2 + 3s + 16 1 s+3 1 1 (sI A ) = 3 s( s + 3) s 2 25 s + 3s + 16 s + 25 25 16s + 25 s 2 The poles of the transforms in the matrix of Equation (d) are obtained as s1 = 1.8059 (d)
(e) (f)
s2 = 0.5971 + 3.673 j
(g) s3 = 0.5971 3.673 j Partial fraction decompositions of the transforms in Equation (d) are of the form 1 (sI A )1 = 1 C1 + 2 (C 2 s + C 3 ) (h ) s + 1.806 s + 1.19 s + 13.8 MATLAB is used to help determine the partial fraction decompositions. For example the elements in the third row and third column of each matrix are obtained with the following worksheet >> D=[1 3 16 25] D= 1 3 16 25
>> N=[1 0 0] N= 1 0 0
401
Chapter 9 p= -0.5971 + 3.6725i -0.5971 - 3.6725i -1.8059 k= [] >> C3=r(1)*(s-p(2))+r(2)*(s-p(1)) C3 = (7042123938002753/18014398509481984+655836290881283/2251799813685248*i)*(s +2688890991237795/4503599627370496+8269686015043673/2251799813685248*i)+( 7042123938002753/18014398509481984655836290881283/2251799813685248*i)*(s+2688890991237795/45035996273704968269686015043673/2251799813685248*i) >> C3A=vpa(simplify(C3),3) C3A = .782*s-1.67 >> The resulting matrices are 0.0669 0.9261 0.0797 1.673 0.1443 0.1208 C1 = 1.673 3.605 0.2182 0.0739 0.0797 0.0669 C 2 = 1.673 1.14 0.1208 1.673 3.605 0.728 (i)
(j)
1.76 1.05 0.0409 (k) C 3 = 1.023 1.11 0.926 1.023 13.8 1.67 The state transition matrix is obtained by inverting the transforms of Equation (h) leading to (t ) = C 1 e 1.806 t + C 2 e 0.595t cos(3.67t ) + 0.273(C 3 0.595C 2 )e 0.595t sin(3.67t ) (l)
402
Chapter 9 9.11 The state matrix derived in the solution of Problem 9.2 is 5 10 A= (a) 12 4 The state transition matrix is 1 (t ) = L1 (sI A ) (b) To this end s + 10 5 sI A = (c) 4 s + 12 and 5 s + 12 (sI A )1 = 2 1 (d ) 4 s + 10 s + 22 s + 100 The following MATLAB workspace is used to determine the partial fraction decompositions for each element of the matrix in Equation (d)
>> clear >> D=[1 22 100] D= 1 22 100 >> N11=[1 12] N11 = 1 12 >> [r11,p11,k11]=residue(N11,D) r11 = 0.3909 0.6091
k11 = []
403
Chapter 9 >> N12=[5] N12 = 5 >> [r12,p12,k12]=residue(N12,D) r12 = -0.5455 0.5455 p12 = -15.5826 -6.4174 k12 = [] >> N21=[4] N21 = 4 >> [r21,p21,k21]=residue(N21,D) r21 = -0.4364 0.4364 p21 = -15.5826 -6.4174 k21 = [] >> N22=[1 10]
404
Chapter 9 N22 = 1 10 >> [r22,p22,k22]=residue(N22,D) r22 = 0.6091 0.3909 p22 = -15.5826 -6.4174 k22 = [] >> From the MATLAB workspace it is clear that the poles of the transforms are -15.58 and 6.42. Use of the MATLAB workspace leads to 0.391 0.546 0.609 0.546 1 (sI A )1 = 1 (e) 0.436 0.609 + s + 6.42 0.436 0.391 s + 15.58 Inversion of Equation (e) leads to 0.391 0.546 15.58t 0.609 0.546 6.42t e (f) + (t ) = e .609 0.436 0.391 0.436 9.12 The state matrix obtained during the solution of Problem 9.5 is 8.33 0 A= 5.56 0 The state transition matrix is defined as 1 (t ) = L1 (sI A ) To this end s + 0.833 0 sI A = s 5.56
(a)
(b)
(sI A )1 =
(c)
405
Chapter 9
0 1
(d)
9.13 The state transition matrix is defined as 1 (t ) = L1 (sI A ) The state matrix derived during the solution of Problem 9.9 is 0 0 1 0 c k 2k 0 m m m A= 0 0 0 1 k k 0 0 m m Thus 1 0 s c k 2k m s+ m m sI A = 0 0 s k k 0 m m MATLABs symbolic capabilities are used to determined (s I A ) 1
(a)
(b)
0 0 1 s
(c)
where D( s ) = m 2 s 4 + cms 3 3mks 2 cks + k 2 It is not possible to proceed further without numerical values for system parameters. 9.14 The state transition matrix derived during the solution of Problem 9.10 is (t ) = C 1 e 1.806 t + C 2 e 0.595 t cos( 3 .67 t ) + 0 .273 (C 3 0 .595 C 2 )e 0.595 t sin( 3 .67 t ) where 0.0669 0.9261 0.0797 1.673 0.1443 0.1208 C1 = 1.673 3.605 0.2182 (e)
(a)
(b)
(c)
406
Chapter 9 1.76 1.05 0.0409 C 3 = 1.023 1.11 0.926 1.023 13.8 1.67 The initial condition vector is given as 1 y (0) = 0 0 The free response is obtained using Equation (9.45) as y (t ) = (t ) y (0) Substitution of Equations (a)-(e) into Equation (f) leads to 0.0669 1 0.9261 0.0797 1.673 0.1443 0.1208 0 e 1.806t (t ) = 1.673 3.605 0.2182 0
0.0739 0.0797 0.0669 1 + 1.673 0.1208 0 e 0.595t cos(3.67t ) 1.14 1.673 3.605 0.728 0 1.76 1.05 0.0409 0.0739 0.0797 0.0669 0.595 1.673 + 0.273 1.023 1.11 0.926 1.14 0.1208 1.023 13.8 1.67 1.673 3.605 0.728 1 0 e 0.595t sin(3.67t ) 0 Simplification of Equation (g) leads to (g)
(d)
(e)
(f )
x(t ) 0.926 0.0739 0.469 x(t ) = 1.673 e 1.81t + 1.673 e 0.595 cos(3.67t ) + 0.551 e =0.595t sin(3.67t ) & & t ) 1.673 1.673 0.551 & x( 9.15 The state transition matrix derived in the solution to Problem 9.11 is 0.391 0.546 15.58t 0.609 0.546 6.42t e + (t ) = e .609 0.436 0.391 0.436 The initial condition vector is given as 0 y (0) = 0.5 The free response is obtained using Equation (9.45) y (t ) = (t ) y (0)
(h)
(a)
(b)
(c)
407
Chapter 9 Substitution of Equations (a) and (b) into Equation (c) leads to 0.391 0.546 0 15.58t 0.609 0.546 0 6.42t y (t ) = e + e .609 0.5 0.436 0.391 0.5 0.436 0.273 15.58t 0.273 6.42t + = e e 0.196 0.305 9.16 Application of Kirchoffs laws to the circuit lead to the following equations t di1 di 2 1 q ( 0) L L + Ri1 + i1 dt = dt dt C0 C
L di 2 di L 1 + Ri2 = 0 dt dt
(d)
(a) (b)
y 3 = i2 Equations (a) and (b) are written in terms of the state variables as 1 q(0) & & Ly 2 Ly 3 + Ry 2 + y1 = C C 1 R q(0) & & y 2 y3 = y1 y 2 LC L LC & & Ly 3 Ly 2 + Ry 3 = 0 & & y3 y 2 =
(d)
R y3 (e) L Equations (c)-(e) are summarized in matrix form as 0 1 0 y1 0 & 0 y1 1 0 1 1 R 0 1 1 y = & 0 y 2 + q (0) (f) 2 LC L LC 0 1 1 y 3 R y3 0 & 0 0 L However the formulation of Equation (f) does not work as the matrix on the left hand side of Equation (f) is singular. This is to be expected as there are only two energy storage devices in the system, thus only two states are available. Notice the subtraction of Equation (e) from Equation (d) 1 R q ( 0) & & 2( y 3 y 2 ) = (g) y1 + ( y 3 y 2 ) LC L LC Equation (g) suggests that the state variables are y1 = q1 y 2 = i2 i1 (h)
408
Chapter 9 Note that Equation (d) can be formulated in terms of the state variables of Equation (h) as q ( 0) 1 R & & y2 = (i) y1 y1 LC L LC Thus the state-space formulation of the system using the state variables of Equation (h) is 1 0 y 1 q(0) R y & 1 1 1 LC L + 1 (j) y = R y 2 LC 1 0 1 & 2 2 2 LC 2 L Substituting given values, & 0 y1 1 333.3 1 y1 1.67 x10 7 = + 0 1 y 2 8.33 x10 6 166.7 y 2 0.5 &
& y1 7.51x10 4 0.500 y1 0.0015 (k ) + y = 6 166.7 y 2 0.5 2 & 8.33x10 where = q (0) /( LC ) is taken as the input. The state matrix and input matrix are defined in Equation (k). The output matrix and transition matrix are C = [1 0] D = [0] (l) (b) The state transition matrix is
(t ) = L1 (sI A )
(m)
To this end s 7.51x10 4 0.5 sI A = 6 s 166.7 8.33x10 s 166.7 0.5 [sI A]1 = 1 8.33x10 6 s 7.51x10 4 D( s ) D( s ) = s 2 7.53x10 5 + 8.34 x10 6 Inversion of Equation (n) leads to 0.6669 7.51x104 t 7.40 x10 4 0.9993 (t ) = + e 4 111.1 1.49 x10 111.1 The system response is
y = (t )Bu( )d
0 t
(n) (o)
0.6669 0.0111t e 1
(p)
y1 = 0.332e 7.51x10
0
( t )
0.332e 0.0111( t ) d
(q)
9.17 The differential equations governing the motion of the system are & x1 0 & 5 x10 7 5 x10 7 x1 0 3000 = 0 & + & 3000 x 2 5 x10 7 5 x10 7 x 2 0
(a)
409
Chapter 9 When premultiplied by the inverse of the mass matrix, Equation (a) becomes & x1 1.67 x1 & 4 1.67 & + 10 1.67 1.67 x & 2 x2 The state vector is defined as x1 x y = 2 x1 & & x2 The state-space formulation of Equation (c) is & 0 0 1 0 y1 y1 y & 0 0 0 1 y2 2 = y 3 1 .67 x10 4 1 .67 x10 4 & 0 0 y3 4 & 1 .67 x10 4 0 0 y 4 y 4 1 .67 x10 The state matrix for the system is 0 0 A= 1.67 x10 4 4 1.67 x10 The state transition matrix is 0 1 0 0
1
(b)
(c)
(d)
1 0 0 0
(e)
(f)
>> c=1.67e4 c= 16700 >> A=[0 0 1 0;0 0 0 1;-c c 0 0;c -c 0 0] A= 0 0 -16700 16700 0 1 0 0 16700 -16700 0 1 0 0 0 0
410
Chapter 9
I= 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1
>> B=(s*I-A)^-1 B= [ 1/s*(s^2+16700)/(s^2+33400), 16700/s/(s^2+33400), (s^2+16700)/s^2/(s^2+33400), 16700/s^2/(s^2+33400)] [ 16700/s/(s^2+33400), 1/s*(s^2+16700)/(s^2+33400), 16700/s^2/(s^2+33400), (s^2+16700)/s^2/(s^2+33400)] [ -16700/(s^2+33400), 16700/(s^2+33400), 1/s*(s^2+16700)/(s^2+33400), 16700/s/(s^2+33400)] [ 16700/(s^2+33400), -16700/(s^2+33400), 16700/s/(s^2+33400), 1/s*(s^2+16700)/(s^2+33400)] >> PHI=ilaplace(B) PHI = [ 1/2+1/2*cos(10*334^(1/2)*t), -1/2*cos(10*334^(1/2)*t)+1/2, 1/6680*334^(1/2)*sin(10*334^(1/2)*t)+1/2*t, 1/2*t1/6680*334^(1/2)*sin(10*334^(1/2)*t)] -1/2*cos(10*334^(1/2)*t)+1/2, 1/2+1/2*cos(10*334^(1/2)*t), 1/2*t1/6680*334^(1/2)*sin(10*334^(1/2)*t), 1/6680*334^(1/2)*sin(10*334^(1/2)*t)+1/2*t] -5*334^(1/2)*sin(10*334^(1/2)*t), 5*334^(1/2)*sin(10*334^(1/2)*t), 1/2+1/2*cos(10*334^(1/2)*t), -1/2*cos(10*334^(1/2)*t)+1/2] 5*334^(1/2)*sin(10*334^(1/2)*t), -5*334^(1/2)*sin(10*334^(1/2)*t), -1/2*cos(10*334^(1/2)*t)+1/2, 1/2+1/2*cos(10*334^(1/2)*t)]
[ [
411
Chapter 9 2*sin(182.75666882497065256033071913464*t)+.500000000000000000000000 00000000*t, .50000000000000000000000000000000*t.27358782758229139604839928014167e2*sin(182.75666882497065256033071913464*t)] [ .50000000000000000000000000000000*cos(182.7566688249706525603307191 3464*t)+.50000000000000000000000000000000, .50000000000000000000000000000000+.5000000000000000000000000000000 0*cos(182.75666882497065256033071913464*t), .50000000000000000000000000000000*t.27358782758229139604839928014167e2*sin(182.75666882497065256033071913464*t), .27358782758229139604839928014167e2*sin(182.75666882497065256033071913464*t)+.500000000000000000000000 00000000*t] 91.378334412485326280165359567320*sin(182.75666882497065256033071913 464*t), 91.378334412485326280165359567320*sin(182.75666882497065256033071913 464*t), .50000000000000000000000000000000+.5000000000000000000000000000000 0*cos(182.75666882497065256033071913464*t), .50000000000000000000000000000000*cos(182.7566688249706525603307191 3464*t)+.50000000000000000000000000000000] 91.378334412485326280165359567320*sin(182.75666882497065256033071913 464*t), 91.378334412485326280165359567320*sin(182.75666882497065256033071913 464*t), .50000000000000000000000000000000*cos(182.7566688249706525603307191 3464*t)+.50000000000000000000000000000000, .50000000000000000000000000000000+.5000000000000000000000000000000 0*cos(182.75666882497065256033071913464*t)] >> PHI1=vpa(PHI,3) PHI1 = [ [ [ [ .500+.500*cos(183.*t), -.500*cos(183.*t)+.500, .274e-2*sin(183.*t)+.500*t, .500*t-.274e-2*sin(183.*t)] -.500*cos(183.*t)+.500, .500+.500*cos(183.*t), .500*t-.274e-2*sin(183.*t), .274e-2*sin(183.*t)+.500*t] -91.5*sin(183.*t), 91.5*sin(183.*t), .500+.500*cos(183.*t), .500*cos(183.*t)+.500] 91.5*sin(183.*t), -91.5*sin(183.*t), -.500*cos(183.*t)+.500, .500+.500*cos(183.*t)]
412
Chapter 9
>> 0.5 + 0.5 cos(183.3t ) 0.5 0.5 cos(183.3t ) 0.5t + 0.0274 sin(183.3t ) 0.5t 0.0274 sin(183.3t ) 0.5 0.5 cos(183.3t ) 0.5 + 0.5 cos(183.3t ) 0.5t 0.0274 sin(183.3t ) 0.5t + 0.0274 sin(183.3t ) = 91.5 sin(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) 0.5 0.5 cos(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) The initial condition vector is 0 0 y0 = 8 0 The system response is determined using Equation (9.34) y = y 0
(h)
0.5 + 0.5 cos(183.3t ) 0.5 0.5 cos(183.3t ) 0.5t + 0.0274 sin(183.3t ) 0.5t 0.0274 sin(183.3t ) 0.5 0.5 cos(183.3t ) 0.5 + 0.5 cos(183.3t ) 0.5t 0.0274 sin(183.3t ) 0.5t + 0.0274 sin(183.3t ) = 91.5 sin(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) 0.5 0.5 cos(183.3t ) 0.5 + 0.5 cos(183.3t ) 91.5 sin(183.3t ) 0 0 X 8 0 4t + 0.2192 sin(183.3t ) 4t 0.2192 sin(183.3t ) = 4 + 4 cos(183.3t ) 4 4 cos(183.3t ) Thus x1 (t ) = 4t + 0.2192 sin(183.3t ) x 2 (t ) = 4t 0.2192 sin(183.3t ) (j) (k)
(i)
9.18 The differential equations governing the system after the coupling engages the bumper are & & x1 0 & 1.2 x10 4 0 x1 5.2 x10 7 5 x10 7 x1 0 3000 (a) = & + 0 & + & 3000 x 2 0 0 x 2 5 x10 7 5 x10 7 x 2 0 Equation (a) may be rewritten as
413
Chapter 9
(b)
(c )
0 1 (d) 0 0 Equation (d) is valid after the bumper is engaged, after x1 = 10 m . For x1 < 10 m the appropriate state-space model is that of Equation (d) of the solution of Problem 9.17 & 0 0 1 0 y1 y1 y & 0 0 0 1 y 2 2 = (e ) y 3 1.67 x10 4 1.67 x10 4 0 0 y3 & 4 & 1.67 x10 4 0 0 y 4 y 4 1.67 x10 Equations (d) and (e) are integrated using the MATLAB program ode45. The initial conditions as formulated in Equation (h) of the solution of Problem 9.17 are 0 0 y0 = (f ) 8 0 The program applies the initial conditions to Equation (e) and solves Equation (e) until x1 = 10 m . After this time it solves Equation (d). The MATLAB m file for this simulation is shown below 1 0 4 0 % Problem 9.18 clear % Specify initial conditions y0=[0 0 8 0]; % Specify final time for integration tf=3; % Call ode45 [t,y]=ode45(@bumper,[0 tf],y0) % Plots results figure plot(t,y(:,1),'-',t,y(:,2),'.') 414
Chapter 9 xlabel('t (s)') ylabel('x (m)') title('Dispalcements of railroad cars of Problem 9.18') legend('x_1','x_2') figure plot(t,y(:,3),'-',t,y(:,4),'.') xlabel('t (s)') ylabel('v (m/s)') title('Velocities of railroad cars of Problem 9.18') legend('v_1','v_2') The required function bumper is % Function for Problem 9.18 function dy=bumper(t,y) if y(1)<10 dy1=y(3); dy2=y(4); dy3=-1.67e4*y(1)+1.67e4*y(2); dy4=1.67e4*y(1)-1.67e4*y(2); else dy1=y(3); dy2=y(4); dy3=-1.73e4*y(1)+1.67e4*y(2)-4*y(3); dy4=1.67e4*y(1)-1.67e4*y(2); end dy=[dy1;dy2;dy3;dy4] The generated plots are
415
Chapter 9
The details are difficult to determine in these plots. The following is a Figure acquired using the zoom in feature on the plot of displacements
416
Chapter 9
9.19 The differential equations governing the motion of the system after the cars engage the bumper are
1 & & 3000& + 5 x10 7 x1 5 x10 7 x 2 + 2 x10 6 x1 1.2 x10 4 x1 .6 x1
& 3000& 5 x10 7 x1 + 5 x10 7 x 2 = 0 x2 Dividing by the mass and rearranging, Equations (a) and (b) are written as 1 & = 1.67 x10 4 x1 + 1.67 x10 4 x 2 6.67 x10 2 x1 4 x1 .6 & & x1 & = 1.67 x10 4 x1 1.67 x10 4 x 2 & x2 Define state variable by
x1 x y = 2 x1 & & x2 Equations (c) and (d) are written using state variables as & y1 = y 3 & y2 = y4 & y 3 = 1.67 x10 y1 + 1.67 x10 y 2 6.67 x10 y1 4 y
4 4 2 4 4 1.6 3
(e )
& (i) y 4 = 1.67 x10 y1 1.67 x10 y 2 Equations (f)-(i) are valid after the bumper is engaged, after x1 = 10 m . For x1 < 10 m the appropriate state-space model is that of Equation (d) of the solution of Problem 9.17
417
Chapter 9 & 0 0 1 0 y1 y1 y & 0 0 0 1 y 2 2 = (e ) y 3 1.67 x10 4 1.67 x10 4 0 0 y3 & 4 & 1.67 x10 4 0 0 y 4 y 4 1.67 x10 Equations (f)-(i) and (e) are integrated using the MATLAB program ode45. The initial conditions as formulated in Equation (h) of the solution of Problem 9.17 are 0 0 y0 = (f ) 8 0 The program applies the initial conditions to Equation (e) and solves Equation (e) until x1 = 10 m . After this time it solves Equation (d). The MATLAB m file for this simulation is shown below % Problem 9.19 clear % Specify initial conditions y0=[0 0 8 0]; % Specify final time for integration tf=3; % Call ode45 [t,y]=ode45(@bumpern,[0 tf],y0) % Plots results figure plot(t,y(:,1),'-',t,y(:,2),'.') xlabel('t (s)') ylabel('x (m)') title('Dispalcements of railroad cars of Problem 9.19') legend('x_1','x_2') figure plot(t,y(:,3),'-',t,y(:,4),'.') xlabel('t (s)') ylabel('v (m/s)') title('Velocities of railroad cars of Problem 9.19') legend('v_1','v_2') The required function bumpern is % Function for Problem 9.19 function dy=bumpern(t,y) if y(1)<10 dy1=y(3); dy2=y(4); dy3=-1.67e4*y(1)+1.67e4*y(2); 418
419
Chapter 9
420
Chapter 9
9.20 The state transition matrix derived in the solution of Problem 9.10 is (t ) = C 1 e 1.806 t + C 2 e 0.595 t cos( 3 .67 t ) + 0 .273 (C 3 0 .595 C 2 )e 0.595 t sin( 3 .67 t ) (a) where 0.0669 0.9261 0.0797 1.673 0.1443 0.1208 C1 = (b) 1.673 3.605 0.2182 0.0739 0.0797 0.0669 C 2 = 1.673 1.14 0.1208 1.673 3.605 0.728 1.76 1.05 0.0409 C 3 = 1.023 1.11 0.926 1.023 13.8 1.67 The input vector determined in the solution of Problem 9.1 is 0 B = 0 1 (c)
(d)
(e)
The input is given as u = [F (t )] = [0.5 sin(20t )] . The response is determined using Equation (9.54)
y = (t )Bu( )d
0 t
0 0.273(C 3 0.595C 2 )e sin[3.67(t )]}0[0.5 sin(20 )]d (f) 1 Substitution of Equations (b), (c), and (d) into Equation (f) leads to 0.669 0.0669 t t 0.1208 + 0.5 e =0.59 ( t ) cos[3.67(t )] sin( 20 )d 0.1208 =1.806 ( t ) sin( 20 )d y = 0.5 e 0 0 0.2182 0.728
0.595 ( t )
+ 0 .5 e
0
0.59 ( t )
(g)
421
Chapter 9
0.669 sin( 20 )d 0.1208 0.2182
y = 0.5e
1.80 t
e
0
1.80
+ 0.5e
0.59 t
0.0669 t 0.0220 cos(3.67t ) 0.0148 sin(3.67t ) e 0.59 cos(3.67 ) sin( 20 )d 0.1208 0.728 0 0.5742 0.0220 t 0.0669 cos(3.67t ) + 0.1208 sin(3.67t ) e 0.59 sin(3.67 ) sin( 20 )d 0.0148 0.5742 0 0.728
+ 0.5e
0.59 t
Integration leads to
422
Chapter 9
0.669 y = 5.17 x10 0.1208 1.80 sin( 20t ) 20 cos( 20t ) + 20e 1.80 t 0.2182 0.0669 0.0220 cos(3.67t ) 0.0148 sin(3.67t ) 4 + 4.46 x10 0.1208 0.728 0.5742 0.59 t 0.59 sin( 23.67t ) 23.67 cos( 23.67t ) + 23.67 e }
5
0.0669 0.0220 cos(3.67t ) 0.0148 sin(3.67t ) + 9.26 x10 0.1208 0.728 0.5742
4
0.0669 0.0220 cos(3.67t ) 0.0148 sin(3.67t ) {0.59 sin( 23.67t ) + 4.46 x10 0.1208 0.728 0.5742 + 23.67 cos( 23.67t ) 23.67 e 0.59 t + 0.0669 0.0220 cos(3.67t ) 0.0148 sin(3.67t ) 9.26 x10 0.1208 0.728 0.5742
4 0.59 t
(j)
9.21 The state transition matrix derived in the solution of Problem 9.10 is (t ) = C 1 e 1.806t + C 2 e 0.595t cos(3.67t ) + 0.273(C 3 0.595C 2 )e 0.595t sin(3.67t ) (a) where 0.0669 0.9261 0.0797 1.673 0.1443 0.1208 C1 = (b) 1.673 3.605 0.2182 0.0739 0.0797 0.0669 C 2 = 1.673 1.14 0.1208 1.673 3.605 0.728 (c)
423
Chapter 9 1.76 1.05 0.0409 C 3 = 1.023 1.11 0.926 1.023 13.8 1.67 The input vector determined in the solution of Problem 9.1 is 0 B = 0 1
(d)
(e)
The input is given as u = [F (t )] = [0.5(u (t ) u (t 3))] . The response is determined using Equation (9.54)
y = (t )Bu( )d
0 t
0 0.273(C 3 0.595C 2 )e sin[3.67(t )]}0[0.5(u ( ) u ( 3)]d 1 Substitution of Equations (b), (c), and (d) into Equation (f) leads to 0.669 t y = 0.5 e =1.806(t ) [u( ) u( 3)]d 0.1208 0 0.2182
0.595 ( t )
(f)
+ 0.5 e
0 t
=0.59(t )
0.0669 cos[3.67(t )][u( ) u( 3)]d 0.1208 0.728 0.0220 sin[3.67(t )][u( ) u( 3)]d 0.0148 0.5742 (g)
+ 0.5 e
0
0.59(t )
Note that
f ( )[u ( ) u ( 3)]d = f ( )d u (t 3) f ( )d
0 3
(h )
424
Chapter 9 Use of Equation (h) to evaluate Equation (g) leads to 0 . 669 y = 0 . 2688 0 . 1208 1 e 1 .860 t (1 e 1 .860 ( t 3 ) )u ( t 3 ) 0 . 2182
0 . 0669 + 0 . 0362 0 . 1208 { 0 .59 t [ 0 . 59 cos( 3 . 67 t ) + 3 . 67 sin( 3 . 67 t ) ] + 0 . 59 + e 0 . 728 0 . 59 t [ 0 .59 cos( 3 .67 t ) + 3 .67 sin( 3 .67 t ) ] u ( t 3 )[ e e 0 .59 ( t 3 ) [ 0 . 59 cos( 3 . 67 ( t 3 )) + 3 . 67 sin( 3 . 67 ( t 3 ))]} 0 . 0220 + 0 . 0362 0 . 0148 { 0 .59 t [ 0 . 59 sin( 3 . 67 t ) + 3 . 67 cos( 3 . 67 t ) ] 3 . 67 + e 0 . 5742 0 . 59 t [ 0 .59 sin( 3 .67 t ) + 3 .67 cos( 3 .67 t ) ] u ( t 3 )[ e e 0 .59 ( t 3 ) [ 0 . 59 sin( 3 . 67 ( t 3 )) + 3 . 67 cos( 3 . 67 ( t 3 ))]} (i)
9.22 The state transition matrix derived in the solution of Problem 9.11 is 0.391 0.546 15.58t 0.609 0.546 6.42t (a) (t ) = + e e .609 0.436 0.436 0.391 The input matrix derived in the solution of Problem 9.2 is 5 0 (b) B= 0 4 The input vector for this problem is given as 0.2 u(t ) = e 0.2t (c) 0.5 The convolution integral solution for the forced response is obtained using Equation (9.54) assuming initial conditions of zero
y (t ) = (t )Bu( )d
0
(d)
Substitution of Equations (a)-(c) into Equation (d) leads to t 0.391 0.546 15.58( t ) 0.609 0.546 6.42( t ) 5 0 0.2 0.2 + y (t ) = e d e e 0.436 .609 0.436 0.391 0 0 4 0.5
t 0.391 0.546 15.58(t ) 0.609 0.546 6.42 ( t ) 1 0.2 = + e d e e 0.436 .609 0.436 0.391 0 2
=e
15.58t
(e)
425
Chapter 9 Simplification of Equation (e) leads to .0456 0.2t 0.274 0.2t e 15.58t ) + e 6.42t ) y (t ) = (e (e .0508 0.196 0.229 0.2t 0.0456 15.58t 0.274 6.42t = e e e 0.247 0.0508 0.196 9.23 The state transition matrix obtained during the solution of Problem 9.12 is 0 e 0.833t = 0.833t 1 1 0.675 e The input matrix obtained during the solution of Problem (9.5) is 0 16.67 B= 5.56 0 The input for this problem is 120 sin(800t ) u= 0 The convolution integral solution for the circuit response is (f)
(a)
(b)
(c)
y = (t )Bu( )d
0 t 0 120 sin(800 ) e 0.833(t ) 0 16.67 = d 5.56 0 0.675(e 0.833( t ) 1) 1 0 0 t e 0.833(t ) 0 2000 sin(800 ) = d t ) 0 0.675(e 0.833( 1) 1 0 t 2000e 0.833( t ) sin(800 ) = d 0.833( t ) 1)sin(800 ) 0 1350(e
(d)
Evaluation of the integrals in Equation (d) leads to y1 (t ) = 0.0260 sin(800t ) + 2.5 cos(800t ) 2.5e 0.833t
(e) (f)
y 2 (t ) = 0.00176 sin(800t ) + 1.6875 1.6875e 0.833t The output relationship obtained from the solution of Problem 9.5 is y v [v B ] = [0 1] 1 + [0 1] 1 v y 2 2 Thus v B (t ) = 0.00176 sin(800t ) + 1.6875 1.6875e 0.833t
(g)
(h)
426
Chapter 9 9.24 The differential equations modeling the system of Figure 9.24 are dh 1.5 1 + 0.2h1 0.2h2 = qi dt dh 1.2 2 0.2h1 + 0.325h2 0.125h3 = 0 dt dh 2.5 3 0.125h2 + 0.525h3 = 0 dt Defining the state variables as y1 = h1 y 2 = h2 y 3 = h3 The state equations become & h1 .1333 .1333 0 h1 0.667 & .2708 .1042 h2 + 0 qi (t ) h2 = .167 & h3 0 0.050 .210 h3 0 The state-transition matrix is where 0 s + .1333 .1333 .167 sI A = s + .2708 .1042 0.050 s + .210 0 Inversion of Equation (g) leads to (t ) = C 1 e 0.391t + C 2 e 0.196 t + C 3 e 0.0289 t where 0.238 0.341 0.196 C1 = 0.428 0.661 0.380 0.118 0.183 0.125 0.124 C 2 = 0.0719 0.256 0.637 C 3 = 0.500 0.138 0.0573 0.425 0.0271 0.201 0.0956 0.825 (g) = L1 {( sI A) 1 } (d)
(e)
(f )
(h )
(i)
(j)
(k)
427
0.238 0.124 0.637 = 0.2 0.428 e 0.391( t ) + 0.0719 e 0.196( t ) + 0.500 e 0.0289 (t ) d 0 0.256 0.138 0.118 0.122 0.127 4.408 0.229 1 e 0.391t + 0.0734[1 e 0.196t ] + 3.460[1 e 0.0289t ] = 0.0604 0.0261 0.955
t
(l)
9.25 The mathematical model for the system is & & 10 &+ 200 x + 4 x10 5 x = 100 sin(200t ) x
&+ 20 x + 4 x10 4 x = 10 sin(200t ) & & x Define state-space variables as y1 = x & y2 = x The state-space formulation for the system is & 1 y1 0 y1 0 y = 4 x10 4 20 y + 1 F (t ) 2 2 & The state transition matrix is calculated from 1 s sI A = 4 4 x10 s + 20
(a )
(b)
(c)
(d) 1 s (e)
(sI A )1 =
Inversion of Equation (e) leads to =
1 s + 20 s + 4 x10 4
2
s + 20 4 x10 4
e 10t cos(199.75t ) + 0.0501 10t sin(199.75t ) e 0.0050e 10t sin(199.75t ) (f) 10t 10t 10t 200.25e sin(199.75t ) e cos(199.75t ) 0.0501 e sin(199.75t ) The system response is determined from
y = (t )Bu( )d
0 t
(g)
428
Chapter 9
t
(h)
+ 1.56 x10 6 sin(199.75t ) 10 sin(399.75t ) (399.75) cos(399.75t ) + 399.75e 10t + 0.0025 cos(199.75t )[0.25 sin(0.25t ) + 10 cos(0.25t ) 0.25e 10t ] 1.56 x10 6 cos(199.75t )[399.75 sin(399.75t ) + 10 cos(399.75t ) 10e 10t ]
]
(i)
9.26 If the mass of the system varies with time, the appropriate form of Newtons law is d & (a) F = dt (mx) Application of Equation (a) to the particle leads to & & & m(t ) &+ ( m + c ) x + kx = F (t ) x (b) where m(t ) = 10e 0.1t (c)
& m = e 0.1t
Defining state-space variables of y1 = x & y2 = x the state-space formulation for the system is & 0 1 0 y1 y1 y = 4 x10 4 e 0.1t 0.1 20e 0.1t y + 10e 0.1t sin(100t ) 2 2 & A MATLAB M file which numerically obtains the solution of Equation (f) is
% Problem 9.26 % Initial conditions clear y0=[0 0]; tf=1; % Nuerical integration using ode45 [t,y]=ode45(@xxy,[0 tf],y0); plot(t,y(:,1)) xlabel('t') ylabel('x') title('Response of mechanical system losing mass')
(d)
(e)
(f)
429
Chapter 9 Execution of the file requires a user defined program of derivatives such as
% Functions defined for Problem 9.26 function dy=xxy(t,y) dy1=y(2); dy2=-4e4*exp(0.1*t)*y(1)+(0.1-20*exp(0.1*t))*y(2)... +10*exp(0.1*t)*sin(100*t); dy=[dy1;dy2];
0.1
0.2
0.3
0.4
0.5 t
0.6
0.7
0.8
0.9
9.27 The state-space formulation for the system is & 1 z1 z1 0 z = 640 20 z 2 2 & The problem is modeled as that of the free response due to initial condition of z1 (0) = 0.0128
(a)
(b)
z 2 (0) = 0 (c) A MATLAB program using ode45 to numerically integrate the differential equation follows
% Problem 9.27 % Initial conditions clear z0=[0.0128 0]; tf=1; % Nuerical integration using ode45 [t,z]=ode45(@zxy,[0 tf],z0); plot(t,z(:,1)) xlabel('t') ylabel('x') title('Response of simplified syspension system')
430
Chapter 9
The required file zxy.m which supplies the vector of derivatives for use in ode45 is
% Functions defined for Problem 9.27 function dy=zxy(t,y) dy1=y(2); dy2=-640*y(1)-20*y(2); dy=[dy1;dy2];
0.1
0.2
0.3
0.4
0.5 t
0.6
0.7
0.8
0.9
& y1 = y 2 & y 2 = 100 sin( y1 ) The initial conditions are of the form y1 (0) = 0
y 2 (0) = 0 (f) The MATLAB program to numerically determine the response of the nonlinear pendulum is
% Problem 9.28 % Initial conditions clear disp('Please input initial angular displacement in degrees') theta0=input('>>') theta1=theta0*180/pi;
431
Chapter 9
z0=[theta1 0]; tf=4; % Nuerical integration using ode45 [t,z]=ode45(@theta,[0 tf],z0); plot(t,z(:,1)) xlabel('t') ylabel('\theta') str=['Response of nonlinear pendulum with \theta_0='... ,num2str(theta0),' degrees'] title(str) figure plot(z(:,1),z(:,2)) xlabel('\theta') ylabel('\omega') str=['State plane for nonlinear pendulum with \theta_0='... ,num2str(theta0),' degrees']; title(str)
572.5
572
0.5
1.5
2 t
2.5
3.5
432
Chapter 9
State plane for nonlinear pendulum with 0=10 degrees 15
10
-5
-10
-15 570.5
571
571.5
572
572.5
573
Response of nonlinear pendulum with 0=45 degrees 2578.5 2578 2577.5 2577 2576.5 2576 2575.5 2575 2574.5 2574 2573.5
0.5
1.5
2 t
2.5
3.5
433
Chapter 9
State plane for nonlinear pendulum with 0=45 degrees 20 15 10 5 0 -5 -10 -15 -20 2573.5 2574 2574.5 2575 2575.5 2576 2576.5 2577 2577.5 2578 2578.5
It is clear from running the file that the period decreases as the initial angle increases. 9.29 A MATLAB program which uses the state-space formulation to model the CSTR of Example 4.17 is % Problem 9.29 % CSTR of Example 4.17 % Numerical simulation clear y0=[0 0 0] tf=44 [t,y]=ode45(@lin,[0 tf],y0) plot(t,y(:,1)) figure plot(t,y(:,2)) figure plot(t,y(:,3))
% Function for Problem 9.29 function dy=lin(t,y) V=1e-3 q=1e-6; alpha=1e-3; lambda=1e-3; R=480; Ts=25; Q=1; rho=760;
434
Chapter 9 cp=.06; E=5; B=exp(-E/(R*Ts)); cai=1e-4; dy1=-1/V*(q*y(2)+alpha*V*B*(y(1)+E/R/Ts^2*y(3)*cai)); dy2=-1/V*(-alpha*V*B*(y(1)+E/R/Ts^2*cai*y(3))+q*y(2)); dy3=-1/(rho*V*cp)*(rho*q*cp*y(3)-lambda*alpha*V*B*(y(1)+E/R/Ts^2*y(3)*cai))+Q; dy=[dy1;dy2;dy3]; 9.30 The state matrix and the input vector obtained in the solution of Problem 9.1 are 1 0 0 0 0 1 (a) A= 25 16 3
0 B = 0 (b) 1 Noting that the state variables are defined such that y1 = x the transfer function X (s) G (s) = is obtained from F (s) V1 (c) G ( s) = sI A where 0 1 0
V1 = 0 s 1 16 =1 s sI A = 0 1 s 0 1 s+3 (e)
(f)
1 s+3 (d)
25 16
= s 2 ( s + 3) + 25 + 16 s = s 3 + 3s 2 + 16 s + 25
Thus
G ( s) = 1 s + 3s + 16 s + 25
3 2
435
Chapter 9 9.31 The state-space formulation obtained during the solution of Example 9.8 is 0 & y1 0 y k + k & 2 2 = 1 m1 y3 & k2 & y4 m1 The transfer function G1,1 ( s ) = 0 0 k2 m2 k 2 m2 1 0 c1 + c 2 m1 c2 m1 0 0 y1 0 1 c2 y2 1 + m2 y 3 m1 c 2 y4 0 m2 0 0 F (t ) 0 1 F2 (t ) 1 m2
(a)
Y1 ( s) is determined as F1 ( s )
G1,1 ( s ) = V1,1 sI A (b)
where
V1,1
0 0 1 = m1 0
0 s k 2 m2 k2 m2
1 0 c1 + c 2 s+ m1 c 2 m2 0 1 c s+ 2 m2
0 1 c 2 m2 c s+ 2 m2
0 1 s = m1 k 2 m2 = 1 m1
1 0 c 2 m2
k2 c + s s + 2 m2 m2
(c)
436
Chapter 9
s 0 k1 + k 2 sI A = m1 k 2 m1 s k2 =s m2 k2 m2 0 s k 2 m2 k2 m2 1 0 c1 + c 2 s+ m1 c 2 m1 0 1 c 2 m2 c s+ 2 m2 s k 2 m2 k2 m2 1 c 2 m2 c s+ 2 m2 c2 s 2 m1 m2 (d)
0 c1 + c 2 s+ m1 c2 m1
1 0 c2 k1 + k 2 m2 m1 c2 k s+ 2 m2 m1
c + c2 = s s s + 1 m1
c s + 2 m2
k 2 c2 k2 c1 + c 2 m m + m s + m 1 2 2 1
c s s + 2 m2
V2,1 sI A
(e)
V2,1
k2 c s 2 m2 m1
437
Chapter 9 9.32 The transfer function of a second-order system is 2s + 3 G (s) = 2 s + 6 s + 25 A state matrix for a state-space formulation of the system is 1 0 A= 25 6
T
(a)
(b)
b1 = b2 = 2
3 = 0.12 25
(d) (e)
Thus the state-space formulation is & 1 y1 0.12 y1 0 = y f (t ) + 2 & 25 6 y 2 2 9.33 The transfer function of a fourth-order system is 3s 2 + 2 s + 5 G ( s) = 4 s + 3s 3 + 10s 2 + 20s + 50 A state matrix that can be used in a state-space model for the system is 1 0 0 0 0 0 1 0 A= 0 0 0 1 50 20 10 3 T The input vector B = [b1 b2 b3 b4 ] is obtained by requiring
V4 = 3s 2 + 2 s + 5
s 1 0 0 s 1 0 0 s 50 20 10 b1 b2 = 3s 2 + 2 s + 5 b3 b4
(f)
(a)
(b)
(c)
438
Chapter 9 The determinant on the left-hand side of Equation (c) is evaluated by expanding about its first column 1 0 b1 s 1 b2 s 0 s b3 50 s 1 b 2 = 3s 2 + 2s + 5 20 10 b4 0 s b3
s b4 s 2 20b3 20sb2 10sb3 50 b3 + b1 s 2 + b2 s = 3s 3 + 2 s + 5 b4 s 3 + (20b2 10b3 50b1 ) s 2 + (20b3 50b2 ) s 50b3 = 3s 2 + 2 s + 5 Equating coefficients of like powers of s in Equation (d) leads to 50b3 = 5 b3 = 0.1 20b3 50b2 = 2 50b2 = 2 + 20(.1) b2 = 0 20b2 10b3 50b1 = 3 50b1 = 3 + 10(.1) b1 = .04 (d) (e) (f) (g)
b4 = 0 (h) Thus a state-space formulation of the equations governing the response of the system is & 1 0 0 y1 0.04 y1 0 y 0 & 0 1 0 y2 0 2 = + f (t ) (i) y3 0 & 0 0 1 y 3 0.1 & y 4 50 20 10 3 y 4 0 9.34 The closed-loop transfer function for the system is 3 s +1 2 + 2 s s + 4s + 16 H ( s) = 3 s +1 1 + 2 + 2 s s + 4s + 16 (2s + 3)( s + 1) = 2 s ( s + 4 s + 16) + (2s + 3)( s + 1)
2 s 2 + 5s + 3 s 3 + 6s 2 + 21s + 3 The state matrix for the state-space formulation of this closed loop system is 1 0 0 0 A= 0 1 3 21 6 = (a)
(b)
439
V3 = 2s 2 + 5s + 3
s 1 b1 0 s b2 = 2s 2 + 5s + 3 3 21 b3 b3 s 2 3b2 3sb1 21sb2 = 2s 2 + 5s + 3 (c) Equation (c) must hold for all s, thus coefficients of like powers of s must be the same on both sides of the equation, 3b2 = 3 b2 = 1 (d) 3b1 21b2 = 5 3b1 = 16 b1 = 16 / 3 b3 = 2 Thus the state-space formulation for this system is & y1 1 0 y1 26 / 3 & 0 & = 0 & 0 1 y 2 + 1 a(t ) y2 & 3 21 6 y 3 2 & y3 (e) (f)
(h )
SIMULINK models are constructed using both the transfer function formulation as well as the state-space formulation. The SIMULINK transfer function model is shown below
2s+3 s Step Transfer Fcn1 s+1 s2 +4s+16 Transfer Fcn Y To Workspace
440
10 t
15
20
25
When developing the state-space model note that since the input vector is determined using V3 the transfer function corresponds to that of y 3 and thus the appropriate output vector is C = [0 0 1] . The SIMULINK model developed using the state-space formulation is
x' = Ax+Bu y = Cx+Du Step A=[0 1 0;0 0 1;-3 -21 -6] B=[16/3;-1;2] C=[0 0 1] D=[0]
X To Workspace
441
10 t
15
20
25
(c)
442
Chapter 9 Substitution of Equation (c) into Equation (a) leads to 2s s + 1 2 C ( s) = B( s) = 2 s A( s) + C ( s) s +s+4 3 2 2 s + s + 4 ( s + 1) C ( s) = A( s ) + C ( s) 6s s 3 2 2 2 s + s + s + s + 4s + 4 6s C ( s ) = A( s ) 6s s 12 C ( s) = 3 A( s ) s + 2s 2 s + 4 Thus the transfer function for the system is C (s) H (s) = A( s ) 12 = 3 s + 2s 2 s + 4 The state matrix for a state-space formulation of the closed loop is 0 1 0 A= 0 0 1 4 1 2
(d)
(e)
(f)
V3 = 12
s 0
4
1 b1 s b2 = 12 1
b3
(g) (h)
b2 3 = 4 4
(i) (j)
(f)
443
A simulation of the model is run with a step input. The response is illustrated below. Note that since the sign on the summing junction in one of the feedback loops is positive the system is unstable.
9.37 The state-space model for the system of Figure P9.24 is obtained in the solution of Problem 9.24 as & h1 .1333 .1333 0 h1 0.667 & .2708 .1042 h2 + 0 qi (t ) h2 = .167 & h3 0 0.050 .210 h3 0
(a)
444
Chapter 9 The SIMULINK model using the state-space formulation is illustrated below
H 0.2u(t) x' = Ax+Bu y = Cx+Du A=[-0.133 0.133 0;0.167 -0.2708 0.1042;0 0.05 -0.210 B=[0.667;0;0] C=[1 1 1] D=[0;0;0]
h (m)
5 t (s)
10
9.38 The differential equations obtained during the solution of Example 6.41 are d 5.73 x10 3 o + 1.48 x10 4 o 2.99 x10 3 w = 0 dt d 4.04x10 4 w 2.99 x10 3 o + 1.35 x10 4 w = 1.05 x10 5 u (t ) dt Equations (a) and (b) are rewritten as d 0 = 2.58 0 + 0.522 w dt d w = 0.0740 0 0.3342 w + 2.60u (t ) dt
(a) (b)
(c) (d)
445
y1 = o
(e) (f)
y2 = w The state-space formulation of the problem is & 0.522 0 y1 2.58 y = 0.0740 0.3342 + 2.60u (t ) 2 & The steps in obtaining the state transition matrix are 0.522 s + 2.58 sI A = 0.0740 s + 0.3342
s + 0.3342 0.522 1 s + 2.58 s + 2.91s + 0.924 0.0740 The state transition matrix is obtained by inverting the transforms of Equation (h) 0.9928 0.2295 2.59t 0.0072 0.2295 0.318t = + e e 0.0325 0.0053 0.0325 0.9947 The response is obtained from
(g)
(sI A )1 =
(h)
(i)
y = (t )Bu( )d
0
t 0.9928 0.2295 2.59 ( t ) 0.0072 0.2295 0.318( t ) 0 = + e u ( )d e 0.0325 0.0053 0.0325 0.9947 0 2.60 t 0.5967 2.59 ( t ) 0.5967 0.318( t ) = + e u ( )d e 0.0138 2.586 0 Evaluation of the integrals of Equation (j) leads to o (t ) = 0.2304(1 e 2.59t ) + 1.867(1 e 0.318t )
(j)
(k) (l)
w (t ) = 0.0053(1 e
2.59 t
) + 8.13(1 e
0.318t
446