CP1 B8 Lecture No. 1 - Fundamental Principles & Methods PDF

Competency Training and Certification Program in Electric Power Distribution System Engineering
Certificate in
Power System Modeling and Analysis
Training Course in
Fundamental Principles and

Methods in Power System Analysis
U. P. NATIONAL ENGINEERING CENTER

NATIONAL ELECTRIFICATION ADMINISTRATION
Training Course in Fundamental Principles and Methods in Power System Analysis 2
Course Outline
1. Circuit Conventions and Notations

2. Power System Representation
3. Per Unit Quantities
4. Symmetrical Components
5. Network Equations and Methods of Solution
U. P. National Engineering Center Competency Training & Certification Program in Electric

National Electrification Administration Power Distribution System Engineering
1
Circuit Conventions & Notations
Voltage and Current Directions

Double Subscript Notation
Voltage, Current and Phasor Notation
Complex Impedance and Phasor Notation


Voltage and Current Directions
Polarity Marking of Voltage Source
Terminals: ZA
Plus sign (+) for the terminal where a I b
positive current comes out + - I
I VA +
Specification of Load Terminals: +
Plus sign (+) for the terminal whereVs
VB ZB
I
positive current enters -
-
Specification of Current Direction: n
o
Arrows for the positive current (i.e.,
from the source towards the load)

2

Double Subscript Notation
The letter subscripts on a voltage indicate the nodes of the circuit
between which the voltage exists.
The first subscript denotes the voltage of that node with respect to the
node identified by the second subscript.
ZA
a b
+ - VS - VA - Vb = 0
VA +
+
Vao - IabZA - Vbn = 0
VS = Vao I = Iab Zb Vb = Vbn
- Vao - Vbn
Iab =
- ZA
o n

Voltage, Current and Phasor Notation
jω t
VS = Vmε
ε jωt = cosωt + j sinωt
θ V
V = V ∠0o volts
( )
I = I ∠ − θ o amperes

3

+
+ R (Resistance)
jω t i(t) VL
VS = Vmε
- L (Inductance)
-
di ( t ) jω t
Applying Kirchoff’s voltage law, Ri ( t ) + L = Vmε
dt
The first order linear differential equation has a particular
jω t
solution of the form i ( t ) = K ε .
jω t jω t jω t j = −1
Hence, RK ε + j ω LK ε = Vmε

Solving for the current
Vm jω t
i( t ) = ε
R + jω L
Dividing voltage by current to get the impedance,
v( t ) V m ε jω t
Z = = = R + jω L
i( t ) Vm jω t
ε
R + jω L
Z
Therefore, the impedance Z is a complex ωL
quantity with real part R and an imaginary (j) θ
part ωL R
4

+
+ R (Resistance)
jω t i(t) VL
VS = Vmε
- C (Capacitance)
-
R
For Capacitive Circuit, .1 φ
1
Z = R − j( ) ωC
Z
ωC
Z= |Z|ejφφ or φ + jsinφ
Z = |Z|(cosφ φ) or Z = |Z|∠φ
∠φ

Z= |Z|ejφφ or φ + jsinφ
Z = |Z|(cosφ φ) or Z = |Z|∠φ
∠φ
ωt + 30°) volts
v = 141.4 cos(ω
ωt) amperes
i = 7.07 cos(ω
Vmax = 141.4 |V| = 100 ∠30
V = 100∠
Imax = 7.07 |I| = 5 ∠0
I = 5∠
100 ∠ 30
Z = = 20 ∠ 30
5 ∠0 20
10
30°
Z = 20(cos 30 + j sin 30 ) = 17.32 + j10 )
17.32
5
Power System Representation
Electrical Symbols
Three-Line and Single-Line Diagram
Equivalent Circuit of Power System
Components
Impedance and Admittance Diagram
Bus Admittance Matrix

Power System representation

Electrical Symbols
G Generator Switch
Circuit Breaker
or Transformer
Fuse
Transmission or
Distribution Line
Bus Node

6
Power System representation
3-phase wye neutral ungrounded
3-phase delta connection
3-phase wye neutral grounded
Current Transformer V Voltmeter

A Ammeter
Potential Transformer R Protective Relay


Three Line Diagram
The three-line diagram is used to represent each phase of a three-
phase power system. Transformer
CTs
Circuit Breaker
Distribution Lines
Main Bus
R Relays
a b c
R R
R
7

Single Line Diagram
The three-line diagram becomes rather cluttered for large power
systems. A shorthand version of the three-line diagram is referred
to as the Single Line Diagram.
Bus
Transformer
CB Distribution Line
Node
R
CT and Relay


Components: Generator
a
Ia
Za R a + jX s
Ea
Eb + Ia +
Ec Ib
Eg Va
Zb Zc b -
Ic -
c
Three-Phase Equivalent Single-Phase Equivalent

8

Components: Transformer
RT + jX T
v
+ Iex r +
r IH r RT = R H + a 2 R X
V HR c jX m aV X
XT = X H + a2 X X
- -
ZT
+ +
Single-Phase Equivalent
- -


Components: Transformer
Primary Y11 Y12 Y13 Y14 Y15 Y16
Secondary
A Y21 Y22 Y23 Y24 Y25 Y26 a
Y31 Y32 Y33 Y34 Y35 Y36
B b
Y41 Y42 Y43 Y44 Y45 Y46
C Y51 Y52 Y53 Y54 Y55 Y56 c
Y61 Y62 Y63 Y64 Y65 Y66
6x6 Admittance Matrix

Core Loss
Three-Phase Equivalent

9

Components: Transmission & Distribution
Lines
T&D Lines can be represented by an infinite series of

resistance and inductance and shunt capacitance.
∆l


Components: Long Transmission Lines
Length = Longer than 240 km. (150 mi.)
Z ' = Z c sinh γ l
+• • +
IS Y' 1 γl Y' IR
Vs = tanh VR
2 Zc 2 2
-• • -
Characteristic Impedance Propagation Constant γ= zy
z
ZC = Single-Phase Equivalent
y
10

Components: Medium-Length
Transmission Lines
Length = 80 – 240 km. (50 - 150 mi.)
Z = (r + jx L )l
+• • +
IS Y 1 / ωc Y IR
Vs = VR
2 2 2
-• • -

Components: Short Transmission Lines
Length = up to 80 km. (50 mi.)
Z = (r + jx L )l
+• • +
Vs Is = IR VR
-• • -
11

Components: Distribution Lines
A Zaa Zab Zac a
B Zba Zbb Zbc b
Equivalent C Zca Zcb Zcc c
π-Network Yaa Yab Yac Yaa Yab Yac
1/2 Yba Ybb Ybc 1/2 Yba Ybb Ybc

Yca Ycb Ycc Yca Ycb Ycc
Capture
Unbalanced
Characteristics Three-Phase Equivalent

Single Phase Equivalent of
Balanced Three-Phase System
Ic
c c
Eco = |E| ∠120° V
Ia ZR
a a
o n
Eao = |E|∠0° V ZR ZR
Ebo= |E| ∠240° V

b b
Ib

12

a
Ia
a
o n E∠0
Eao = |E|∠ 0° V ZR Ia = = I∠ − θ
Z R ∠θ
o n
ZR E∠240
ZR Ib = = I∠( 240 − θ )
Z R ∠θ
Ebo= |E| ∠240° V Ib
b b
c c
Eco = |E| ∠120° V Ic E∠120
Ic = = I∠( 120 − θ )
ZR Z R ∠θ
o n
ZR
Note: Currents are Balanced

Ic
Single Phase c
Eco = |E| ∠120° V c
Representation
Ia ZR
of a Balanced a a
Three-Phase o n
Eao = |E|∠0° V ZR
System ZR
Ebo= |E| ∠240° V

b b
Ib
a a
Ia ZR
Eao = |E|∠ 0° V
o n
13

Impedance and Admittance Diagrams
a c b
Bus Gen Line 1 3 2
1 a 1-3
2 b 2-3
3 c 1-4
4 2-4
3-4
4
Single Line Diagram

0
Ea za Ea Ec Eb
1 za zc zb
Generator
1 3 2
0 zd ze
z13 zf zh
zg
1 3
Line
4
0 Impedance Diagram
14

VL VL
Zg
+ Eg = ISZP
Eg Is Zp
IL IL Zg = Zp
-
The two sources will be equivalent if VL

and IL are the same for both circuits.
0
I1 = Ea/za
y01 = 1/za I1 I3 I2
y01 y03 y02
I2 = Eb/zb
y02 = 1/zb 1 3 2
I3 = Ec/zc y13 y23
y03 = 1/zc y14
y24
y13 = 1/zd y14 = 1/zf
y23 = 1/ze y24 = 1/zg 4
y34 = 1/zh Admittance Diagram
15

Applying Kirchoff’s Current Law
at node 1:
I 1 = V 1 y 01 + (V 1 − V 3 ) y 13 + (V 1 − V 4 ) y 14
at node 2:
I 2 = V 2 y 02 + (V 2 − V 3 ) y 23 + (V 2 − V 4 ) y 24
at node 3:
I 3 = V 3 Y03 + (V 3 − V 2 ) y 23 + (V 3 − V 4 ) y 34 + (V 3 − V 1 ) y 13
at node 4:
0 = (V 4 − V1 ) y 14 + (V 4 − V 2 ) y 24 + (V 4 − V 3 ) y 34

Rearranging the equations,
I1 = V1 ( y01 + y13 + y14 ) −V3 y13 −V4 y14

I 2 = V 2 ( y 02 + y 23 + y 24 ) − V 3 y 23 − V 4 y 24
I 3 = V3 ( y03 + y23 + y34 + y13 ) − V2 y23 − V4 y34 − V1 y13
0 = V 4 ( y 14 + y 24 + y 34 ) − V 1 y 14 − V 2 y 24 − V 3 y 34

16
Per Unit Quantities
The Per Unit System

Per Unit Impedance
Changing Per Unit Values
Consistent Per Unit Quantities of Power
System
Advantages of Per Unit Quantities

The Per Unit System

Per Unit Value
Actual Value
Per Unit Value =
Base Value
Per-unit Value is a dimensionless quantity

Per-unit value is expressed as decimal
Actual Value
Percent =
100
17
The Per Unit System

Per Unit Value
Actual Value of Power
Per Unit Power =
Base Power
Actual Value of Voltage
Per Unit Voltage =
Base Voltage
Actual Value of Current
Per Unit Current =
Base Current
Actual Value of Impedance

Per Unit Impedance =
Base Impedance
The Per Unit System

Per Unit Calculations
PU Voltage
PU Current =
PU Impedance
PU Power = PU Voltage x PU Current

18
The Per Unit System

Example: I
+
+ Zline = 1.4 ∠75° Ω
Zload = 20 ∠30° Ω 2540 ∠0° V

Vs = ?
-
-
Determine Vs
The Per Unit System

Choose: Base Impedance = 20 ohms (single phase)
Base Voltage = 2540 volts (single phase)
PU Impedance of the load = 20∠30°/20 = ______ p.u.
PU Impedance of the line = 1.4∠75°/20 = ______ p.u.
PU Voltage at the load = 2540∠0° /2540 = ______ p.u.
Line Current in PU = PU voltage / PU impedance of the load

= ______ / ______ = ______ p.u.
PU Voltage at the Substation = Vload(pu) + IpuZLine(pu)
= ________ + _______ x _______ = _______ p.u.
19
The Per Unit System

1.0 ∠-30° p.u.
+ 0.07 ∠75° p.u. +
1.05∠2.70° 1.0∠30° p.u. 1.0∠0° p.u.
- -
The magnitude of the voltage at the substation is
1.05 p.u. x 2540 Volts = _______ Volts
The Per Unit System

Establishing Base Values
1. Base values must satisfy fundamental electrical
laws (Ohm’s Law and Kirchoff’s Laws)
2. Choose any two electrical parameters
• Normally, Base Power and Base Voltage are chosen
3. Calculate the other parameters
• Base Impedance and Base Current

20
The Per Unit System

For a Given Base Power and Base Voltage,
Base Power
Base Current =
Base Voltage
Base Voltage (Base Voltage)2

Base Impedance = =
Base Current Base Power
The Per Unit System

For Single Phase System, For Three Phase System,
Pbase(1φ) Pbase(3φ)
Ibase = ------------ Ibase = ------------
Vbase(1φ) √3Vbase(LL)
Vbase(1φ) Vbase(LN)
Zbase = ------------ Zbase = ------------
Ibase(1φ) Ibase(L)
[Vbase(1φ)]² [Vbase(LL)]²
= ------------ = ------------
Pbase(1φ) Pbase(3φ)
21
The Per Unit System

• Base Values can be established from Single Phase or Three
Phase Quantities
1
Base MVA 1φ = Base MVA 3φ
3
Base kV LL
Base kV 1φ =
3
• Base MVA is the same base value for Apparent, Active and
Reactive Power
• Base Z is the same base value for Impedance, Resistance and
Reactance
The Per Unit System

Example:
Base kVA3φφ = 30,000 kVA Base kVA1φφ = 10,000 kVA
= 30 MVA = 10 MVA
Base kVLL = 120 kV Base kVLN = 69.282 kV
Base Z = (120)2/30 Base Z = (69.282)2/10
= 480 ohms = 480 ohms
30 x1000 10 x1000
Base Current = Base Current =
3 ( 120 ) 69.282
= 144.34 Amps = 144.34 Amps
22
Per Unit Impedance

Generators
Manufacturers provide the following impedance in per unit:
1.
2.
3.
4.
Armature Resistance, Ra
Direct-axis Reactances, Xd”, Xd’ and Xd
Quadrature-axis Reactances, Xq”, Xq’ and Xq
Negative Sequence Reactance, X2
} Positive
Sequence
Impedances
5. Zero Sequence Reactance, X0

{}
The Base Values used by manufacturers are:

1. Rated Capacity (MVA, KVA or VA)
2. Rated Voltage (kV or V)

Per Unit Impedance

Transmission and Distribution Lines
R(Ω
=
)
R( pu )
Z base
X L( Ω )
X L ( pu ) =
Z base
X C( Ω )
X C ( pu ) =
Z base
23
Per Unit Impedance

Transformers
The ohmic values of resistance and leakage reactance of a
transformer depends on whether they are measured on the
high- or low-tension side of the transformer
The impedance of the transformer is in percent or per unit with
the Rated Capacity and Rated Voltages taken as base Power and
Base Voltages, respectively
The per unit impedance of the transformer is the same
regardless of whether it is referred to the high-voltage or low-
voltage side
The per unit impedance of the three-phase transformer is the
same regardless of the connection

Per Unit Impedance

Example
A single-phase transformer is rated 110/440 V, 2.5 kVA.
The impedance of the transformer measured from the low-
voltage side is 0.06 ohms. Determine the impedance in per unit
(a) when referred to low-voltage side and (b) when referred to
high-voltage side
Solution
0.110 2
Low-voltage Zbase = = ______ ohms
2.5 / 1000
PU Impedance, Zpu = = ______ p.u.

24
Per Unit Impedance
If impedance had been measured on the high-voltage

side, the ohmic value would be
2
 440 
Z = 0 . 06   = _______ ohms
 110 
High Voltage, Zbase = = _______ ohms
PU Impedance, Zpu = = _______ p.u.

Note: PU value of impedance referred to any side of the
transformer is the same
Changing Per-Unit Values
Example:
Consider a three-phase transformer rated 20
MVA, 69 kV/13.2 kV voltage ratio and a reactance
of 7%. The resistance is negligible.
a) What is the equivalent reactance in ohms referred to the

high voltage side?
b) What is the equivalent reactance in ohms referred to the low
voltage side?
c) Calculate the per unit values both in the high voltage and
low voltage side at 100 MVA.

25
SOLUTION:
a) Pbase = 20 MVA
Vbase = 69 kV (high voltage)
( kV)²
Zbase = = ________ ohms
( MVA)
Xhigh = Xp.u. x Zbase = _______ x _______= _______ ohms

SOLUTION:
b) Pbase = 20 MVA
Vbase = 13.2 kV (low voltage)
( kV)²
Zbase = = ________ ohms
( MVA)
Xlow = Xp.u. x Zbase = _______ x _______= _______ ohms

26

c) Pbase = 100 MVA
Vbase,H = 69 kV
(69)²
Zbase,H = = ________ ohms
100
ohms
Xp.u.,H = = ______ p.u.
ohms
Vbase,L = 13.2 kV
(13.2)²
Zbase,L = = _______ ohms
100
ohms
Xp.u.,L = = ______ p.u.
ohms
Note that the per unit quantities are the same regardless of the voltage level.

Three parts of an electric system are designated A, B
and C and are connected to each other through transformers,
as shown in the figure. The transformer are rated as follows:
A-B 10 MVA, 3φ, 13.8/138 kV, leakage reactance 10%

B-C 10 MVA, 3φ, 138/69 kV, leakage reactance 8%
300 Ω/ φ
SOURCE A B C LOAD PF=100 %
A-B B-C
Determine the voltage regulation if the voltage at the load is

66 kV.
27

Solve using actual quantities
XAB=10% XBC=8%
300 Ω/ φ
VA Vc PF=100%
A-B B-C
13.8/138 kV 138/69 kV


SOLUTION USING PER UNIT METHOD:
Pbase = 10 MVA
VA,base = 13.8 kV
VB,base = 138 kV (69)²
VC,base = 69 kV ZC,base = -------- = _____ ohms
10
ZLOAD,p.u. = ---------- = ______ + j _____ p.u.
VC,p.u. = -------------- = _______ p.u.
Ip.u. = ---------------- = _______ p.u.

28

VA = _______ + ( ________ ) x ( ________ + ________ )
= ________ + j ________ p.u.
= _________ p.u.
VNL - VL
V.R. = ---------------- x 100%
VL
V.R. = ------------------------ x 100%

=

Consider the previous example, What if transformer A-B is

20 MVA instead of 10 MVA. The transformer nameplate
impedances are specified in percent or per-unit using a
base values equal to the transformer nameplate rating.
The PU impedance of the 20 MVA transformer cannot be

added to the PU impedance of the 10 MVA transformer
because they have different base values
The per unit impedance of the 20 MVA can be referred to 10

MVA base power

29

Convert per unit value of 20 MVA transformer,
Pbase = 20 MVA (Power Rating)
Vbase,H = 138 kV (Voltage Rating)
(138)²
Zbase,H = ---------- = _______ ohms
20
Xactual,H = 0.10 p.u. x _______ ohms = _______ ohms
At Pbase = 10 MVA (new base)
(138)² The per unit impedance of
Zbase,H = ---------- = 1904.4 ohms
10 the 20MVA and 10 MVA
95.22 transformer can now be
Xp.u.(new) = ---------- = 0.05 p.u. added.
1904.4
Note that the transformer can have different per unit

impedance for different base values (i.e., the actual ohmic
impedances of the equipment is independent of the selected
base values), then
Zactual = Zpu1 • Zbase1 Zactual = Zpu2 • Zbase2
Z pu 1 ⋅ Z base 1 = Z pu 2 ⋅ Z base 2
Z
Z pu 2 = Z pu 1 ⋅ base 1
Z base 2

30

Recall: Z base =
(base voltage )
2
base Power
Then, (kV LL , base 1 ) 2
MVA 3 φ , base 1
Z = Z
pu 2 pu 1
(kV LL , base 2
2
)
MVA 3 φ , base 2
2
or,  kV   MVA 3 φ ,base 2 
Z pu 2 = Z pu 1  LL ,base 1   
  MVA 
 kV LL ,base 2   3 φ ,base 1 

[ Vbase( given ) ] 2 Pbase( new )
Z p .u .( new ) = Z p .u .( given ) x x
[ Vbase( new ) ] 2 Pbase( given )
Example
A three-phase transformer is rated 400 MVA, 220Y/22 ∆ kV.
The impedance measured on the low-voltage side of the
transformer is 0.121 ohms (approx. equal to the leakage
reactance). Determine the per-unit reactance of the transformer
for 100 MVA, 230 kV base values at the high voltage side of the
transformer.
31

Solution
On its own base the transformer reactance is
( )
X= = ________ pu
( )2
( )
On the chosen base the reactance becomes
( )2
X=( )x x ( = )________ pu
( )2

Consistent Per Unit Quantities

of Power System
Procedure:
a) Establish Base Power and Base Voltages
• Declare Base Power for the whole Power
System
• Declare Base Voltage for any one of the Power
System components
• Compute the Base Voltages for the rest of the
Power System Components using the voltage
ratio of the transformers
Note: Define each subsystem with unique Base Voltage based on separation due to
magnetic coupling

32

of Power System
b) Compute Base Impedance and Base Current
• Using the Declared Base Power and Base
Voltages, compute the Base Impedances and
Base Currents for each Subsystem
c) Compute Per Unit Impedance
• Using the declared and computed Base Values,
compute the Per Unit values of the impedance
by:
Dividing Actual Values by Base Values
Changing Per Unit Impedance with change in Base
Values

of Power System
Use Base Power = 100 MVA
T1 T2
G2
L1
G1
G3
Generator 1 (G1): 300 MVA; 20 kV; 3φ; Xd” = 20 %
Transmission Line(L1): 64 km; XL = 0.5 Ω / km
Transformer 1 (T1): 3φ; 350 MVA; 230 / 20 kV; XT1 = 10 %
Transformer 2 (T2): 3-1φ; 100 MVA; 127 / 13.2 kV; XT2 = 10 %
Generator 2 (G2): 200 MVA; 13.8kV, Xd” = 20 %
Generator 3 (G3): 100 MVA; 13.8kV, Xd” = 20 %
33

of Power System
T1 T2
Transmission Line G2
G1
G3
XT1 XLINE XT2
XG1 XG2 XG3
E1 E2 E3


of Power System
a) & b) Establish Base Power, Base Voltages, Base
Impedance, and Base Current
Base Power:
Sub-
Vbase (kV) Zbase (Ohm) Ibase (Amp)
System

34

of Power System
c) Compute Per Unit Impedance
G1:
G2:
G3:
L1:
T1:
T2:

Advantages of
Per-Unit Quantities
The computation for electric systems in per-unit
simplifies the work greatly. The advantages of Per Unit
Quantities are:
1. Manufacturers usually specify the impedances of
equipments in percent or per-unit on the base of the
nameplate rating.
2. The per-unit impedances of machines of the same type and
widely different rating usually lie within a narrow range.
When the impedance is not known definitely, it is generally
possible to select from tabulated average values.

35
Advantages of
Per-Unit Quantities
3. When working in the per-unit system, base voltages can be
selected such that the per-unit turns ratio of most
transformers in the system is equal to 1:1.
4. The way in which transformers are connected in three-
phase circuits does not affect the per-unit impedances of
the equivalent circuit, although the transformer connection
does determine the relation between the voltage bases on
the two sides of the transformer.
5. Per unit representation yields more meaningful and easily
correlated data.

Advantages of
Per-Unit Quantities
6. Network calculations are done in a much more handier
fashion with less chance of mix-up
• between phase and line voltages
• between single-phase and three-phase powers, and
• between primary and secondary voltages.

36
Symmetrical Components
Sequence Components of Unbalanced Phasor

Sequence Impedance of Power System
Components
Practical Implications of Sequence
Components of Electric Currents

Sequence Components of
Unbalanced Phasor
In a balanced Power System,
Generator Voltages are three-phase balanced
Line and transformer impedances are balanced
Loads are three-phased balanced
Single-Phase Representation and Analysis can

be used for the Balanced Three-Phase Power
System

37
Unbalanced Phasor
In a practical Power Systems,
Lines are not transposed.
Single-phase transformers used to form three-phase
banks are not identical.
Loads are not balanced.
Presence of vee-phase and single phase lines.
Faults
Single-phase Representation and Analysis
cannot be use for an unbalanced three-phase
power system.

Unbalanced Phasor
Any unbalanced three-phase system of phasors
may be resolved into three balanced systems of
phasors which are referred to as the symmetrical
components of the original unbalanced phasors,
namely:
a) POSITIVE-SEQUENCE PHASOR
b) NEGATIVE-SEQUENCE PHASOR
c) ZERO-SEQUENCE PHASOR

38
Unbalanced Phasor
REFERENCE PHASE SEQUENCE: abc
Phase c
Positive Sequence Phasors are
120° three-phase, balanced and have
120° Phase a the phase sequence as the
120° original set of unbalanced
phasors.
Phase b
Negative Sequence Phasors are three-phase, balanced but with a
phase sequence opposite to that original set of unbalnced
phasors.
Zero Sequence Phasors are single-phase, equal

in magnitude and in the same direction.
Unbalanced Phasor
Each of the original unbalanced phasor is the sum
of it’s sequence components. Thus,
Va = Va1 + Va2 + Va0

Vb = Vb1 + Vb2 + Vb0
Vc = Vc1 + Vc2 + Vc0
Where,
Va1 – Positive Sequence component of Voltage Va
Va2 – Negative Sequence component of Voltage Va
Va0 – Zero Sequence component of Voltage Va
39
Unbalanced Phasor
OPERATOR “a”
An operator “a” causes a rotation of 120° in the
counter clockwise direction of any phasor.
a = 1 ∠ 120° a² = 1 ∠ 240° a³ = 1 ∠ 0°
Vb in terms of Va Vc in terms of Va
Vb = a² Va Vc = a Va
Vb1 = a² Va1 Vc1 = a Va1
Vb2 = a Va2 Vc2 = a2 Va2
Vb0 = Va0 Vc0 = Va0
Unbalanced Phasor
Writing again the phasors in terms of phasor Va
and operator “a”,
Vb = Va0 + a²Va1 + aVa2
Vc = Va0 + aVa1 + a²Va2
Computing for Va0, Va1 & Va2
V a0 =
1
3
[V a + V b + V c ] 1
[
V a 1 = V a + aV b + a 2 V c
3
]
V a2 =
1
3
[
V a + a 2 V b + aV c ]
40
Unbalanced Phasor
Vc = 8 ∠143.1 EXAMPLE:
Determine the symmetrical
components of the following
unbalanced voltages.
Va = 4 ∠0
Vb = 3 ∠-90
Unbalanced Phasor
For Phasor Va:
1
V a0 = ( V a + V b + V c )
3
1
= ( 4 ∠ 0 + 3 ∠ - 90 + 8 ∠ 143.1)
3
= 1 ∠ 143.05
1
V a 1 = ( V a + aV b + a 2V c )
3
1
= [4 ∠ 0 + (1 ∠ 120)(3 ∠ - 90) + (1 ∠ 240)(8 ∠ 143.1) ]
3
= 4.9 ∠ 18.38
41
Unbalanced Phasor
For Phasor Va:
1
V a 2 = ( V a + a 2Vb + aV c )
3
1
= [4 ∠0 + (1 ∠ 240)(3 ∠ - 90) + (1 ∠120)(8 ∠143.1) ]
3
= 2.15 ∠ − 86.08

Unbalanced Phasor
Components of Vb can be obtained by operating the
sequence components of phasor Va.
V b0 = V a0
= 1 ∠ 143.05 = 1 ∠ 143.05
V b1 = a 2V a 1
= (1 ∠ 240)(4.9 ∠ 18.38)
= 4.9 ∠ 258.38
= 4.9 ∠ - 101.62
V b2 = aV a2
= (1 ∠ 120)(2.15 ∠ - 86.08)
= 2.15 ∠ 33.92
42
Unbalanced Phasor
Similarly, components of phasor Vc can be obtained by
operating Va.
V c0 = V a0
= 1 ∠ 143.05 = 1 ∠ 143.05
V c1 = aV a 1
= (1 ∠ 1 2 0)(4.9 ∠ 18.38)
= 4.9 ∠ 1 3 8.38
V c2 = a 2 V a2
= (1 ∠ 2 4 0)(2.15 ∠ - 86.08)
= 2.15 ∠ 1 5 3.92
Unbalanced Phasor
Vc1 Va0 Vb0 Vc0
Va1 Zero Sequence Components

Vc2 Vb2
Va2
Negative Sequence Components
Vb1
Positive Sequence Components
43
Unbalanced Phasor
The results can be checked either mathematically or
graphically.
V a = V a0 + V a1 + V a2
= 1 ∠ 143 . 05 + 4.9 ∠ 18.38 + 2.15 ∠ - 86.08
= 4 ∠0
V b = V b0 + V b1 + V b2
= 1 ∠ 143.05 + 4.9 ∠ - 101.62 + 2.15 ∠ 33.92
= 3 ∠ - 90
V c = V c0 + V c1 + V c2
= 1 ∠ 143 . 05 + 4.9 ∠ 138.38 + 2.15 ∠ 153.92
= 8 ∠ 143.1
Unbalanced Phasor
Add Sequence Components Graphically
Va1
Va2 Vb0
Va = 4 ∠0
Va0
Vc2 Vb1 Vb = 3 ∠-90
Components of Va
Vc1
Vc = 8 ∠143.1 Vb2
Vc0 Components of Vb
Components of Vc
44
Sequence Impedance of
Power System Components
Sequence Networks
F F F
+ + +
Ia1 Ia2 Ia0
Z1 Z2 Z0
Va1 Va2 Va0
+
Vf
- - -
Va1 = V f – I a1 Z 1 V =-I Z V =-I Z

a2 a2 2 ao ao o
Positive Sequence Negative Sequence Zero Sequence

Sequence Impedance of
Power System Components
In general,
Z1 ≠ Z2 ≠ Z0 for generators
Z1 = Z2 = Z0 for transformers
Z1 = Z2 ≠ Z0 for lines

45

ZERO-SEQUENCE CURRENTS:
IA0 Ia0
A a
IC0 B Ic0 b
C 3I0 c 3Io
IB0 Ib0
The neutral return (ground) carries the in-phase zero-

sequence currents.


ZERO-SEQUENCE CURRENTS:
IC0 Ia0
I0 = 0
a
C
A IA0
I0 = 0 B Ic0 b
I0 = 0 c 3Io
IB0 Ib0
In-phase zero-sequence currents

circulates in the delta-connected
The neutral return (ground)
transformer windings.
carries the in-phase zero-
sequence currents.
There is “balancing ampere-turns”
for the zero-sequence currents.
46

NEGATIVE-SEQUENCE CURRENTS:
A three-phase unbalanced load produces a reaction field which
rotates synchronously with the rotor-field system of generators.
Any unbalanced condition will have negative sequence components.
This negative sequence currents rotates counter to the
synchronously revolving field of the generator.
The flux produced by sequence currents cuts the rotor field at twice
the rotational velocity, thereby inducing double frequency currents
in the field system and in the rotor body.
The resulting eddy-currents are very large and cause severe heating
of the rotor.

Network Equations and Methods of

Solution

Network Equations
Matrix Representation of System of Equations
Type of Matrices
Matrix Operations
Direct Solutions of System of Equations
Iterative Solutions of System of Equations

47
a c b
Bus Gen Line 1 3 2
1 a 1-3
2 b 2-3
3 c 1-4
4 2-4
3-4
4
Single Line Diagram
Ea za Ea Ec Eb
1 za zc zb
Generator
1 3 2
0 zd ze
z13 zf zh
zg
1 3
Line
4
0 Impedance Diagram
48
0
I1 = Ea/za
y01 = 1/za I1 I3 I2
y01 y03 y02
I2 = Eb/zb
y02 = 1/zb 1 3 2
I3 = Ec/zc y13 y23
y03 = 1/zc y14 y24
y13 = 1/zd y14 = 1/zf
y23 = 1/ze y24 = 1/zg 4
y34 = 1/zh Admittance Diagram
Network Equations
Applying Kirchoff’s Current Law
at node 1:
I 1 = V 1 y 01 + (V 1 − V 3 ) y 13 + (V 1 − V 4 ) y 14
at node 2:
I 2 = V 2 y 02 + (V 2 − V 3 ) y 23 + (V 2 − V 4 ) y 24
at node 3:
I 3 = V 3 Y03 + (V 3 − V 2 ) y 23 + (V 3 − V 4 ) y 34 + (V 3 − V 1 ) y 13
at node 4:
0 = (V 4 − V1 ) y 14 + (V 4 − V 2 ) y 24 + (V 4 − V 3 ) y 34
49
Network Equations
Rearranging the equations,
I 1 = V 1 ( y 01 + y 13 + y 14 ) − V 3 y 13 − V 4 y 14
I 2 = V 2 ( y 02 + y 23 + y 24 ) − V 3 y 23 − V 4 y 24
I 3 = V 3 ( y 03 + y 23 + y 34 + y 13 ) − V 2 y 23 − V 4 y 34 − V 1 y 13
0 = V 4 ( y 14 + y 24 + y 34 ) − V 1 y 14 − V 2 y 24 − V 3 y 34
In matrix form,
I1  y01 + y13 + y14 0 - y13 - y14 V1 
I   0 y02 + y23 + y24 - y23 - y24 V 
 2 =   2 
I3   - y13 - y23 y03 + y23 + y34 + y13 - y34 V3 
    
0   − y14 − y24 − y34 y14 + y24 + y34V4 

Network Equations
The standard form of n independent equations:
I1   Y 11 Y 12 Y 13 L Y 1 n  V 1 
  
   Y 21 Y 23 L Y 2 n  V 2
I2 
Y 22
   
I3  =  Y 31 Y 32 Y 33 L Y 3 n  V3 
    
M  M M M M  M 
 
I  Y n 1 Yn2 Y n 3 L Y nn  V 
 n   n
[ I ] = [Ybus][V]
Ybus is also called Bus Admittance Matrix

50
Network Equations
 Y 11 Y 12 Y 13 L Y 1 n 
 
 Y 21 Y 22 Y 23 L Y 2 n 
 
[YBUS] =  Y 31 Y 32 Y 33 L Y 3 n 
 
M M M M 
 
Y n 1 Yn2 Y n 3 L Y nn 
 
Ypp = self-admittance, the sum of all admittances terminating on the

node (diagonal elements)
Ypq = mutual admittance, the negative of the admittances connected
directly between the nodes identifed by the double subscripts

Matrix Representations of
System of Equations
System of n Linear Equations
In the following system of equations:
a11 x1 + a12 x 2 + a13 x 3 = y 1

a 21 x1 + a 22 x 2 + a 23 x 3 = y 2
a 31 x1 + a 32 x 2 + a 33 x 3 = y 3
x1, x 2, and x3, are unknown variables, a11, a12,… …, a33
are the coefficient of these variables and y1, y2, and y3
are known parameters.
51
System of Equations
Writing again the phasors in terms of phasor Va
and operator “a” (Symmetrical Components),
Vb = Va0 + a²Va1 + aVa2
Vc = Va0 + aVa1 + a²Va2
Rearranging and writing in matrix form
Va  1 1 1  Va 0 
V  = 1 a 2 a  Va1 
 b 
Vc  1 a a 2  Va 2 
System of Equations
Let 1 1 1 
A =  1 a2 a 
 1 a a 2 
Using the numerical techniques, the inverse of A can be
obtained as,
1 1 1
1
A −1 = 1 a a 2 
3
1 a 2 a 
52
System of Equations
The symmetrical sequence components can be obtained by
pre-multiplying the original phasors (Va, Vb and Vc) by the
inverse of A,
Va 0  1 1 1  Va 
V  = 1 a
1
a 2  Vb 
 a1  3 
Va 2  1 a 2 a  Vc 
Thus,
V a0 =
1
3
[V a + V b + V c ] Va1 =
1
3
[
V a + aV b + a 2 V c ]
V a2 =
1
3
[
V a + a 2 V b + aV c ]
System of Equations
The coefficients form an array
 a11 a12 a13 

A = a21 a22 a23 
 
a31 a32 a33 
which is called the Coefficient Matrix of the system of
equations.
53
System of Equations
Similarly, the variables and parameters can be

written in matrix form as.
 x1   y1 
X =  x2  and Y =  y2 
   
 x3   y3 
System of Equations
System of Equations in Matrix Form
The system of equations in matrix notation is
 a11 a12 a13   x1   y1 

a a 22 a23   x2  =  y 2 
 21    
 a31 a32 a33   x3   y 3 
AX = Y
54
Definition of a MATRIX
A matrix consists of a rectangular array of elements
represented by a single symbol.
[A] is a shorthand notation for the matrix and aij
designates an individual element of the matrix.
A horizontal set of elements is called a row and a
vertical set is called a column.
The first subscript i always designates the number of
the row in which the element lies.
The second subscript j designates the column.
For example, element a23 is in row 2 and column 3.

Definition of a MATRIX
 a11 a12 a13 K a 1n 
a a22 a23 K a2n 
 21 
[A] = [a ij ] =  a31 a32 a33 K a3n 
 
 M M M 
am1 am2 am3 K amn 
The matrix has m rows and n columns and is said to
have a dimension of m by n (or m x n).
[aij]mxn
55
Definition of a Vector
A vector X is defined as an ordered set of elements. The
components x1, X2…, Xn may be real or complex
numbers or functions of some dependent variable.
 x1 
x 
X =  2
M 
 
 xn 
“n” defines the dimensionality or size of the vector.

Matrices with only one row (n = 1) are called Row

Vectors while those with one column (m=1) are called
Column Vectors. The elements of a vectors are
denoted by single subscripts as the following:
 c1 
c 
C =  2 
R = [r1 r2 L rn ]  M 
 
cm 
Thus, R is a row vector of dimension n while C is a
column vector of dimension m.
56
Type of Matrices
Square Matrix
Upper Triangular Matrix
Lower Triangular Matrix
Diagonal Matrix
Identity or Unit Matrix
Symmetric Matrix
Skew-symmetric Matrix
Null Matrix

Type of Matrices
A square matrix is a matrix in which m = n.
For a square, the main or principal diagonal consists

of the elements of the form aii; e.g., for the 3 x 3
matrix shown
a11 a12 a13 

A = a21 a22 a23 
 
a31 a32 a33 
the elements a11, a22, and a33 constitute the principal
diagonal.
57
Type of Matrices
An upper triangular matrix is one where all the

elements below the main diagonal are zero.
u11 u12 u13 

U=0 u22 u23 
 
 0 0 u33 
Type of Matrices
A lower triangular matrix is one where all elements

above the main diagonal are zero.
l11 0 0
L = l21 l22 0
 
l31 l32 l33 

58
Type of Matrices
A diagonal matrix is a square matrix where all
elements off the diagonal are equal to zero.
Note that where large blocks of elements are zero,
they are left blank.
d11 0 0
D=  0 d22 0
 
 0 0 d33 
Type of Matrices
An identity or unit matrix is a diagonal matrix where
all elements on the main diagonal are equal to one.
1 0 0
I =  0 1 0 
 0 0 1 

59
Type of Matrices
A symmetric matrix is one where aij = aji for all i’s and j’s.
5 1 2 a12 = a21 = 1
S = 1 3 7 a13 = a31 = 2
 
2 7 8  a23 = a32 = 7
Type of Matrices
A skew-symmetric matrix is a matrix which has the
property aij = -aji for all i and j; this implies aii = 0
0 −5 3
K = 5 6
a12 = −5
0
  a 21 = +5
− 3 −6 0
60
Type of Matrices
The null matrix is matrix whose elements are

equal to zero.
0 0 0
N = 0 0 0
 
0 0 0 

Matrix Operations
Addition of Matrices
Product of a Matrix with a Scalar
Multiplication of Matrices
Transpose of a Matrix
Kron Reduction Method
Determinant of a Matrix
Minors and Cofactors of a Matrix
Inverse of a Matrix

61
Two matrices A = [aij] and B = [bij] can be added
together if they are of the same order (mxn). The sum
C = A + B is obtained by adding the corresponding
elements.
C = [cij] = [aij + bij]

Example:
1 4 0 5 2 6
A =  B = 
2 7 3  0 1 1 
then,
(1 + 5) (4 + 2) (0 + 6) 6 6 6
A+ B=  =
(2 + 0) (7 + 1) (3 + 1) 2 8 4
(1 − 5) (4 − 2) (0 − 6) − 4 2 − 6
A− B =  =
(2 − 0) (7 − 1) (3 − 1)  2 6 2 
62
Example:
1+ j2 4 − j1 6 − j3 3 + j2 2 − j1 7 + j5
A = 4 + j1 5 + j3 1 + j1 B = 2 + j1 4 + j6 5 + j4
   
6 + j3 1 − j1 8 + j9 7 − j5 5 − j4 6 + j5
then,
(1+ j2) +(3+ j2) (4− j1) +(2− j1) (6 − j3) +(7 + j5)
A+ B= (4+ j1) +(2+ j1) (5+ j3) +(4+ j6) (1+ j1) +(5+ j4) 

(6 + j3) +(7 − j5) (1− j1) +(5− j4) (8+ j9) +(6 + j5)
 4 + j4 6 − j2 13+ j2 
A + B =  6 + j2 9 + j9 6 + j5 
 
13− j2 6 − j5 14+ j14
− 2 + j0 2 + j0 −1− j8
A− B =  2 + j0 1− j3 − 4 − j3
 
−1+ j8 − 4 + j3 2 + j4 

63

A matrix is multiplied by a scalar k by multiplying
all elements mn by k , that is,
 ka11 ka12 L ka1n 

 ka ka22 L ka2 n 
kA = Ak =  21
 M M M M 
 
kam1 kam 2 L kamn 


Example:
4 3
A= 5 2 and k =3
 
6 1
4 3  12 9
B = kA = 3 5 2 B =  15 6
   
6 1 18 3 
64

Example:
4 + j1 3 − j6
A =  5 - j2 2 + j5 and k = 3
 
6 + j3 1 − j4
4 + j1 3- j6 12+ j3 9 − j18

B = kA=3 5 - j2 2 + j5 B = 15− j6 6 + j15
   
6 + j3 1- j4 18+ j9 3 − j12
Two matrices A = [aij] and B = [bij] can be multiplied in

the order AB if and only if the number of columns of A is
equal to the number of rows of B .
That is, if A is of order of (m x l), then B should be of
order (l x n).
If the product matrix is denoted by C = A B, then C is of
order (m x n). The elements cij are given by
l
cij = ∑ aik bkj for all i and j.
k =1
65
An easy way to check whether two matrices can
be multiplied.
[A ]m x l [B ]l x n = [C ]m x n
Interior dimensions are equal

multiplication is possible
Exterior dimensions define
the dimensions of the result

Example:  a11 a12 
b b 
A = a21 a22  and B =  11 12 
a31 b21 b22  2 x 2
a32  3 x 2
then
 a11 a12 
 b b12 
C = A B = a 21 a22   11
  b21 b22 
 a31 a32 
cij = ∑ aik bkj c 11 = a 11 b 11 + a 12 b 21
66
 (a11b11 + a12b21 ) (a11b12 + a12b22 ) 

C = AB = ( a21b11 + a22b21 ) ( a21b12 + a22b22 )
 (a31b11 + a32b21 ) ( a31b12 + a32b22 )
cij = ∑ aik bkj

Example:
1 4  7 8 
and B=
A = 2 5  
9 0  2 x 2
3 6  3 x 2
then
1 4
7 8
C = AB =  2 5  
0 
6  
9
 3
67
( 1x7 + 4 x9 ) ( 1x8 + 4 x0 )
C = AB = ( 2 x7 + 5 x9 ) ( 2 x8 + 5 x0 )
( 3 x7 + 6 x9 ) ( 3 x8 + 6 x0 )
 43 8 
C =  59 16 
 
 75 24 

Example:
1+ j2 4− j1 6 − j3 3+ j2 2− j1 7 + j5
A= 4+ j1 5+ j3 1+ j1 B= 2+ j1 4+ j6 5+ j4
 
   
6 + j3 1− j1 8 + j9 7 − j5 5− j4 6 + j5
c 11 = (1 + j 2 )(3 + j 2 ) + (4 − j 1 )(2 + j 1 ) + (6 − j 3 )(7 − j 5 ) = 35 − j 41

c 12 = (1 + j 2 )(2 − j 1 ) + (4 − j 1 )(4 + j 6 ) + (6 − j 3 )(5 − j 4 ) = 44 − j 16
c 13 = (1 + j 2 )(7 + j 5 ) + (4 − j 1)(5 + j 4 ) + (6 − j 3 )(6 + j 5 ) = 72 + j 42
c21 = (4 + j1)(3 + j 2 ) + (5 + j 3 )(2 + j1) + (1 + j1)(7 − j 5 ) = 29 + j 24

68
Example:
c 21 = (4 + j 1 )(3 + j 2 ) + (5 + j 3 )(2 + j 1 ) + (1 + j 1 )(7 − j 5 ) = 29 + j 24
c 22 = (4 + j 1 )(2 − j 1 ) + (5 + j 3 )(4 + j 6 ) + (1 + j 1 )(5 − j 4 ) = 20 + j 41
c 23 = (4 + j 1 )(7 + j 5 ) + (5 + j 3 )(5 + j 4 ) + (1 + j 1 )(6 + j 5 ) = 37 + j73
c 31 = (6 + j 3 )(3 + j 2 ) + (1 − j 1)(2 + j 1) + (8 + j 9 )(7 − j 5 ) = 116 + j 43
c 32 = (6 + j 3 )(2 − j 1) + (1 − j 1)(4 + j 6 ) + (8 + j 9 )(5 − j 4 ) = 101 + j 15
c 33 = (6 + j 3 )(7 + j 5 ) + (1 − j 1)(5 + j 4 ) + (8 + j 9 )(6 + j 5 ) = 39 + j 144
 35 − j 41 44 − j16 72 + j 42 
[A] x[B] =  29 + j 24 20 + j 41 37 + j73 

116 + j 43 101 + j15 39 + j144

If the rows and columns of an m x n matrix are
interchanged, the resultant n x m matrix is the
transpose of the matrix and is designated by AT.
a a21 a31 
For the matrix A =  11
a12 a22 a32  2 x 3
 a11 a12 
The transpose is AT = a21 a22 
 
 a31 a32  3 x 2
69
Example:
1 3 5
A=
then, 2 4 6  2 x 3
1 2
AT =  3 4
 
5 6  3 x 2

Example:
1 + j 4 3 − j6 5 + j 2
A=
2 − j 5 4 + j1 6 − j 3
then,
1 + j 4 2 − j 5
AT =  3 − j 6 4 + j 1
 
5 + j 2 6 − j 3

70
Determinant of a 2 x 2 Matrix
The solutions of two simultaneous equations:
a11 x1 + a12 x2 = y1 (1)
a21 x1 + a22 x2 = y2 (2)
can be obtained by eliminating the variables one
at a time. Solving for x2 in terms of x1 from the second
equation and substituting this expression for x2 in the
first equation, the following is obtained:
y2 a21
x2 = − x1
a22 a22
substituting x2 and solving for x1
y 2 a21
a11 x1 + a12 ( − x1 ) = y1
a22 a22
a11 a22 x1 + a12 y 2 − a12 a21 x1 = a22 y1
( a11 a22 − a12 a21 ) x1 = a22 y1 − a12 y2
a22 y1 − a12 y2
x1 =
a11 a22 − a12 a21

71
Then, substituting x1 in either equation (1) or (2), x2 is
obtained
a11 y2 − a21 y1
x2 =
a11a22 − a12a21
The expression (a11a22 – a12a21) is the value of the
determinant of the coefficient matrix A, denoted by |A|.
a11 a12
| A| =
a21 a22
The determinant obtained by striking out the ith

row and jth column is called the minor element aij.
Example:
a11 a12 a13
a a13
a 21 a22 a 23 = 12
a32 a33
a31 a32 a33
The minor of a 21 = ( a 12 a 33 − a 32 a 13 )

72
The cofactor of an element aij designated by Aij is
Aij = ( − 1 ) i + j (minor of a ij )
Example: A21 = ( − 1 ) 2 + 1 ( the min or of a 21 )
= (-1) 3 ( the min or of a 21 )
A21 = - 1 (the minor of a 21 )
Since the minor of a 21 = ( a 12 a 23 − a 32 a 13 )
∴ the cofactor of A21 = − 1( a 12 a 33 − a 32 a 13 )

1 1 2
 1
Example: A = - 1 -2

- 6 4 2
1 1 2
A11 = ( −1 )1+1 − 1 −2 1 = 1[( −2 )( 2 ) − ( 1 )( 4 )] = −8
−6 4 2
1 1 2
A12 = ( −1 ) − 1
1+ 2
−2 1 = −1[( −1 )( 2 ) − ( 1 )( −6 )] = −4
−6 4 2

73
1 1 2
A13 = ( −1 )1+3 − 1 − 2 1 = 1[( −1 )( 4 ) − ( −2 )( −6 )] = −16
−6 4 2
1 1 2
A21 = ( −1 ) 2 +1
−1 −2 1 = −1[( 1 )( 2 ) − ( 4 )( 2 )] = 6
−6 4 2
1 1 2
A22 = ( −1 ) 2+ 2
−1 −2 1 = 1[( 1 )( 2 ) − ( 2 )( −6 )] = 14
−6 4 2

1 1 2
A23 = ( −1 ) 2 +3
−1 −2 1 = −1[( 1 )( 4 ) − ( 1 )( −6 )] = −10
−6 4 2
1 1 2
A31 = ( − 1 ) 3+1
−1 −2 1 = 1[( 1 )( 1 ) − ( −2 )( 2 )] = 5
−6 4 2
1 1 2
A32 = ( −1 ) 3+ 2
−1 −2 1 = −1[( 1 )( 1 ) − ( −1 )( 2 )] = −3
−6 4 2

74
1 1 2
A33 = ( −1) − 13+3
−2 1 = 1[(1 )( −2 ) − ( −1 )(1 )] = −1
−6 4 2
Therefore the cofactors of matrix A are:
A11 = −8 A21 = 6 A31 = 5

A12 = −4 A22 = 14 A32 = −3
A13 = −16 A23 = −10 A33 = −1

and in matrix form:

 A11 A12 A13 
A =  A21 A22 A23 
 
 A31 A32 A33 
− 8 −4 − 16
A=  6 14 − 10
 
 5 −3 − 1 

75
Inverse of a Matrix
Division does not exist in matrix algebra except in the
case of division of a matrix by a scalar. However, for a
given set of equations.
a 11 x 1 + a 12 x 2 + a 13 x 3 = y 1
a 21 x 1 + a 22 x 2 + a 23 x 3 = y 2
a 31 x 1 + a 32 x 2 + a 33 x 3 = y 3
or in matrix form [AX] = [Y]. It is desirable to

express x1, x2, and x3 a function of y1, y2, and y3, i.e..
[X] = [BY], where B is the inverse of A designated
by A-1.
Inverse of a Matrix
If the determinant of A is not zero, the equations can
be solved for x ’s as follows;
A 11 A 21 A 31
x1 = y1 + y2 + y3
| A| | A| | A|
A 21 A 22 A 32
x2 = y1 + y2 + y3
| A| | A| | A|
A 13 A 23 A 33
x3 = y1 + y2 + y3
| A| | A| | A|
where A11, A12, …, A33 are cofactors of a11, a12,,a33 and |A|
is the determinant of A.
76
Inverse of a Matrix
Thus,
 A 11 A 21 A 31 
| A | | A | | A |
  A+
B = A -1
A
=  12
A 22 A 32  or A = -1
| A | | A | | A | | A|
 A 13 A 23 A 33 
 
| A | | A | | A |
A+ is called the adjoint of A. It should be noted that the
elements of adjoint A+ are the cofactors of the elements of
A, but are placed in transposed position.
Inverse of a Matrix
Example: Get the inverse of A
1 1 2
A = - 1 -2 1 A -1 =
A+
  | A|
- 6 4 2
the Adjoint of A is
A11 A21 A31  − 8 6 5

A+ = A12 A22 A32  =  − 4 14 − 3
   
A13 A23 A33  −16 −10 −1
77
Inverse of a Matrix
The determinant of A is
1 1 2
| A |= − 1 − 2 1
−6 4 2
−2 1 −1 1 −1 −2
| A |= ( −1 )1+1 1 + ( −1 )1+2 1 + ( −1 )1+3 2
4 2 −6 2 −6 4
| A |= 1( − 2 )( 2 ) − ( 1 )( 4 )( 1 ) − ( 1 )( − 1 )( 2 )
+ ( 1 )( − 6 )( 1 ) + ( 2 )( − 1 )( 4 ) − ( 2 )( − 6 )( − 2 )
= − 4 − 4 + 2 − 6 − 8 − 24 = 2 − 46
| A |= − 44
Inverse of a Matrix
Hence, the inverse of matrix A is
 −8 6 5  −−448 6 5

 −4  −44 −44
=  −−444 
 14 − 3  14 −3
+  − 16 − 10 − 1 −44 −44 
=
A
A −1 =
A − 44  −−16
44
−10
−44
−1
−44

 −8 6 5 
= −  − 4 14 − 3
1
A −1
44
− 16 − 10 − 1
78

 y1   a11 a12 a13 a14   x1 
 y  a a22 a23 a24   x2 
 2  =  21  
 y3  a31 a 32 a33 a34   x3 
    
 0  a41 a42 a43 a44   x4 
 y1  a11 a12 a13 a14  x1 

 y  a a22 a23 a24 x2 
 2  =  21  
 y3  a31 a32 a33 a34 x3 
    
 0  a41 a42 a43 a44 x4 

Y1  A1 A2 X1 

 0  = A A4 X2 
   3
 y1  a11 a12 a13  a14  x1 
[Y1 ] =  y 2  [ A1 ] = a21 a22 a23  [A2 ] = a24  [X1 ] = x2 

 y 3  a31 a32 a33  a34  x3 
[A ] = [a
3 41
a42 a43 ] [A4 ] = [a44] [ X ] = [x ]
2 2
[Y ] = ([ A ] − [ A ][ A ] [ A ])[ X ]
1 1 2 4
−1
3 1

79
Example:  y1  1 2 3 4   x1 
 y  8 7 6 5  x2 
 2 =   
 y3  3 4 5 6   x3 
    
 0  2 4 7 8   x4 
 1 2 3 4  
 
[Y1 ] =  8 7 6  − 5[8] [2
  
−1
4 7 ][ X 1 ]
 3 4 5  6  
 

Example:
 1 2 3  4  
 1 
[Y1 ] =  8 7 6  − 5    [2
  8 
4 7 ][ X 1 ]
 3 4 5  6  
 
 1 2 3  0.5  
 
[Y1 ] =  8 7 6  − 0.625[2
  
4 7 ][ X 1 ]
 3 4 5  0.75  
 
 1 2 3  1 2 3.5  
 
[Y1 ] =  8 7 6  − 1.25
 
2.5 4.375 [ X 1 ]

 3 4 5  1.5 3 5.25  


80
The 4x4 matrix can be reduced to a

3x3matrix.
 0 0 − 0.50
∴ [Y1 ] = 6.75 4.5 1.625 [ X 1 ]
 
1.50 1.0 − 0.25

Direct Solutions
of System of Equations
Cramer’s Rule
Matrix Inversion Method
Gaussian Elimination Method
Gauss-Jordan Method

81
Solutions of System of
Equations by Cramer’ s Rule
The system of three linear equations in three
unknowns x1, x2, x3:
a11 x1 + a12 x 2 + a13 x3 = y1
a 21 x1 + a 22 x 2 + a 23 x3 = y 2
a31 x1 + a 32 x 2 + a 33 x3 = y 3
written in matrix form as :
 a11 a12 a13   x1   y 1 
a a 22 a 23   x 2  =  y 2 
 21     or AX = Y
 a 31 a 32 a 33   x3   y 3 
can be solved by Cramer’s Rule of determinants. The

determinant of coefficient matrix A is
a 11 a 12 a 13
| A |= a 21 a 22 a 23
a 31 a 32 a 33
x1 can be obtained by :
y1 a 12 a 13
y2 a 22 a 23
y3 a 32 a 33
x1 =
| A|
Note that values in the numerator are the values of the
determinant of A with the first column were replaced by the Y
vector elements.
82
Similarly, x2 and x3 can be obtained by:

a 11 y1 a 13
a 21 y2 a 23
a 31 y3 a 33
x2 =
| A|
and
a 11 a 12 y1
a 21 a 22 y2
a 31 a 32 y3
x3 =
| A|
Solutions of System of Equations by

Cramer’s Rule
Example: x 1 + x 2 + 2x 3 = 3
- x 1 − 2x 2 + x 3 = 7
- 6x 1 + 4x 2 + 2x 3 = 14
 1 1 2   x1   3 
- 1 - 2 1 x2  =  7 
    
 - 6 4 2   x 3   14 
83

Cramer’s Rule
1 1 2 1 1 2
A = - 1 -2 1 | A|= -1 −2 1 = − 44
 
- 6 4 2 −6 4 2
y1 a 12 a 13 3 1 2
y2 a 22 a 23 7 −2 1
y3 a 32 a 33 14 4 2 88
x1 = x1 = = = −2
| A| - 44 − 44


Cramer’s Rule
a 11 y1 a 13 1 3 2
a 21 y2 a 23 −1 7 1
a 31 y3 a 33 −6 14 2 44
x2 = x2 = = = −1
| A| - 44 − 44
a 11 a 12 y1  1 1 3
− 1 −2 7 
a 21 a 22 y2  
a 31 a 32 y3 x3 =  − 6 4 14 
=
− 132
=3
x3 = - 44 − 44
| A|
Therefore, x1 = - 2 x 2
= -1 x 3
=3
84
Equations by Matrix Inversion
The system of equations in matrix form can be
manipulated as follows:
AX = Y
A -1 AX = A -1Y
IX = A -1Y
X = A -1Y
Hence, the solution X can be obtained by multiplying
The inverse of the coefficient matrix by the constant
matrix Y.
Example:
x 1 + x 2 + 2x 3 = 3
- x 1 − 2x 2 + x 3 = 7
- 6x 1 + 4x 2 + 2x 3 = 14
 1 1 2
A = - 1 -2 1
 
 - 6 4 2 
85
 −8 6 5
A-1 = −  − 4 − 3
1 From slide no. 173
14
44  
− 16 − 10 − 1
 −8 6 5  3 
1 
X =A Y =− -1
−4 14 − 3  7 
44   
 − 16 − 10 − 1 14 
 − 8( 3 ) + 6( 7 ) + 5( 14 ) 
1 
X =A Y =−
-1
− 4( 3 ) + 14( 7 ) + − 3(14) 
44  
− 16( 3 ) + − 10( 7 ) + − 1( 14 )

 88   − 2 
X = A -1Y = −
1  44  =  − 1 
44    
 − 132   3 
Therefore:
x1 = - 2
x2 = - 1
x3 = 3

86
The following are the rules in matrix

manipulation:
(1) Interchange rows
(2) Multiply row by constant
(3) Add rows


Example:
x1 + x2 + 2x3 = 3 1 1 2 M 3
−1 −2 M 7
- x1 − 2x2 + x3 = 7 
1

- 6x1 + 4x2 + 2x3 = 14 −6 4 2 M 14
Add row 1 to row 2 to get row 2.
Add 6 times row 1 to row 3 to get row 3.
1 1 2 M 3
0 −1 3 M 10
 
0 10 14 M 32
87

1 1 2 M 3
0 −1 3 M 10
 
0 10 14 M 32
Multiply row 2 by 10 then add to row 3 to
obtained row 3.
1 1 2 M 3
0 −1 3 M 10 
 
0 0 44 M 132


By Back Substitution:
0x1 + 0 x2 + 44 x3 = 132
0 x1 − x2 + 3( 3 ) = 10
x1 + ( −1 ) + 2( 3 ) = 3
Therefore:
x3 = 3 x2 = - 1 x1 = -2

88
Gaussian a11 a12a13 M c1 

a a22 a23 M c2 
Elimination  21 
a31 a32 a33 M c3 
Method
⇓ Forward
elimination
a11 a12 a13 M c1 
The two phases of  a22 a'23 M c2' 
'
Gauss Elimination:  
forward elimination  a"33 M c3" 
& back substitution.
⇓
The primes indicate
the number of times x3 = c3" / a"33
that the coefficients Back
and constants have
x2 =( c'2 −a'23x3 ) / a'22 substitution
been modified. x1 =( c1 −a12x2 −a13x3 ) / a11
Gauss-Jordan Method
From Gauss Elimination Method slide
no.192
1 1 2 M 3
0 −1 3 M 10 
 
0 0 44 M 132
Multiply row 2 by -1.
1 1 2 M 3 
0 1 −3 M − 10
 
0 0 44 M 132
89
Gauss-Jordan Method
Divide row 3 by 44.
1 1 2 M 3 
0 1 −3 M − 10
 
0 0 1 M 3 
Multiply row 2 by -1 then add to row 1 to get row 1.
1 0 5 M 13 
0 1 −3 M − 10
 
0 0 1 M 3 
Gauss-Jordan Method
Multiply row 3 by -5 then add to row 1 to get row 1.
Multiply row 3 by 3 then add to row 2 to get row 2.
1 0 0 M − 2
0 1 0 M − 1
 
0 0 1 M 3 

90
Gauss-Jordan Method
Therefore: 1 0 0   x1   − 2 
0 1 0   x2  =  − 1
    
0 0 1   x 3   3 
Then, x1 = − 2
x2 = −1
x3 = 3
Gauss-Jordan Method
The Gauss-Jordan method is a variation of
Gauss Elimination. The major differences is
that when an unknown is eliminated in the
GJM, it is eliminated from all other equations
rather than just the subsequent ones.
In addition, all rows are normalized by
dividing them by their pivot elements. Thus,
the elimination steps results in an identity
matrix rather than a triangular matrix.

91
 a 11 a 12 a 13 M c1 
Gauss-Jordan a a 22 a 23 M c2 
Method  21 
 a 31 a 32 a 33 M c 3 
↓
Graphical depiction of the
Gauss-Jordan Method. 1 0 0 M c 1(n) 
The superscript (n) means 0 1 0 M c 2(n) 

that the elements of the
0 0 1 M c 3(n) 
right-hand-side vector
have been modified n ↓
times (for this case, n=3). x1 = c 1(n)
x2 = c 2(n)
x3 = c 3(n)

Iterative Solutions
Gauss Iterative Method

Gauss-Seidel Method
Newton-Raphson Method

92
Iterative Solutions
An iterative method is a repetitive process for

obtaining the solution of an equation or a system of
equation. It is applicable to system of equations
where the main-diagonal elements of the coefficient
matrix are larger in magnitude in comparison to the
off-diagonal elements.
The Gauss and Gauss-Seidel iterative techniques are
for solving linear algebraic solutions and the Newton-
Raphson method applied to the solution of non-linear
equations.
Iterative Solutions
The solutions starts from an arbitrarily

chosen initial estimates of the unknown
variables from which a new set of estimates
is determined. Convergence is achieved when
the absolute mismatch between the current
and previous estimates is less than some
pre-specified precision index for all the
variables.

93

Example:
4x 1 − x 2 + x 3 = 4
x 1 + 4x 2 + x 3 = 6
x 1 + x 2 + 3x 3 = 5
Assume a convergence index of ε = 0.001 and the
following initial estimates:
a) x1 0 = x 2 0 = x3 0 = 0.0
b) x1 0 = x 2 0 = x 3 0 = 0.5

Solution:
a) The system of equation must be expressed
in standard form.
1
x 1k + 1 = ( 4 + x 2k - x 3k )
4
1
x 2k + 1 = ( 6 - x 1k - x 3k )
4
1
x 3k + 1 = ( 5 - x 1k - x 2k )
3

94

a) with x1 0 = x 2 0 = x3 0 = 0
1
Iteration 1 (k = 0): x11 = ( 4 + 0 - 0 ) = 1.0
4
1
x 21 = ( 6 - 0 - 0 ) = 1.5
4
1
x31 = ( 5 - 0 - 0 ) = 1.6667
3
∆ x 10 = 1 − 0 = 1
∆ x 20 = 1 . 5 − 0 = 1 . 5
∆ x 30 = 1 . 6667 − 0 = 1 . 6667
max ∆ x 30 = 1.6667

Iteration 2 (k = 1):
1
x1 2 = ( 4 + 1.5 - 1.6667 ) = 0.958333
4
1
x2 2 = ( 6 - 1.0 - 1.6667 ) = 0.833333
4
1
x3 2 = ( 5 - 1.0 - 1.5 ) = 0.833333
3
∆x1 = 0.958325 − 1 = 0.041667
1
∆x21 = 0.833333 − 1.5 = 0.66667

∆x31 = 0.833333 − 1.6667 = −0.83334
max ∆x31 = 0.83334

95

Iteration 3 (k = 2): 1
x1 3 = ( 4 + 0.8333 - 0.8333 ) = 1.0
4
1
x 2 3 = ( 6 - 0.9583 - 0.8333 ) = 1.0521
4
1
x 3 3 = ( 5 - 0.9583 - 0.8333 ) = 1.0695
3
∆ x 1 = 1 − 0 . 958325 = 0 . 041667
2
∆ x 22 = 1 . 0521 − 0 . 833325 = 0 . 21877

∆ x 32 = 1 . 0695 − 0 . 8333 = 0 . 23617
max ∆ x 32 = 0.23617


1
x1 4 = ( 4 + 1.0521 - 1.0695 ) = 0.9956
4
1
x2 4 = ( 6 - 1.0 - 1.0695 ) = 0.9826
4
1
x3 4 = ( 5 - 1.0 - 1.0521 ) = 0.9826
3
∆x1 = 0.9956 − 1 = −0.0044
3
∆x23 = 0.9826 − 1.0521 = −0.0695

∆x33 = 0.9826 − 1.0695 = −0.0869
max ∆x33 = 0.0869

96

x1 5 = ( 4 + 0.9826 - 0.9826 ) = 1.0
4
1
x 2 5 = ( 6 - 0.9956 - 0.9826 ) = 1.0054
4
1
x 3 5 = ( 5 - 0.9956 - 0.9826) = 1.0073
3
∆ x 1 = 1 − 0 . 9956 = 0 . 0044
4
∆ x 24 = 1 . 0054 − 0 . 9826 = − 0 . 0228

∆ x 34 = 1 . 0073 − 0 . 9826 = 0 . 0247
max ∆ x 34 = 0.0247


x1 6 = ( 4 + 1.0054 - 1.0073 ) = 0.9995
4
1
x 2 6 = ( 6 - 1.0 - 1.0071 ) = 0.9982
4
1
x 3 6 = ( 5 - 1.0 - 1.0054 ) = 0.9982
3
∆x1 = 0.9995 − 1 = −0.0005
5
∆x 25 = 0.9982 − 1.0054 = −0.0072

∆x 35 = o .9982 − 1.0073 = −0.0091
max ∆x35 = 0.0091

97

1
x1 7 = ( 4 + 0.9982 - 0.9982 ) = 1.0
4
1
x 2 7 = ( 6 - 0.9995 - 0.9982 ) = 1.0006
4
1
x 3 7 = ( 5 - 0.9995 - 0.9982) = 1.0008
3
∆ x 1 = 1 − 0 . 9995 = 0 . 0005
6
∆ x 62 = 1 . 0006 − 0 . 9982 = 0 . 0024

∆ x 36 = 1 . 0008 − 0 . 9982 = 0 . 0026
max ∆ x 36 = 0.0026


1
x1 8 = ( 4 + 1.0006 - 1.0008 ) = 0.9995
4
1
x2 8 = ( 6 - 1.0 - 1.0008 ) = 0.9998
4
1
x3 8 = ( 5 - 1.0 - 1.0008 ) = 0.9998
3
∆x1 = 0.9995 − 1 = −0.0005
7
∆x27 = 0.9998 − 1.0006 = −0.0008

∆x37 = 0.9998 − 1.0008 = −0.0010
max ∆x37 = 0.0010
98
The Gauss iterative method has converged at

iteration 7. The method yields the following
solution.
x 1 = 0.9995
x 2 = 0.9998
x 3 = 0.9998


b) with x 1 0 = x 2 0 = x 3 0 = 0.5
x1 1 =
x21 =
x31 =
∆ x 10 =
∆ x 20 =
∆ x 30 =
max ∆ x =

99

x1 2 =
x2 2 =
x3 2 =
∆x11 =
∆x21 =
∆x31 =
max ∆x 1 =

x1 3 =
x2 3 =
x3 3 =
∆x12 =
∆x22 =
∆x32 =
max ∆x 2 =
100

x1 4 =
x2 4 =
x3 4 =
∆x13 =
∆x23 =
∆x33 =
max ∆x 3 =

x1 5 =
x2 5 =
x3 5 =
∆x14 =
∆x24 =
∆x34 =
max ∆x 4 =

101

x1 6 =
x2 6 =
x3 6 =
∆x15 =
∆x25 =
∆x35 =
max ∆x 5 =

x1 7 =
x2 7 =
x3 7 =
∆x16 =
∆x26 =
∆x36 =
max ∆x 6 =
102

x18 =
x2 8 =
x3 8 =
∆x17 =
∆x27 =
∆x37 =
max ∆x7 =

x 1
=
x 2
=
x 3
=
Note: Number of iterations to achieve

convergence is also dependent on
initial estimates

103

Given the system of algebraic equations,
a 11 x 1 + a 12 x 2 + L + a 1n x n = y 1
a 21 x 1 + a 22 x 2 + L + a 2n x n = y 2
↓ ↓ ↓ ↓
a 31 x 1 + a 32 x 2 + L + a nn x n = y 3
In the above equation, the x’s are unknown.


From the first equation,
a 11 x 1 = y 1 − a 12 x 2 K − a 1n x n
1
x1 = ( y 1 − a 12 x 2 K − a 1n x n )
a 11
Similarly, x2, x3…xn of the 2nd to the nth equations can
be obtained.
1
x2 = (b 2 − a 21 x 1 − a 23 x 3 − K a 2n x n )
a 22
↓ ↓ ↓ ↓
1
xn = ( bn − a n1 x 1 − a n2 x 2 − K − a n,n -1 x n -1 )
a nn
104
In general, the jth equation may be written

as
1
xj = ( b j − ∑ i = 1 a ji x i )
n
equation “a”
a jj i≠ j
j = 1, 2,K n


In general, the Gauss iterative estimates are:
k +1 y1 a a a
= − 12 x 2 − 13 x 3 − ... − 1n x n
k k k
x1
a 11 a 11 a 11 a 11
k +1 y2 a k a k a k
x2 = − 21 x 1 − 23 x 3 − ... − 2n x n
a 22 a 22 a 22 a 22
k +1 yn a a a
= − n1 x 1 − n2 x 2 − ... − n, n - 1 x n - 1
k k k
xn
a nn a nn a nn a nn
where k is the iteration count
105

From an initial estimate of the unknowns (x10,
x20,…xn0), updated values of the unknown variables
are computed using equation “a”. This completes one
iteration. The new estimates replace the original
estimates. Mathematically, at the kth iteration,
1
x kj + 1 = ( b j − ∑ i=1 a
n
ji
xk ) equation “b”
a jj i≠ j
i
j = 1, 2,K n

A convergence check is conducted after each
iteration. The latest values are compared with their
values respectively.
∆ x k = x kj + 1 − x kj equation “c”
j = 1, 2,K n
The iteration process is terminated when
max | ∆ x kj | < ε (convergen t)
k = itermax (non - convergent )
106
Gauss-Seidel Method
Example: Solve the system of equations using the
Gauss-Seidel method. Used a
convergence index of ε = 0.001
4x 1
− x2 + x3 = 4
x 1 + 4x 2
+ x3 =6
x 1 + x 2 + 3x 3
= 5
x 1 0 = x 2 0 = x 3 0 = 0.5
Gauss-Seidel Method
Solution:
a) The system of equation must be expressed
in standard form.
1
x 1k + 1 = ( 4 + x 2k - x 3k )
4
1
x 2k + 1 = ( 6 - x 1k + 1 - x 3k )
4
1
x 3k + 1 = ( 5 - x 1k + 1 - x 2k + 1 )
3
107
Gauss-Seidel Method
with x 1 0 = x 2 0 = x 3 0 = 0.5
Iteration 1 (k =0): 1
x11 = ( 4 + 0.5 - 0.5 ) = 1.0
4
1
x 21 = ( 6 - 1.0 - 0.5 ) = 1.125
4
1
x31 = ( 5 - 1.0 - 1.125 ) = 0.9583
3
∆ x 10 = 1 − 0 . 5 = 0 . 50
∆ x 20 = 1 . 125 − 0 . 50 = 0 . 625
∆ x 30 = 0 . 9583 − 0 . 50 = 0 . 4583
max | ∆ x 20 | = 0.625
Gauss-Seidel Method
1
x1 2 = ( 4 + 1.125 - 0.9583 ) = 1.0417
4
1
x2 2 = ( 6 - 1.0417 - 0.9583 ) = 1.0
4
1
x3 2 = ( 5 - 1.0417 - 1.0 ) = 0.9861
3
∆x1 = 1.0417 − 1 = 0.0417
1
∆x21 = 1 − 1.125 = −0.125

∆x31 = 0.9861 − 0.9583 = 0.0323
max | ∆x21 | = 0.125

108
Gauss-Seidel Method
1
x1 3 = ( 4 + 1.0 - 0.9861 ) = 1.0035
4
1
x 2 3 = ( 6 - 1.0035 - 0.9891 ) = 1.0026
4
1
x 3 3 = ( 5 - 1.0035 - 1.0026 ) = 0.9980
3
∆ x1 = 1 .0035 − 1 .0417 = − 0 .0382
2
∆ x 22 = 1 .0026 − 1 = 0 .0026
∆ x 32 = 0 .9980 − 0 .9861 = 0 .0119
max | ∆ x 32 | = 0.0119

Gauss-Seidel Method
1
x1 4 = ( 4 + 1.0026 - 0.9980 ) = 1.0012
4
1
x2 4 = ( 6 - 1.0012 - 0.9980) = 1.0002
4
1
x3 4 = ( 5 - 1.0 - 1.0012 - 1.0002) = 0.9995
3
∆x13 = 1.0012 − 1.0035 = 0.0023
∆x23 = 1.0002 − 1.0026 = −0.0024
∆x33 = 0.9995 − 0.9980 = 0.0015
max | ∆x23 | = 0.0024
109
Gauss-Seidel Method
1
x1 5 = ( 4 + 1.0002 - 0.9995 ) = 1.0002
4
1
x 2 5 = ( 6 - 1.0002 - 0.9995) = 1.0001
4
1
x 3 5 = ( 5 - 1.0002 - 1.0001) = 0.9999
3
∆ x14 = 1 .0002 − 1 .0012 = −0 .001
∆ x 24 = 1 .0001 − 1 .0002 = −0 .0001
∆ x 34 = 0 .9999 − 0 .9995 = 0 .0004
max | ∆ x 4 | = 0.001 < ε

Gauss-Seidel Method
The Gauss-Seidel Method has converged after 4

iterations only with the following solutions:
x 1 = 1.0002
x 2 = 1.0001
x 3 = 0.9999

110
Gauss-Seidel Method
The Gauss-Seidel method is an improvement over the
Gauss iterative method. As presented in the previous
section, the standard form of the jth equation may be
written as follows.
1
xj = (bj − ∑
n
i=1
a ji x i ) j = 1, 2, K n
a jj i≠ j
From an initial estimates (x10, x20,…xn0), an updated

value is computed for x1 using the above equation with j
set to 1.This new value replaces x10 and is then used
together with the remaining initial estimates to compute
a new value for x2. The process is repeated until a new
estimate is obtained for xn. This completes one iteration.

Gauss-Seidel Method
Note that within an iteration, the latest computed
values are used in computing for the remaining
unknowns. In general, at iteration k,
1 α
x k +1 = (bj − ∑
n
i=1
a ji
xi )
j
a jj i≠ j
j = 1, 2, K n
where α = k if i > j
= k + 1 if i < j
After each iteration, a convergence check is
conducted. The convergence criterion applied is the
same with Gauss Iterative Method.
111
Gauss-Seidel Method
An improvement to the Gauss Iterative Method
k +1 y1 a k a k
x1 = − 12 x 2 − ... − 1n x n
a 11 a 11 a 11
k +1 y a k +1 a k
x2 = 2 − 21 x 1 − ... − 2n x n
a 22 a 22 a 22
k+1 y aij k+1 a k+1 a k+1 a k+1
xi = i − xi −...− i,i-1 xi-1 − i,i+1 xi+1 − in xn
aii aii aii aii aii
k +1 yn a k +1 a k +1
xn = − n1 x 1 − ... − n, n - 1 x n - 1
a nn a nn a nn
Newton Method
Example:
Find the root of f ( x ) = e−x − x
Note: the equation is non-linear and can not be solved by the
direct and iterative methods for solving linear equations.
A non-linear equation can be expanded using Taylor

Series
f " ( xi )
f ( x i + 1 ) = f ( x i ) + f ' ( x i )(∆ x ) + (∆ x )2 + ...
2!
f n ( xi )
+ ... + (∆ x )n
n! ∆ x = xi +1 − xi
112
Newton Method
f(x)
f ( xi +1 ) ≅ f ( xi ) + f ' ( xi )(xi+1 − 1)
f(xi)
0 X At the intersecti on with

Xi
the x - axis, f ( x i + 1 ) = 0
Xi+1
(xi – xi+1)
f (x i )
xi +1 = xi −
f ' (xi )

Newton Method
The first order Taylor Series expansion of f ( x ) = −e − x − x
can be obtained by getting the first derivative
f ' ( x ) = −e − x − 1
At the intersection of the slope (i.e., derivative) with
the original function, f ( x i + 1 ) = 0 , hence
f ( x i + 1 ) = f ( x i ) + f ' ( x i )( x i − x i + 1 )
0 = f ( x i ) + f ' ( x i )( x i − x i + 1 )
f ( xi )
x i+1 = xi −
f ' ( xi )
113
Newton Method
e − xi − x i xi+1 = xi −
f ( xi )
Substituting, x i +1 = xi − f ' ( xi )
− e − xi − 1
Using initial guess x0=0
e −0 − 0 1
x1 = 0 − −0
=− = 0 .5 ∆ x = 0 .5 − 0 = 0 . 5
−e −1 −1−1
The next estimate is
e − 0 .5 − 0 . 5
x 2 = 0 .5 − = 0 .566311
− e − 0 .5 − 1
∆ x = 0 . 566311 − 0 . 50 = 0 . 066311
Newton Method
x3 = ∆ x = _________ − ________ =
x4 = ∆ x = _________ − ________ =
x5 = ∆ x = _________ − ________ =
The true value of the root is 0.56714329

114
Example: Solve the non-linear equation
x12 − 4x2 = 4
2x1 − x2 = 2
use: x1 = 1, x2 = −1
0 0
Solution: First, form the Jacobian

∂f 1 ∂f 1
= 2 x1 = −4
f 1 = x 12 − 4 x 2 ∂ x1 ∂x2
f 2 = 2 x1 − x 2 ∂f2 ∂f2
= 2 = −1
∂ x1 ∂x2
In Matrix form
 ∂f 1 ( x ) ∂f 1 ( x 0 ) 
0
 y1 − f 1 ( x )  ∂x1 ∂x2   ∆x1 

0
 y − f ( x ) =  ∂f ( x 0 ) 
∂f 2 ( x 0 )  ∆x2 
 2 2 0   2
 ∂x1 ∂x2 
Jacobian Matrix
4 − ( x12 − 4 x2 )  2 x1 - 4   ∆x1 
 2 − ( 2 x − x ) =
 1 2 
 2
 − 1 ∆x2 
115
Iteration 0:
f 1 ( x 0 ) = 12 − 4( −1 ) = 5 , y1 = 4
∂f 1
= 2(1) = 2
∂x1
∂f 1
= −4 f 2 ( x 0 ) = 2( 1 ) − ( − 1 ) = 3 , y 2 = 2
∂x2
∂f 2
=2
∂x1
∂f 2
= −1
∂x 2
The equations are: In matrix form:
4 − 5 = ( 2 )∆x1 + ( −4 )∆x2
0 0
 − 1  2 − 4   ∆ x10 
 − 1 =  2  
− 1   ∆ x 2 0 
2 − 3 = ( 2 )∆x1 + ( −1 )∆x2
0 0
  
Solving,
∆ x1 = − 0 .5
0
∆ x = 0
0
1
2
Thus, x1 = 1 + ( −0.5 ) = 0.5
1
x2 = −1 + 0 = −1

116
Repeating the process with the new estimates,
Iteration 1:
f 1 ( x1 ) = ( 0.5 )2 − 4( −1 ) = 4.25 , y1 = 4
∂f 1 ( x1 )
= 2( 0.5 ) = 1.0
∂x1
∂f 1 ( x1 ) f 2 ( x 1 ) = 2 ( 0 .5 ) − ( − 1 ) = 2 , y2 = 2
= −4 ∂f 2 ( x 1 )
∂x2 = 2
∂x1
∂f 2 ( x 1 )
= −1
∂x2

The equations are: In matrix form:
4 − 4.25 = ∆x1 − 4 ∆ x 2  − 0.25   1 − 4   ∆ x1 1 

1 1
 0  = 2  
− 1   ∆x 2 1 
2 − 2 = 2 ∆ x1 − ∆ x 2
1 1
  
Solving,
∆ x 1 = 0 . 03571
1
∆ x 2 = 0 . 07143
1
Thus,
x 1 = 0 .5 + 0 .03571 = 0 .53571
2
x 2 = − 1 + 0 .07143 = − 0 .92857
2

117
Repeating the process with the new estimates,
Iteration 2:
f 1 ( x 2 ) = ( 0.53571 )2 − 4( −0.92857 ) = 4.001265 , y1 = 4

∂f 1 ( x 2 )
= 2( 0.53571 ) = 1.07142
∂x1
∂f 1 ( x 1 )
= −4 f 2 ( x 2 ) = 2( 0.53571 ) − ( −0.92857 ) = 1.99999 ≅ 2
∂x2
∂f 2 ( x 2 )
=2 y2 = 2
∂x 1
∂f 2 ( x 2 )
= −1
∂x 2
The equations are:
4 − 4 . 001265 = 1 . 07142 ∆ x 1 − 4 ∆ x 2
2 2
2 − 2 .0 = 2 ∆ x1 − ∆ x 2
2 2 In matrix form:
 − 0 .001265  1 .07142 − 4   ∆ x1 2 
 =  
− 1   ∆ x 2 2 
 0   2
Solving,
∆ x 1 = − 0 . 00018
2
∆ x 2 = 0 . 00036
2
Thus, x 1 = 0 . 53571 − 0 . 00018 = 0 . 53553

3
x 2 = − 0 . 92857 − 0 . 00035 = − 0 . 92893

3

118
Substituting to the original equation:
f 1 ( x1 ) = ( 0.53553 )2 − 4( −0.92893 ) = 4.0025124 , y1 = 4

3
f 2 ( x1 ) = 2( 0.53553 ) − ( −0.92893 ) = 1.99928 , y 2 = 2

3
y 1 − f 1 = − 0 . 0025
y 2 − f 2 = 0 . 00072 Therefore,
x 1 = 0 .53553
x 2 = − 0 .92893
Note the rapid convergence of the Newton-Raphson
Method.
The Newton-Raphson method is applied when the
system of equations is non-linear.
Consider a set of n non-linear equations in n unknowns.
y 1 = f 1 ( x 1 , x 2 ,K , x n )
y 2 = f 2 ( x 1 , x 2 ,K , x n )
M
y n = f n (x 1 , x 2 , K , x n )

119
The system of non-linear equations can be linearized
using Taylor’s Series
∂f1 0 ∂f ∂f
y1 = f1 ( x0 ) + ( x )∆x1 + 1 ( x0 )∆x2 + K + 1 ( x0 )∆xn
0 0 0
∂x1 ∂x2 ∂xn

∂f 2 0 ∂f ∂f
y2 = f 2 ( x0 ) + ( x )∆x1 + 2 ( x0 )∆x2 + K+ 2 ( x0 )∆xn
0 0 0
∂x1 ∂x2 ∂xn
M M M M
∂f n 0 ∂f ∂f
yn = f n ( x0 ) + ( x )∆x1 + 1 ( x0 )∆x2 + K+ 1 ( x0 )∆xn
0 0 0
∂x1 ∂x2 ∂xn

Where:
X0 = (x10,x20, …, xn0)
= set of initial estimates
fi(x0) = the function fi (x1,x2, …, xn)

evaluated using the set of initial
estimates.
∂f i ( x 0 )
= the partial derivatives of the
∂x j function fi(x1,x2,…,xn) evaluated using
the set of original estimates.

120
The equation may be written in matrix form as
follows:
f 1 ( x 0 )   1∂x1   ∆x 0 
∂f ( x o ) ∂f 1 (x 0 ) ∂f 1 (x 0 )
 y1 − ∂x 2 K ∂x n
   1
 0
∂f 2 ( x 0 ) ∂f 2 ( x 0 )  0
 y2 − f 2 ( x 0 )  ∂f 2∂(xx ) K   ∆ x 
= 1 ∂x 2 ∂x n
 M 
2
 M 
   M0 M M 
 0
∂f n ( x 0 ) 0
 y n − f n ( x 0 )  ∂f n∂(xx ) K ∂f n∂(xnx )  ∆xn 
 1 ∂x 2
The matrix of partial derivatives is known as the

Jacobian. The linearized system of equations may be
solved for ∆x’s which are then used to update the initial
estimates.
At the kth iteration:
k =1
x j = x j + ∆x j j = 1, 2,...,n
k k
Convergence is achieved when
y j − f j ( xk ) ≤ ε 1 j = 1,2,...,n
or
∆x j ≤ ε 2 j = 1, 2, ..., n
k
Where ε1 and ε2 are pre-set precision indices.

121

122

CP1 B8 Lecture No. 1 - Fundamental Principles & Methods PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CP1 B8 Lecture No. 1 - Fundamental Principles & Methods PDF

Uploaded by

Copyright:

Available Formats

Competency Training and Certification Program in Electric Power Distribution System Engineering

Fundamental Principles and

U. P. NATIONAL ENGINEERING CENTER

Training Course in Fundamental Principles and Methods in Power System Analysis 2

1. Circuit Conventions and Notations

U. P. National Engineering Center Competency Training & Certification Program in Electric

Circuit Conventions & Notations

Voltage and Current Directions

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 4

Circuit Conventions & Notations

U. P. National Engineering Center Competency Training & Certification Program in Electric

Circuit Conventions & Notations

Training Course in Fundamental Principles and Methods in Power System Analysis 6

Circuit Conventions & Notations

U. P. National Engineering Center Competency Training & Certification Program in Electric

Circuit Conventions & Notations

Training Course in Fundamental Principles and Methods in Power System Analysis 8

Circuit Conventions & Notations

Circuit Conventions & Notations

Training Course in Fundamental Principles and Methods in Power System Analysis 10

Circuit Conventions & Notations

Power System Representation

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 12

Power System representation

U. P. National Engineering Center Competency Training & Certification Program in Electric

Power System representation

3-phase wye neutral ungrounded

3-phase delta connection

3-phase wye neutral grounded

Current Transformer V Voltmeter

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 14

Power System Representation

Power System Representation

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 16

Power System Representation

U. P. National Engineering Center Competency Training & Certification Program in Electric

Power System Representation

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 18

Power System Representation

6x6 Admittance Matrix

U. P. National Engineering Center Competency Training & Certification Program in Electric

Power System Representation

T&D Lines can be represented by an infinite series of

U. P. National Engineering Center Competency Training & Certification Program in Electric

Training Course in Fundamental Principles and Methods in Power System Analysis 20

Power System Representation

Power System Representation

Training Course in Fundamental Principles and Methods in Power System Analysis 22

Power System Representation

Power System Representation

1/2 Yba Ybb Ybc 1/2 Yba Ybb Ybc

Training Course in Fundamental Principles and Methods in Power System Analysis 24

Power System Representation

Ebo= |E| ∠240° V

U. P. National Engineering Center Competency Training & Certification Program in Electric

Power System Representation

Training Course in Fundamental Principles and Methods in Power System Analysis 26

Power System Representation