Professional Documents
Culture Documents
ISA Transactions
journal homepage: www.elsevier.com/locate/isatrans
Research note
article info a b s t r a c t
Article history: In this work, the normalized Internal Model Control (IMC) filter time constant is designed to achieve
Received 30 May 2008 a specified value of the maximum sensitivity for stable first and second order plus time delay process
Received in revised form models, respectively. Since a particular value of the maximum sensitivity results in an almost constant
20 August 2008
percentage overshoot to controller setpoint change, an empirical relationship between the normalized
Accepted 15 September 2008
Available online 9 October 2008
IMC filter time constant and percentage overshoot is presented. The main advantage of the proposed
method is that only a user-defined overshoot is required to design a PI/PID controller. Simulation
Keywords:
examples are given to demonstrate the value of the proposed method.
Internal model control © 2008 ISA. Published by Elsevier Ltd. All rights reserved.
Maximum sensitivity
PID Controller
Robustness
0019-0578/$ – see front matter © 2008 ISA. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.isatra.2008.09.002
A. Ali, S. Majhi / ISA Transactions 48 (2009) 10–15 11
Table 1 Table 2
Controller parameters for various process models Ms∗ , Ms , percentage overshoots and λn
Gp (s) Kp Ti Td Ms∗ Ms Overshoot (%) λn
K e−θ s 1 τ τ
τ s+ 1 K λ+θ
= 1
θ K λn +1
τ 1.2 1.2 0 3.5568
K e−θ s 1 2ξ τ 1 2ξ τ τ 1.3 1.3 0 2.1384
τ 2 s2 +2ξ τ s+1 K λ+θ
= θ K λn +1
2ξ τ 2ξ 1.4 1.4 0.17 1.4458
1.5 1.6 3.48 1.0375
1.6 1.7 8.79 0.7691
2. Controller design 1.7 1.8 14.51 0.5796
shows the IMC-based PI/PID controller settings for the FOPTD and
SOPTD process models, where λn = λ/θ . The loop transfer function Nyquist plots given in [9]. Therefore, the condition that ω2 should
L(s) using (1) and (3) or (2) and (4) becomes have repeated roots gives
L(s) = Gp (s)Gc (s)
4c 2 λ2 + λ(8c 2 θ + 4θ c − 8θ ) + 4c θ 2 (1 + c ) − 7θ 2 = 0. (10)
e−θ s
= Dividing the above equation by θ , (10) becomes
2
(λ + θ )s
e−θ s 4c 2 λ2n + λn (8c 2 + 4c − 8) − 7 + 4c + 4c 2 = 0. (11)
= (5)
zs The solution of (11) gives
where z = λ + θ . Consequently, the sensitivity function S (s) √
2 + 2 1 − c − c − 2c 2
becomes λn = . (12)
2c 2
1
S (s) = The λn obtained by solving (11) for various values of c and the
1 + L(s)
corresponding percentage overshoots are given in Table 2. Ms is
zs the true value of the maximum sensitivity corresponding to the
= . (6)
zs + e−θ s obtained set of controller parameters and is calculated using the
robust control toolbox of MATLAB. The difference between Ms∗ and
Substituting the delay term by a 1/1 Padé approximation, the
Ms is because of the 1/1 Padé approximation of the delay term.
sensitivity function can be written in the frequency domain as
An important point to be observed is that the overshoot remains
−θ ω2 z + j2ωz constant for a particular value of Ms for a wide range of FOPTD and
S (jω) = . (7) SOPTD processes, respectively. Analytical expressions correlating
−ω θ z + 2 + jω(2z − θ )
2
Ms and λn to percentage overshoot (ov ) obtained using the curve
Squaring the magnitude of both sides of (7), we get fitting toolbox are
where y = = 1 1
is the inverse of the user specified
λn = −47.89 × 10−5 × (ov)3
s M∗ kS (jw)k ∞
maximum sensitivity. Simplification of (8) gives + 0.0144 × (ov)2 − 0.1698 × (ov) + 1.474. (14)
ω4 θ 2 z 2 c 2 + ω2 (4z 2 c 2 + 8θ z − θ 2 ) − 4 = 0 (9) The normalized IMC filter parameter and the corresponding
p maximum sensitivity for smooth and tight control are obtained by
where c = 1 − y2 . For y to be the minimum distance of the calculating the respective values for 0 and 10 percent overshoot
Nyquist curve from the critical point, the Nyquist curve of the and are given in Table 3. The proposed tuning rules for stable
loop transfer function should touch the circle with centre (−1, 0) FOPTD and SOPTD process models are given in Tables 4 and 5,
and radius y at only one frequency as is evident from the various respectively.
12 A. Ali, S. Majhi / ISA Transactions 48 (2009) 10–15
Fig. 1. Step responses for G1 (s), G2 (s), G3 (s) and G4 (s): (–) smooth control, (. . . ) tight control, (- -) SIMC.
A. Ali, S. Majhi / ISA Transactions 48 (2009) 10–15 13
Fig. 2. Step responses for G5 (s) and G6 (s): (–) smooth control, (. . . ) tight control, (- -) SIMC, (—·—) Wang et al.
2.1. Limitations of IMC-based controller design SOPTD processes to demonstrate the effectiveness of the proposed
tuning method. The performance of the proposed controller is
As the IMC approach is based on pole zero cancellation, methods compared with the IMC-based approach proposed in [5]. The
which incorporate IMC design principles result in good set point results are also compared with Wang et al.’s [10] method for SOPTD
responses. However, the IMC settings result in a long settling time process models. The PID controller is implemented in the widely
for the load disturbances for lag dominant processes which is used parallel form given by
not desirable in the control industry. A good trade off between
disturbance response and robustness is achieved by selecting 1 sTd
Gc (s) = Kp 1+ + (16)
Ti = 8θ for lag dominant processes [5]. As suggested in [5], the sTi 0.1sTd + 1
integral time constant is therefore taken as Ti = min(T , 8θ ) for 1
the PI controller and Ti = min(2ξ τ , 8θ ) for the PID controller. G1 (s) = e− s
Even though the modified integral settings degrades the setpoint s+1
performance, it is observed that the proposed controller settings 1
G2 (s) = e−5s
gives satisfactory performance for both setpoint tracking and load s+1
disturbance rejection for lag dominant processes. 100
Extensive simulation results have shown that the closed loop G3 (s) = e− s
100s + 1
performance of second order processes with θ ξ /τ less than 0.6 can
be improved by the following settings of the proportional gain of 3
G4 (s) = e−10s . (17)
the PID controller. 100s + 1
0.2 ξ τ θξ The FOPTD process models considered in [11] are given by
Kp = if 0.01 < < 0.1
K θ λn + 1 τ (17). The time constant and the plant delay are the same for G1 (s)
1.2 ξτ θξ whereas G2 (s) represents a delay dominated plant. G3 (s) is a lag
Kp = if 0.1 ≤ < 0.6. (15) dominant process with θ /τ ratio of 0.01 and G4 (s) represents the
K θ λn + 1 τ transfer function of an important viscosity loop in a polymerization
process. Fig. 1 shows the plant output for a unit change in the step
3. Simulation study input and a unit step change in the load disturbance. The controller
tuned for smooth control gives plant output with no overshoot for
The proposed tuning scheme is applied to FOPTD process G1 (s) and G2 (s) whereas settings for tight control yield faster and
models with various ratios of dead time to time constant and oscillatory output. The SIMC gives a response which is in between
14 A. Ali, S. Majhi / ISA Transactions 48 (2009) 10–15
Fig. 3. Step responses for G5 (s): smooth control for nominal (–) and perturbed (. . . ) process model, Chen et al. for nominal (- -) and perturbed (—·—) process model.
Fig. 4. Step responses for G7 (s) for zero percent overshoot: (–) proposed, (. . . ) Mnif.
these two outputs as is evident from Fig. 1. Unlike the SIMC, the overshoot and satisfactory load disturbance rejection for G5 (s).
proposed method can be used to obtain a family of responses The controller tuned for tight control gives a faster response with
starting from the smooth response to tight control. As G3 (s) and less overshoot for G5 (s) as compared to Wang et al.’s PID as is
G4 (s) represent lag dominant processes, the IMC controller settings evident from Fig. 2. The load disturbance response is the same
result in poor load disturbance rejection and hence the integral for all the three methods. As the θ ξ /τ ratio for G6 (s) is 0.02,
time constant suggested in [5] is considered. It can be observed the modified proportional gain suggested in Section 2 is used.
from Fig. 1 that this degrades the system setpoint performance Satisfactory performance is achieved for the oscillatory plant G6 (s)
and the overshoot is not zero for G3 (s) and G4 (s). However, the by the proposed method whereas Wang et al.’s PID gives large
proposed controller gives satisfactory performance for both the overshoot and long settling time which is not desirable in the
setpoint tracking and disturbance rejection even for these lag control industry.
dominant processes, respectively. It can be thus concluded that the The plant model G5 (s) is considered again and the results
proposed tuning method gives satisfactory performance for a wide are compared with Chen et al.’s [4] method to illustrate the
range of stable FOPTD process models. robustness of the proposed tuning method. The controller settings
The SOPTD plant models considered are given by the following corresponding to gain and phase margins of 3.14 and 61.4◦ (as
transfer functions: these are the best settings reported in [4] from robustness point
of view) are considered in this work. The step responses for the
1 nominal plant model and an assumed 20% uncertainty in both
G5 (s) = e−0.5s (18)
(s + 1)(s + 5)2 the steady state gain and the delay are shown in Fig. 3. It can be
concluded from Fig. 3 that the robustness of the proposed method
1
G6 (s) = e−0.1s (19) towards assumed parameter perturbations are quite satisfactory.
( + 0.4s + 1)
s2 Finally, the thermoelectric device TB-127-1.4-1.2 (Kyrotherm)
1 which is used in biomedical Thermocycler for PCR analysis
Gˆ5 (s) = e−0.606s . (20) and does not tolerate any overshoot in the output response is
7. 724s2 + 32.317s + 25.220 considered. A simplified model of the device is given by
G5 (s) represents a higher order non-oscillatory process and the
7
approximate second order plus time delay model is given by (20). G7 (s) = e−16s . (21)
The second example is an oscillatory plant with ξ = 0.2. The 100s + 1
plant outputs and the corresponding control signals are shown For a fair comparison, the results are compared with Mnif’s
in Fig. 2. The proposed smooth controller gives zero percentage method [12] because both the proposed and Mnif’s methods
A. Ali, S. Majhi / ISA Transactions 48 (2009) 10–15 15
Fig. 5. Step responses for G7 (s) for 5% overshoot: (–) proposed, (. . . ) Mnif.
gives the controller parameters such that the response of the user defined value of the percent overshoot. The usefulness of the
compensated system has overshoot close to a prescribed value. proposed method is illustrated by several simulation examples.
The controller parameters proposed in [12] are Kp = 0.31 and
Ti = 100 whereas the proposed method gives Kp = 0.36 and References
Ti = 100. The performances of the closed loop system is evaluated
using these controller settings by giving a unit step input in the [1] O’Dwyer A. Handbook of PI and PID controller tuning rules. 2nd edn. Imperial
set point and a negative step input of 0.50 at t = 300 s. It college press; 2006.
[2] Rivera DE, Morari M, Skogestad S. Internal model control. 4. PID controller
can be observed from the system outputs shown in Fig. 4 that design. Ind Eng Chem Process Des Dev 1986;25:252–65.
the proposed method results in superior performance for both [3] Liu K, Shimizu T, Inagaki M, Ohkawa A. New tuning method for IMC controller.
the setpoint and disturbance rejection. The step responses for 5% Chem Eng Japan 1998;31(3):320–4.
[4] Chen C, Huang HP, Hsieh CT. Tuning of PI/PID controllers based on specification
overshoot specification are shown in Fig. 5. Exactly 5% overshoot
on maximum closed loop amplitude ratio. Chem Eng Japan 1999;32(6):783–8.
is obtained by the proposed method whereas Mnif’s method gives [5] Skogestad S. Simple analytic rules for model reduction and PID controller
7% overshoot, thereby proving the accuracy of the proposed tuning tuning. J Process Control 2003;13:291–309.
formulas. [6] Ho WK, Hang CC, Cao L. Tuning of PID controllers based on gain and phase-
margin specifications. Automatica 1995;31(3):497–502.
[7] Ho WK, Xu W. PID tuning for unstable processes based on gain and phase
4. Conclusion margin specifications. IEE Control Theory Appl 1998;145(5):392–6.
[8] Skogestad S. Tuning for smooth PID control with acceptable disturbance
The sufficiency of the maximum sensitivity as a tuning param- rejection. Ind Eng Chem Res 2006;45:7817–22.
[9] Åström KJ, Panagopoulos H, Hägglund T. Design of PI controllers based on non-
eter for controller design is illustrated in this paper and guidelines convex optimization. Automatica 1998;34(5):585–601.
are provided regarding the selection of this tuning parameter for [10] Wang QG, Lee TH, Fung HW, Bi Q, Zhang Y. PID tuning for improved
smooth and tight control for stable FOPTD and SOPTD process mod- performance. IEEE Trans Control Syst Technol 1999;7(4):457–65.
[11] Shamsuzzoha M, Lee M. An enhanced performance PID filter controller for first
els, respectively. Furthermore, an empirical relationship between order time delay processes. Chem Eng Japan 2007;40(6):501–10.
the percentage overshoot and the normalized IMC filter parameter [12] Mnif F. New tuning rules of PI-like controllers with transient performances for
is presented to facilitate controller tuning corresponding to any monotonic time-delay systems. ISA Trans 2008;47(4):401–6.