optimality of pid control for process control applications
DESCRIPTION
Optimality of PID control for process control applications. Sigurd Skogestad Chriss Grimholt NTNU, Trondheim, Norway. AdCONIP, Japan, May 2014. TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: A A A A A A A A A. Trondheim, Norway. Operation hierarchy. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/1.jpg)
1
Optimality of PID control for process control applications
Sigurd SkogestadChriss Grimholt
NTNU, Trondheim, Norway
AdCONIP, Japan, May 2014
![Page 2: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/2.jpg)
3
Trondheim, Norway
![Page 3: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/3.jpg)
4
![Page 4: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/4.jpg)
5
Operation hierarchy
CV1
MPC
PID
CV2
RTO
u (valves)
![Page 5: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/5.jpg)
6
Outline
1.Motivation: Ziegler-Nichols open-loop tuning + IMC
2.SIMC PI(D)-rule
3.Definition of optimality (performance & robustness)
4.Optimal PI control of first-order plus delay process
5.Comparison of SIMC with optimal PI
6.Improved SIMC-PI for time-delay process
7.Non-PID control: Better with IMC / Smith Predictor? (no)
8.Conclusion
![Page 6: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/6.jpg)
7
PID controller
•Time domain (“ideal” PID)
•Laplace domain (“ideal”/”parallel” form)
•For our purposes. Simpler with cascade form
•Usually τD=0. Then the two forms are identical.
•Only two parameters left (Kc and τI)
•How difficult can it be to tune???– Surprisingly difficult without systematic approach!
e
![Page 7: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/7.jpg)
8
Trans. ASME, 64, 759-768 (Nov. 1942).
Disadvantages Ziegler-Nichols:1.Aggressive settings2.No tuning parameter3.Poor for processes with large time delay (µ)
Comment:Similar to SIMC for integrating process with ¿c=0:Kc = 1/k’ 1/µ¿I = 4 µ
![Page 8: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/8.jpg)
9
Disadvantage IMC-PID:1.Many rules2.Poor disturbance response for «slow» processes (with large ¿1/µ)
![Page 9: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/9.jpg)
10
Motivation for developing SIMC PID tuning rules
1.The tuning rules should be well motivated, and preferably be model-based and analytically derived.
2.They should be simple and easy to memorize.
3.They should work well on a wide range of processes.
![Page 10: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/10.jpg)
11
2. SIMC PI tuning rule1.A
pproximate process as first-order with delay• k = process gain• ¿1 = process time constant• µ = process delay
2.Derive SIMC tuning rule:
Reference: S. Skogestad, “Simple analytic rules for model reduction and PID controller design”, J.Proc.Control, Vol. 13, 291-309, 2003
c ¸ - : Desired closed-loop response time (tuning parameter)
Open-loop step response
Integral time rule combines well-known rules:IMC (Lamda-tuning): Same as SIMC for small ¿1 (¿I = ¿1)Ziegler-Nichols: Similar to SIMC for large ¿1 (if we choose ¿c= 0)
![Page 11: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/11.jpg)
12
Derivation SIMC tuning rule (setpoints)
![Page 12: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/12.jpg)
13
Effect of integral time on closed-loop response
I = 1=30
Setpoint change (ys=1) at t=0 Input disturbance (d=1) at t=20
![Page 13: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/13.jpg)
14
SIMC: Integral time correction•S
etpoints: ¿I=¿1(“IMC-rule”). Want smaller integral time for disturbance rejection for “slow” processes (with large ¿1), but to avoid “slow oscillations” must require:
•Derivation:
•Conclusion SIMC:
![Page 14: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/14.jpg)
15Typical closed-loop SIMC responses with the choice c=
![Page 15: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/15.jpg)
16
SIMC PI tuning rule
c ¸ - : Desired closed-loop response time (tuning parameter)•For robustness select: c ¸
Two questions:• How good is really the SIMC rule?• Can it be improved?
Reference: S. Skogestad, “Simple analytic rules for model reduction and PID controller design”, J.Proc.Control, Vol. 13, 291-309, 2003
“Probably the best simple PID tuning rule in the world”
![Page 16: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/16.jpg)
17
How good is really the SIMC rule?
Want to compare with:
•Optimal PI-controller
for class of first-order with delay processes
![Page 17: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/17.jpg)
18
•Multiobjective. Tradeoff between
– Output performance – Robustness– Input usage– Noise sensitivity
High controller gain (“tight control”)
Low controller gain (“smooth control”)
• Quantification– Output performance:
• Rise time, overshoot, settling time
• IAE or ISE for setpoint/disturbance
– Robustness: Ms, Mt, GM, PM, Delay margin, …
– Input usage: ||KSGd||, TV(u) for step response
– Noise sensitivity: ||KS||, etc.
Ms = peak sensitivity
J = avg. IAE for setpoint/disturbance
Our choice:
3. Optimal controller
![Page 18: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/18.jpg)
19
IAE = Integrated absolute error = ∫|y-ys|dt, for step change in ys or d
Cost J(c) is independent of:1. process gain (k)2. setpoint (ys or dys) and disturbance (d) magnitude3. unit for time
Output performance (J)
![Page 19: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/19.jpg)
20
4. Optimal PI-controller: Minimize J for given Ms
Optimal PI-controller
Chriss Grimholt and Sigurd Skogestad. "Optimal PI-Control and Verification of the SIMC Tuning Rule". Proceedings IFAc conference on Advances in PID control (PID'12), Brescia, Italy, 28-30 March 2012.
![Page 20: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/20.jpg)
21
Optimal PI-settings vs. process time constant (1 /θ)
Optimal PI-controller
Ziegler-Nichols
Ziegler-Nichols
![Page 21: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/21.jpg)
22
Ms=2
Ms=1.2
Ms=1.59|S|
frequency
Optimal PI-controller
Optimal sensitivity function, S = 1/(gc+1)
![Page 22: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/22.jpg)
23
Ms=2
Optimal PI-controller
4 processes, g(s)=k e-θs/(1s+1), Time delay θ=1.Setpoint change at t=0, Input disturbance at t=20,
Optimal closed-loop response
![Page 23: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/23.jpg)
24
Ms=1.59
Optimal PI-controller
Setpoint change at t=0, Input disturbance at t=20,g(s)=k e-θs/(1s+1), Time delay θ=1
Optimal closed-loop response
![Page 24: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/24.jpg)
25
Ms=1.2
Optimal PI-controller
Setpoint change at t=0, Input disturbance at t=20,g(s)=k e-θs/(1s+1), Time delay θ=1
Optimal closed-loop response
![Page 25: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/25.jpg)
26
Infeasible
UninterestingPareto-optimal PI
![Page 26: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/26.jpg)
27
Optimal performance (J) vs. Ms
Optimal PI-controller
![Page 27: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/27.jpg)
31
5. What about SIMC-PI?
![Page 28: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/28.jpg)
32
SIMC: Tuning parameter (¿c) correlates nicely with robustness measures
Ms
GM
PM
¿c=µ ¿c=µ
![Page 29: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/29.jpg)
33
What about SIMC-PI performance?
![Page 30: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/30.jpg)
34Comparison of J vs. Ms for optimal and SIMC for 4 processes
![Page 31: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/31.jpg)
35
Conclusion (so far): How good is really the SIMC rule?
•Varying C gives (almost) Pareto-optimal tradeoff between performance (J) and robustness (Ms)
C = θ is a good ”default” choice
•Not possible to do much better with any other PI-controller!
•Exception: Time delay process
![Page 32: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/32.jpg)
36
6. Can the SIMC-rule be improved?
Yes, for time delay process
![Page 33: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/33.jpg)
37
Optimal PI-settings vs. process time constant (1 /θ)
Optimal PI-controller
![Page 34: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/34.jpg)
38
Optimal PI-settings (small 1)
Time-delay processSIMC: I=1=0
0.33
Optimal PI-controller
![Page 35: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/35.jpg)
39
Step response for time delay process
θ=1
Optimal PI
NOTE for time delay process: Setpoint response = disturbance responses = input response
![Page 36: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/36.jpg)
40
Pure time delay process
![Page 37: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/37.jpg)
41
Two “Improved SIMC”-rules that give optimal for pure time delay process
1. Improved PI-rule: Add θ/3 to 1
1. Improved PID-rule: Add θ/3 to 2
![Page 38: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/38.jpg)
42Comparison of J vs. Ms for optimal-PI and SIMC for 4 processes
CONCLUSION PI: SIMC-improved almost «Pareto-optimal»
![Page 39: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/39.jpg)
7. Better with IMC or Smith Predictor?
Surprisingly, the answer is:
NO, worse
![Page 40: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/40.jpg)
Smith Predictor
K: Typically a PI controller
Internal model control (IMC): Special case with ¿I=¿1
Fundamental problem Smith Predictor: No integral action in c for integrating process
c
![Page 41: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/41.jpg)
45
Optimal SP compared with optimal PI
SP = Smith Predictor with PI (K)
¿1=20 since J=1 for SPfor integrating process
¿1=20
¿1=0
¿1=8
¿1=1
Small performance gain with Smith Predictor
![Page 42: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/42.jpg)
46
Additional drawbacks with Smith Predictor•N
o integral action for integrating process•S
ensitive to both positive and negative delay error•W
ith tight tuning (Ms approaching 2): Multiple gain and delay margins
![Page 43: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/43.jpg)
47
Step response, SP and PI
g(s) = k e¡ s
s+1
µ= 1
y
time time time
Smith Predictor: Sensitive to both positive and negative delay error
SP = Smith Predictor
![Page 44: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/44.jpg)
48
Delay margin, SP and PI
SP = Smith Predictor
![Page 45: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/45.jpg)
50
8. Conclusion Questions for 1st and 2nd order processes with delay:
1.How good is really PI/PID-control?– Answer: Very good, but it must be tuned properly
2.How good is the SIMC PI/PID-rule?– Answer: Pretty close to the optimal PI/PID, – To improve PI for time delay process: Replace 1 by 1+µ/3
3.Can we do better with Smith Predictor or IMC?– No. Slightly better performance in some cases, but much worse delay margin
4.Can we do better with other non-PI/PID controllers (MPC)?– Not much (further work needed)
•SIMC: “Probably the best simple PID tuning
rule in the world”
![Page 46: Optimality of PID control for process control applications](https://reader035.vdocuments.net/reader035/viewer/2022062522/56813507550346895d9c5482/html5/thumbnails/46.jpg)
51
11th International IFAC Symposium on Dynamics and Control of Process and
Bioprocess Systems (DYCOPS+CAB). 06-08 June 2016
Location: Trondheim (NTNU)
Organizer: NFA (Norwegian NMO) + NTNU (Sigurd Skogestad, Bjarne Foss, Morten Hovd, Lars Imsland, Heinz Preisig, Magne Hillestad, Nadi Bar),
Norwegian University of Science and Technology (NTNU), Trondheim
Welcome to: