Experiment 7: Pid Controller Tuning of Simulated Gas Mass Flow Process
Experiment 7: Pid Controller Tuning of Simulated Gas Mass Flow Process
SIMULATION TOOL
Process Model : SPC-211
Control Loop : Single Loop Flow Feedback Control ( FIC-91 )
Process Control & Instrumentation Laboratory BKF3791
EXPERIMENTAL PROCEDURE
1. Select the gas mass flow process
2. Call up the controller faceplate FIC-91 and put the controller to Manual Mode
3. Check maximum and minimum value for process variable, PV by setting the MV to
0% ( MVmax ) and let the process stabilize for a few minutes ( see your graph )
4. Complete the table below for process variable, PVmax
5. Repeat step 3-4 for MVmin ( 100% )
MV2
MV1
PV2
63.2%
PV1
td tc
PV2 PV1
PVmin
Kp
PVmax
MV2 MV1
MVmax MVmin
Time delay, td -
Time constant, tc -
Kp
Response Rate, RR RR
tc
Lambda, tC
0.8,
td 10
Process Control & Instrumentation Laboratory BKF3791
11. By using the data from the previous step, calculate the value of gain, K c , reset I
and rate, D for Cohen-Coon Tuning Relations, Zeigler and Nichols Tuning Relations
and Internal Model Control Tuning Relations and complete the table below
1 c D
P 1 - -
K p D 3 c
D
D 30 3
1 c D c
PI 0.9 -
Kp D 12 c
9 20 D
c
D
D 32 6 14 D
1 c 16 c3 D c
PID D
K p D 12 c 11 2
13 8 D c
c
Performance Test
12. Put the controller FIC-91 to Manual mode and set the controller setting accordingly
to calculate value for Cohen-Coon Tuning Relations ( P Control ) obtained from
previous step
13. Set the set point, SP to 18kg/Hr and manually adjust the MV so that the PV
fluctuated at the set point. Let the process stabilize for a few minutes
14. Put the controller FIC-91 to Auto Mode and stimulate a set point change by
changing the set point to 25 kg/Hr and let the process stabilize at new set point.
15. Save your graph for discussion
16. Repeat step 12-15 for PI and PID controller setting using Cohen-Coon tuning
method. Save the graph for discussion.
17. Repeat step 12-16 for Zeigler and Nichols and Internal Model Control
18. Discuss your result
QUESTIONS
1. Base on your experiment, which is the best tuning for the process ?
2. How do you clarify the best tuning for the process ?
3. What is the function of doing a performance test in this experiment ?