On the tuning of PID parameters

Thread Content

I’m not majoring in instrumentation. But now it is necessary to tune the PID parameters, and I would like to ask all fellow users for their advice: 1. I have found two sets of PID tuning values for this device from other locations. The differences are significant; for example: Zhenhai 126PIC-5305F-101 has a radiation chamber value of 100,200, while Lanzhou 31PIC-2007F-201 has values of 0, 50, and 80. Why are the differences so large? Is it the different tuning methods? I found some information on setting parameters in this section, and there is a passage that says: During the actual tuning process, we usually first turn off the integral and derivative terms, set TI to infinity and TD to zero, thereby turning it into a pure proportional control system. The initial proportionality is set based on empirical data; thereafter, the proportionality control is gradually adjusted according to the PV curve so that the system achieves a PV curve with a 4:1 damping oscillation, and then an integral action is added. Before applying the gain, the scale factor should be increased to about 1.2 times the original value. Adjust the integration time TI from large to small until the PV curve of the system again exhibits a 4:1 damped oscillation. To introduce differential action, the differential time is calculated as TD=(1/3～1/4) TI. At this point, the proportionality gain can be set to its original value or lower, after which the differential time is adjusted from small to large until the PV curve meets the desired criteria. One thing to note is that during the trial process, if TI and TD need to be changed, the ratio should remain constant. Should TI be set to infinity? How big? I tried to do it using this method. The trend is okay. But the result was close to that of Zhenhai earlier. The parameters set previously were similar to those in Lanzhou. Is it wrong?