Seeking PID tuning experience
Thread Content
Recently, while tuning the PID loops of newly installed equipment, I have a few questions I’d like to discuss with everyone: 1. In your company, is PID tuning handled by the process team or by the instrumentation team? Our company specializes in instruments. 2. Are those tuning mnemonics that are widely available online useful? Personally, I don’t think it’s very effective. 3. For those liquid level situations where the response is very slow, in the tower top pressure control loop, one adjustment cycle takes about an hour. Do anyone have any good experience regarding the tuning of such control loops? 3. Besides the temperature loop, can differential effects be applied to the slowly responding levels and pressures?
PID is generally adjusted by the instruments in coordination with the process.
The tuning mnemonic is useful; adjustments are based on empirical values.
Bro, do you have any experience with PID tuning? We could exchange ideas: handshake
It can only be said that those with slow reaction times should use differentiation as little as possible
A conventional PID control loop usually only requires proportional and integral control
The tuning of PID parameters should be determined based on the requirements of the process, with the assistance of instrumentation； Generally, the precision of control and the stability of the system are contradictory; it is necessary to find an appropriate balance. For example, in terms of level control, a level deviation of 5% is acceptable; it is possible to make adjustments and the valve position remains stable, with little impact on the processes upstream and downstream. This makes it a suitable approach, so a proportional control system is sufficient – there is no need to eliminate residual errors or use feedforward control; the main thing is to adjust the proportional coefficient.
Personally, I think one should first estimate the feedback cycle roughly and establish a general framework using some rules of thumb; after that, it’s a matter of guessing based on experience. Differentiation is generally used for fine-tuning once stability is achieved.