Seeking experience in PID tuning
Thread Content
I’m an operator. We rarely change the PID parameters of the instruments; I’ve only read brief introductions to this topic in old manual books, and I’m still quite confused. I wonder if other operators have any experience in this area – for example, adjusting the values of PID parameters based on parameter curves. I thought about posting a question in the automation control section, but I’m afraid the explanations there might be too technical. I just want to learn something that’s easy to understand.
I’d like to learn something simple and easy to understand. Here are some common mnemonics for PID control: To determine the optimal parameter settings, start by checking values from low to high; first use the proportional term, then the integral term, and finally add the derivative term. If the curve oscillates frequently, increase the proportional gain; if the curve wanders around in large swings, decrease the proportional gain. If the curve takes a long time to return to its normal position, reduce the integral time. If the curve has long fluctuation cycles, extend the integral time further. If the oscillation frequency is high, first reduce the derivative term. If there is significant error and slow response, increase the derivative time. An ideal curve should have two peaks, with the first peak being higher than the second, in a 4:1 ratio. By observing and making adjustments while analyzing the situation, the quality of tuning will be good. For the engineering-based determination of PID controller parameters, the following empirical values can be used as references for P.I.D parameters in various control systems: For temperature T: P=20~60%, T=180~600s, D=3-180s; for pressure P: P=30~70%, T=24~180s; for liquid level L: P=20~80%, T=60~300s; for flow rate L: P=40~100%, T=6~60s.
Indeed, I basically follow the formula mentioned above to make the settings. In practice, it’s quite troublesome – one has to keep trying different values. I usually need a stable system, with a high gain setting of over 100%, so that adjustments can be made quickly and instantly; in such cases the gain is set low, and then the integral term is adjusted to stabilize the curve. Apart from temperature, differential terms are generally not used
The ones on the second floor – I’ve all written them down in my notebook: loveliness: