How should the PID parameters of the primary and secondary controllers in a cascade control system be set?

For the engineering tuning of PID controller parameters, the following are some empirical values for P.I.D parameters in various control systems: Temperature T: P=20~60%, T=180~600s, D=3-180s; Pressure P: P=30~70%, T=24~180s; Liquid level L: P=20~80%, T=60~300s; Flow rate L: P=40~100%, T=6~60s. Generally, a value is set first, and then adjustments are made gradually until the control requirements are met. Nowadays, valve positioners are essentially required for all control valves

Setting of PID parameters: This is done based on experience and familiarity with the manufacturing process, by referring to the measured values and the setpoint curve in order to adjust the values of P, I, and D. For the engineering tuning of PID controller parameters, the following are some empirical values for P.I.D parameters in various control systems: Temperature T: P=20~60%, I=180~600s, D=3–180s; Pressure P: P=30~70%, I=24~180s; Level L: P=20~80%, I=60~300s; Flow Rate L: P=40~100%, I=6~60s. 　　 Common mnemonics from the book:　　 To determine the optimal parameter settings, check in ascending order.　　 Start with 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 loops, reduce the proportional gain.　　 If the curve takes a long time to return to its normal position, decrease the integration time.　　 If the fluctuations in the curve have a long period, increase the integration time further.　　 If the oscillation frequency of the curve is high, first reduce the derivative term.　　 Large errors lead to slow fluctuations. The differential time should be increased. The ideal curve consists of two waves, with the first one being higher and the second one lower, in a 4:1 ratio. Observe, adjust, and analyze thoroughly – this way, the quality of the adjustments will not be poor. Here, an empirical method is introduced. This method is essentially a trial-and-error approach; it is an effective technique developed through production practice and has been widely applied in the field. The basic procedure of this method is to first determine a set of controller parameters based on operational experience and put the system into closed-loop operation; thereafter, a step disturbance is artificially introduced (such as by changing the setpoint of the controller), and the step response curve of the controlled variable or the controller output is observed. If the control quality is deemed unsatisfactory, the regulator parameters are adjusted based on the impact of each tuning parameter on the control process. Repeat this testing until satisfaction is achieved. The empirical method is simple and reliable, but it requires certain on-site operation experience, and subjectivity and one-sidedness can easily arise during adjustment. When a PID regulator is used, there are multiple tuning parameters, and as the number of trial adjustments increases, it becomes difficult to determine the optimal tuning parameters. This post was last edited by zlky2005 on 2009-3-25 15:19.]

First, adjust the secondary circuit separately, and then add the primary circuit; the primary circuit should not have a derivative term. The secondary loop is primarily proportional, with addition of integration and then differentiation. More or less

Just now it was said on the 4th floor that differentials should not be used, but I think that’s not feasible. Since cascade control systems are generally applied in situations where there is a large capacity lag, it’s necessary to use differentials. Therefore, the main controller usually uses a PID control scheme, while the secondary controller can suffice with a P control scheme. However, when the secondary parameters are flow rate and pressure, a certain effect of T can be applied. This is not to eliminate residual errors; rather, it’s because the time constant for these flow and pressure variables is too small, which results in the P value of the secondary regulator not being able to be too low, leading to a weak regulating effect. The integration term is used to prevent the secondary parameters from deviating too far from the set values. So the question is: the main regulator uses 3 PID parameters, while the secondary regulator, since it deals with steam flow, uses 2 PI parameters. Of course, a valve positioner must be installed.

1. A control valve has two sets of “controllers”; it’s possible to set up two PID controllers, something I really hadn’t noticed before. I hope someone can give an accurate answer; I’ll also pay attention to it when I return to work after the holiday. 2. We also have some cascade control valves, but in reality they are all used to control one parameter only; true cascading control is not implemented.