Quantitative setting method for distributed PI and PID controller with chemical industrial double input and output

Abstract

A quantitative adjusting method includes designing expected response transfer function form and its dynamic demodulation factor form for two distributed control closed loops according to transfer function matrix model, deriving out two ideal expection distributed closed-loop controllers form, obtaining realization form of rational approximation by expansion series of Maclaurin and taking the first two items of expansion series to form distributed PI controller and taking the first three to form PID controller.

Description

technical field [0001] The invention relates to a method used in the technical field of industrial process control, in particular to a quantitative tuning method for a distributed PI and PID controller in a chemical industry dual-input-output process. Background technique [0002] Dual-input-output process is the most common multi-variable process in chemical production, and in order to facilitate operation and control, many high-dimensional multi-variable processes are usually decomposed into several dual-input-output subsystems for processing in practice. Decentralized control structure, also known as multiloop control structure, is widely used in chemical industry due to its advantages of simple structure, low cost and strong independent workability of each closed loop. However, the stability of this control structure is easily damaged by the uncertainty of the actual controlled process, so the tuning and adjustment requirements for the two closed-loop PI (proportional-in...

AI Technical Summary

Problems solved by technology

However, the stability of this control structure is easily damaged by the uncertainty of the actual controlled process, so the tuning and adjustment requirements for the two closed-loop PI (proportional-integral) or PID (proportional-integral) controllers in this control structure Relatively high, in order to ensure the stability of this control structure, the two closed-loop PI (or PID) controllers are usually adjusted relatively loosely, but this results in a relatively low working efficiency of the control system, resulting in a large consumption of raw materials and energy , is not conducive to economic production and operation

In the current chemical practice, the well-developed closed-loop controller tuning methods in the single variable process field, such as the Z-N method and the C-C method, are usually used to tune the two distributed closed-loop controllers of the dual-input-output process, and then respectively in the two The demodulation coefficient is set in the closed-loop controller to reduce the coupling effect between the two control closed-loops. Their main disadvantage is that the regulation level of the control system is low and the control quality is not high.

[0003] After searching the existing technical documents, it is found that American scholars Chen and Professor Seborg wrote "Decentralized PI control systems based on Nyquist stability analysis" (Design of distributed PI control systems based on Nyquist stability analysis, Published in Journal of Process Control, 2003, 13, 27-39.), it is proposed to use the multivariable Nyquist stability criterion to tune the distributed closed-loop PI and PID based on the critical gain and phase obtained from multivariable closed-loop identification in the frequency domain The method of the controller, referred to here as Chen's method, although the control effect has been significantly improved, but because the numerical optimization design method is used to tune the controller, the amount of data calculation is relatively large, and it is not convenient for online adjustment and setting. , and the relevant professional theoretical knowledge used is relatively large, which is not easy to be mastered and promoted by engineers and technicians

It should be pointed out that although the tuning methods of distributed closed-loop PI (or PID) controllers given by other recently published literatures can achieve significant improvement compared with the simple tuning methods commonly used in practice, they basically use numerical values. Optimized iterative method to optimize and solve the controller, the amount of data calculation is large and cumbersome, which is not conducive to online adjustment and engineering practice

Images