PID control algorithm applicable to multi-constraint targets

Abstract

The invention discloses a PID control algorithm applicable to multi-constraint targets, comprising the following steps: S1, if there is only one CV (Controlled Variable), calculating out the output value variation corresponding to the CV according to a PID algorithm, and if there is zero CV, letting the output value variation Delta(u) be zero; S2, calculating out the maximum constraint Delta(u)-max and the minimum constraint Delta(u)-min of the output value variation corresponding to each CCV (Constrained Controlled Variable) according to the PID algorithm; S3, performing comparison step by step according to the reverse order of importance, wherein the principle used is as follows: (1) if the less important Delta(u) is greater than the more important Delta(u)-max, the Delta(u)-max is used as a new Delta(u); (2) if the less important Delta(u) is smaller than the more important Delta(u)-min, the Delta(u)-min is used as a new Delta(u); and (3) if the less important Delta(u) is between Delta(u)-max and Delta(u)-min, the Delta(u) continues to be used as a new Delta(u); and S4, outputting an output value variation Delta(u) generated after comparison as the final result of a controller.

Description

technical field [0001] The patent of the present invention relates to the field of automatic control, in particular to a novel PID control algorithm, which can be applied to the process of dealing with multi-constrained targets. Background technique [0002] A typical control system consists of Controlled Variable (CV), Constrained Controlled Variable (CCV), Manipulated Variable (MV) and Disturbance Variable (DV). [0003] A controlled variable (CV) is a process output variable that needs to be controlled around some target value to improve process operation or product quality performance. [0004] Constraint variables (CCVs) are process output variables that need to be controlled within a certain range for safe plant operation, equipment capability limitations, or process optimization. [0005] A manipulated variable (MV) is a process input variable that can be adjusted to ensure that the manipulated variable (CV) is around a target value or the constraint variable (CCV) i...

AI Technical Summary

Problems solved by technology

[0028] The third and fourth types of processes are more complicated than the first and second types of processes, so they cannot be processed by the classic PID algorithm

However, there are still quite a few processes that cannot be simplified, and this part requires the use of multivariable control algorithms, such as model predictive control, which are beyond the scope of this article.

Images