Predictive wisdom PI control method for large time delay system

Abstract

The reason why a time delay system is difficult to control is that information feedback is not timely, thus a controller based on feedback information seems to be powerless to the kind of systems. Aiming at the control problem of the time delay systems, a predictive wisdom proportional-integral (PWPI) control method that does not depend on a controlled object model is invented. The PI control method regards nominal model parameters of a time delay object as model parameters of a Smith predictor, combines output of the Smith predictor with actual output of the object to form compensation outputof the system, designs a predictive wisdom PI controller, determines speed factors of the PWPI controller according to a time delay estimated value and a time constant nominal value, theoretically analyzes the stability of a closed-loop control system and the robustness of the controller, is an effective control method, and has a wide application value in the field of time delay system control.

Description

technical field [0001] It relates to a control method for a large time-delay system, in particular to a predictive intelligent PI control method. Background technique [0002] Time lag widely exists in industrial processes such as petroleum, chemical industry, power grid, nuclear reactor, and wastewater treatment. These industrial processes can often be simplified as one or more first-order plus-purity lag systems. Because of the existence of the time lag phenomenon, the control quantity of the system cannot respond to the dynamic changes of the system in a timely manner, which leads to the undesirable phenomena of large overshoot and long response time in the entire system, and even oscillation or instability. It brings great challenges to the design of time-delay system controllers. Therefore, solving the time-delay system control problem is of great significance to the precise tracking of the target trajectory and the stability of the system. For the control problems o...

AI Technical Summary

Problems solved by technology

The existing PI control law is constituted by the independent weighted summation of two physical quantity links with different attributes, the proportion of the error and the integral. Then, what is the attribute (dimension) of the control quantity (control force)? Because physical quantities of different dimensions cannot be arithmetically operated, it is impossible to accurately understand the mathematical model of the PI control law in a physical sense;

[0005] (2) Lack of scientific thinking on collaborative control

[0006] (3) Because of the irrationality of the PI control law model, the problem of PI controller gain tuning has been a difficult problem in the field of control science and control engineering for nearly a century.

Images