Iterative feedback tuning (IFT) based design method of UDE control system

Abstract

The invention discloses an IFT based design method of a UDE control system. Transfer functions of different modules of the UDE control system are normalized, and a filter parameter is determined by considering a target of minimizing an integral absolute error value during change of the closed-loop bandwidth and disturbance step; an empirical formula of the system robustness index Mst, the normalization parameter x of the controller and the normalized time lag tau<-> are derived; an IFT algorithm is used to optimize an object nominal model so that the model can describe practical objects; and the normalized time lag tau<-> is calculated according to the know object nominal model, the controller normalization parameter x is calculated by combining a given robustness index Mst, a reference model parameter is determined by weighting the system robustness and the response speed, and design of the UDE control system is completed. According to the invention, the IFT method is applied to design of the UDE control system for the first time, model parameter identification and controller parameter setting can be completed at the same time by completing a specific closed-loop experiment, and requirements for the robustness, setting value tracking performance and disturbance inhibition performance are met simultaneously.

Description

technical field [0001] The invention relates to the technical field of industrial process control, in particular to a UDE control system design method based on iterative feedback tuning. Background technique [0002] Due to the existence of unmodeled dynamics, parameter changes and various unknown disturbances in large-scale industrial process control, good control effects are usually not obtained. In order to solve this problem, Zhong Qingchang proposed an uncertain disturbance estimation controller in 2004. The main idea is to process model uncertainty and external disturbances into a disturbance lumped item, and use a filter with a reasonable bandwidth to estimate and eliminate The disturbance lumped term. Because UDE control has the characteristics of low order, two degrees of freedom, and simple design, it shows broad development prospects in terms of robustness improvement and disturbance suppression. However, the performance of this original UDE design is significan...

AI Technical Summary

Problems solved by technology

However, the performance of this original UDE design is significantly reduced when dealing with industrial processes with time delays. Based on this, someone modified the original UDE control and proposed a modified uncertain disturbance estimation control suitable for time-delay systems, so that UDE Control is more widely used

However, obtaining an accurate model of the controlled process from the field is generally very difficult and time-consuming

Images