Adaptive sliding-mode control method of nonlinear system

Abstract

The invention provides an adaptive sliding-mode control method of a nonlinear system, and belongs to the artificial intelligence and control field. The method is used for solving the problem of asymptotic stability of a closed-loop control system. The method is characterized by, for the nonlinear system, adopting an LS-SVM structure to approximate an ideal state feedback controller so as to construct a new feedback controller; and carrying out sliding-mode control to compensate for LS-SVM regression approximation error and / or uncertain external disturbance, and determining weight parameter vectors based on adaptive laws. The beneficial effects are that the feedback linearization controller is designed by fully utilizing the approximation capability of a nonlinear function of LS-SVM regression, and the sliding-mode control is introduced to compensate for the LS-SVM regression approximation error and the influence of uncertain external disturbance on system output, and thus adjustment ofthe LS-SVM weight parameters can be realized.

Description

technical field [0001] The invention belongs to the field of artificial intelligence and control, and relates to an adaptive sliding mode control method of a nonlinear system. Background technique [0002] Sliding mode variable structure control of nonlinear uncertain systems has always been a hot spot in the control field, and many scholars have achieved research results in this field. Since the sliding mode control of the nonlinear system requires a rough mathematical model of the known system, the dependence of the sliding mode control on the system model is increased. With the development of artificial intelligence theory, fuzzy logic and neural network are introduced into the design of sliding mode control, which can effectively reduce the dependence of sliding mode control on the system model. Literature [3] studied the adaptive fuzzy sliding mode control of nonlinear system based on high-gain observer, and literature [4] studied the adaptive sliding mode control of n...

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problem that the closed-loop control system is asymptotically stable, the present invention proposes the following scheme:

This method makes full use of the nonlinear function approximation ability of LS-SVM regression to design a feedback linearization controller, introduces sliding mode control to compensate the approximation error of LS-SVM regression and the influence of uncertain external disturbance on the system output, and performs LS-SVM weight parameter adjustment, and finally verified the design scheme through a simulation example, indicating that the present invention can solve the asymptotically stable problem of the closed-loop control system

Images