Disturbance-rejection neural network control

Author

Xiu, Y.M. ; Zhao, Z.Y.

Author_Institution

Dept. of Autom., Tsinghua Univ., Beijing, China

Volume

2

fYear

1993

fDate

25-29 Oct. 1993

Firstpage

1841

Abstract

A disturbance-rejection neural network control scheme is presented for control of an unknown nonlinear plant. In the scheme, a multilayer neural network is employed to learn the inverse dynamics of the unknown plant and acts as a feedforward controller to control the plant. The effect of disturbances on the output is suppressed by using a paralleled closed-loop control system. The design technique of the compensator in the closed-loop system is discussed. Simulation results show that the presented control scheme works well in the presence of disturbances.

Keywords

closed loop systems; compensation; feedforward; feedforward neural nets; learning (artificial intelligence); neurocontrollers; nonlinear control systems; compensator; disturbance-rejection neural network control; feedforward controller; inverse dynamics; learning; multilayer neural network; paralleled closed-loop control system; unknown nonlinear plant; Automatic control; Automation; Control systems; Extrapolation; Feedforward neural networks; Multi-layer neural network; Neural network hardware; Neural networks; Nonlinear dynamical systems; Open loop systems;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on

Print_ISBN

0-7803-1421-2

Type

conf

DOI

10.1109/IJCNN.1993.717013

Filename

717013