Adaptive wavelet neural network control for dc motors via second-order sliding-mode approach

Abstract

This paper proposes an adaptive wavelet neural network control (AWNNC) system which is composed of a neural controller and a smooth compensator via second-order sliding-mode approach. The neural controller utilizes a wavelet neural network to approximate an ideal second-order sliding-mode controller and the smooth compensator is designed to guarantee the system stability without occurring chattering phenomena. Moreover, to speedup the convergence of the tracking error, a proportional-integral-derivative type adaptation tuning mechanism is derived based on Lyapunov stability theory. Finally, the proposed AWNNC method is implemented on a field programmable gate array chip and is applied to a DC motor to show its effectiveness. The experimental results verify the system stabilization and the favorable tracking performance can be achieved by the proposed AWNNC system even under the change of the command trajectory and frequency.