Design and Implementation of a Sliding Mode Controller Using a Gaussian Radial Basis Function Neural Network Estimator for a Synchronous Reluctance Motor Speed Drive

Abstract

This article presents a sliding mode control using a Gaussian radial basis function neural network speed control design for robust stabilization and disturbance rejection of the synchronous reluctance motor. In the conventional sliding mode control design, it is assumed that the upper boundary of parameter variations and external disturbances is known and the sign function is used. This causes high-frequency chattering and high gain. A new sliding mode controller using a Gaussian radial basis function neural network estimator is proposed for the synchronous reluctance motor. The proposed method utilizes the Lyapunov function candidate to guarantee convergence and to track the speed command of the synchronous reluctance motor asymptotically. The estimator of parameter variations and external disturbances is designed to estimate the lump unknown uncertainty value in real time. Experiments were conducted to validate the proposed method.

Keywords:

• sliding mode control
• Gaussian radial basis function neural network
• synchronous reluctance motor
• Lyapunov function

Acknowledgment

This work is supported by the National Science Council in Taiwan, Republic of China (grant NSC96-2221-E-224-089).