An Adaptive Self-Tuned Wavelet Controller for IPM Motor Drives

Abstract

This paper presents an intelligent control system for interior permanent magnet motor drives using a wavelet neural network. The wavelet neural network combines the capability of artificial neural networks for learning from processes and the capability of wavelet decomposition for identification and control of dynamic systems. A four-layer wavelet neural network is adopted to implement the proposed controller for IPM motor drives. The inputs of the wavelet neural network are the tracking speed error and the change of speed error. The output of the wavelet neural network is the command torque current. The proposed wavelet neural network controller is trained on-line with adaptive learning rates to control the rotor position of the IPM motor drive system. The adaptive learning rates are derived using discrete Lyapunov stability theorem so that the convergence of the tracking error is guaranteed in the closed- loop system. The performance of this newly devised wavelet controller is evaluated by simulation and experimental results. The complete vector control scheme incorporating the proposed wavelet neural network controller is successfully implemented in real-time using the ds1102 digital signal processor board for the laboratory 1-hp IPM motor. In order to prove the superiority of the proposed controller over the conventional controllers a comparison between the proposed and the conventional proportional-integral (PI) and proportional-integral-derivative (PID) controllers based systems is made at various dynamic operating conditions.