Abstract: This paper presents a improved learning control algorithm in switched reluctance motors(SRM) for torque ripple reduction with a CMAC neural network.Based on maintaining the advantage of the original agorithm the approach allows the generation of optimal current profiles in terms of minimizing torque ripple and copper loss as the motor operates at low speeds,and of minimizing torque ripple,copper loss and rate of change of current as the motor runs at high speeds.

Key words: adaptive control, CMAC neural network, constrained LMS algorithm, torque ripple reduction, switched reductance motor