Variable speed control in switched reluctance drive based on least square-support vector machine

Abstract

Because of internal highly nonlinear characteristics in switched reluctance motor (SRM), it is difficult to obtain satisfying operating performance by using classical linear control methods. In this paper, a control method based on least square support vector machine (LS-SVM) is proposed. PI speed controller in SRM control system is taken place by a controller based on LS-SVM, which is generated from a group of input-output data LS-SVM algorithm trained. It can be concluded that SRM control system, based on the result of simulation in Matlab/Simulink, has good dynamic performance by using the speed controller based on LS-SVM.