Optimization of Single-phase Induction Motor Design using Radial Basis Function Network

This paper presents a radial basis function (RBF) model for optimal design of single-phase induction motor. The RBF network is a new generation of artificial neural network (ANN) of auto configuring nature and extremely fast training procedure. The induction motor design optimization is formulated as a nonlinear programming problem and Simulated Annealing (SA) is used for arriving at the optimal design. RBF network is trained with this optimal data. The model so developed is applied to two test motors and the results are compared with those obtained from SA, GA and conventional method. Test results reveal that the proposed scheme determines the optimal geometry of induction motor efficiently, accurately and quickly.