Radial Basis Function Network based Design Optimization of Induction Motor

Abstract

The application of radial basis function (RBF) network model for optimum design of induction motor (ODIM) is presented. The method utilizes simulated annealing (SA) technique to provide optimum design as training data to the RBF network. RBF is a new generation of artificial neural networks (ANN's) of auto configuring nature and extremely fast training procedure. The RBF network model so developed is applied to a set of test data and results are compared with those obtained from the optimization technique (SA) results. Test results reveal that the proposed model determines the optimal dimensions of three phase induction motor along with the performance parameters efficiently and accurately