Design and multiobjective optimization of the parameters of switched reluctance motor

Abstract

Switched reluctance motor (SRM) has been the subject of many investigations in recent years. The SR technology is a feasible and economically advantageous option for future electromotor production in Iran. However, computational intensive analysis methods (e.g. finite element (FE) analysis) and iterative search for optimal values of design parameters (e.g. genetic algorithms) cannot be jointly carried out online. In this paper, we outline a three-stages design process for determination of the optimal electromotor specifications. In the first stage, FE analysis is used for computation of performance characteristics associated with various design parameters. Next, an interpolator is trained to establish a mapping from design parameters to performance characteristics. Finally the latter values are optimized via multicriteria genetic algorithm.