A Fast Design Technique of SRM using Combined CAD and Genetic Algorithm

In this paper, an optimal design method of SR machine for a hydraulic pump system is researched. In order to get a proper efficiency and output power of hydraulic pump system in the restricted outer dimension, dominant design parameters of SR machine are learned by genetic algorithm with CAD program. The genetic algorithm is used to select the stator/rotor pole arc and the switching on/off angle of SR machine to improve the operating efficiency. Since a modified CAD program designs dominant parameters from genetic algorithm and returns the calculation results to genetic algorithm routine, all the design procedure is automatically implemented exception of proper fitness function design. In order to verify the proposed design method, the prototype SR machine with parameters determined by genetic algorithm is tested, and test results show high efficiency at rated output power of hydraulic pump system.