OPTIMIZATION WITH PARTICLE SWARM AND GENETIC ALGORITHM OF FLUX REVERSAL MACHINE

With the objective to identify optimization methods more suited to design unconventional electric machines, in this work we show the relevance of using a stochastic method newly proposed called particle swarm optimization (PSO), characterized by a constriction coefficient ensuring quick convergence. This method is applied to design and optimization of low speed flux reversal machine (FRM (50 rpm, 10 kW) dedicated to direct drive applications. The optimization results of this machine by PSO combined with the finite element method (FEM) are compared in terms of robustness, convergence and simplicity, to those obtained by genetic algorithm (GA) coupled with the FEM. Conclusion study are important in future development of PSO in design electrical machines.