Comparison between PSO and Genetic Algorithms and for optimizing of Permanent Magnet Synchronous Generator (PMSG) machine design

Abstract

This paper proposes application of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) in the design of direct-driven permanent magnet synchronous generator machine (PMSGs) for wind turbine applications. The power rating of these machines is in the mega watt (MW) level. The constraints and requirements of the generator are outlined. The proposed design scheme optimizes various PMSG parameters like Pole pair number, Linear current density, Air gap thickness, Rotor outer diameter, Relative width of the permanent magnet etc to achieve certain objectives like maximizing efficiency, increasing Torque, improving power factor etc. The results obtained by GA algorithm and those by PSO algorithm are compared. The performance of Particle Swarm Optimization is found to be better than the Genetic Algorithm, as the PSO carries out global search and local searches simultaneously, whereas the Genetic Algorithm concentrates mainly on the global search. Results show that the proposed PSO optimization algorithm is easy to develop and apply and produced competitive designs compared to the GA algorithm.