Particle Swarm Optimisation for the Design of Brushless Permanent Magnet Machines

This paper describes the use of a particle swarm optimisation (PSO) algorithm in the design optimisation of two surface magnet brushless machines. Two PSO algorithms are presented: the global best PSO (gbest PSO) and local best PSO (lbest PSO). The aim is to determine the multi-polar permanent magnet (PM) array that maximises the motor´s output. Two parameters are optimised: the arc lengths and the directions of magnetisation of the PM segments. The proposed optimisation system combines a two-dimensional (2D) finite element method (FEM) with a PSO algorithm. The objective function used in the optimisation procedure is based on maximising phase flux linkage for a given volume of permanent magnet. The calculation results are compared with radial, uniform and discrete sinusoidal Halbach magnetisations of the rotor