Experimental Optimal Design of Slotless Brushless PM Machines Based on 2-D Analytical Model

Abstract

This paper presents an effective optimal design procedure for brushless permanent magnet (BLPM) machines based on an adaptive metaheuristic optimization technique. 2-D analytical expressions for the calculation of the magnetic flux density, electromagnetic torque, back electromotive force, and self-inductance and mutual-inductance are employed to optimally design a slotless brushless motor with surface-mounted magnets. The proposed approach combines the computational accuracy of the 2-D analysis and the computational speed due to the analytical expressions. The objective functions are the power losses and the motor volume to be simultaneously minimized subject to three constraints: 1) the required electromagnetic torque; 2) the required maximum rotational velocity; and 3) the limit of the stator core flux density. The optimization problem is solved using a fuzzy adaptive particle swarm optimization (FAPSO) technique. To evaluate the efficacy and effectiveness of the FAPSO technique, its results are compared with those of the conventional PSO and genetic algorithms. To investigate the influence of the armature current waveforms on the design results, three different optimization problems with different armature current waveforms (ideal rectangular, six-step, and sinusoidal), but identical objectives, constraints, and optimization variables are defined and solved. Finally, a BLPM machine has been designed and manufactured to experimentally show the effectiveness of the proposed technique.