Automated Multi-Objective Design Optimization of PM AC Machines Using Computationally Efficient FEA and Differential Evolution

The design optimization methods described in this paper are employing an ultrafast computationally efficient finite element analysis technique. A minimum number of magnetostatic solutions are used for the analysis, which makes possible the study of thousands of candidate motor designs with typical PC-workstation computational resources. A multi-objective differential evolution algorithm that considers a large number of independent stator and rotor geometric variables and performance criteria, such as average and ripple torque, losses, and efficiency, is used. The optimization method is demonstrated on different permanent magnet (PM) ac synchronous motors in the kilowatt and megawatt power ranges. For the low-power PM ac machine study, a nine-slot six-pole topology is considered. For the high-power PM ac machines, four case studies were carried out with the following: fractional-slot embedded surface PM (SPM), fractional-slot interior PM (IPM), integer-slot SPM, and integer-slot IPM, respectively. Four motor topologies are systematically compared based on optimal Pareto sets. The design optimization of IPM motors includes an additional search for an optimum operating torque angle corresponding to the maximum-torque-per-ampere condition.