Permanent Magnet Machine Design Practice and Optimization

Abstract

The complexity of the permanent magnet (PM) machine structure makes the optimal design of the PM machine always a difficult task. The multiple objectives of an optimal design make most classic optimization algorithms inapplicable, due to the nonlinearity and some discontinous variables. In this paper, two interior permanent magnet (IPM) machine design practices with modular stator structure and conventional stator structure are discussed. The design process is directly coupled with finite-element analysis (FEA) with the machine design guidelines embedded in the optimization process. Multivariable optimization methods based on Monte Carlo and Differential Evolution algorithms are applied in the design phase and the results are compared in this paper.