IPM motor rotor design by means of FEA-based multi-objective optimization

The design optimization of IPM motors for wide speed ranges is pursued by means of a FEA-based multi-objective genetic algorithm (MOGA). Respect to previous works in the literature, the proposed approach evaluates the motor performance with a very limited number of simulations, making FEA optimization more attractive. The 3 goal functions (motor torque, torque ripple and flux weakening capability) are evaluated by means of 7 static FEA runs, that means nearly 20 seconds per tentative motor with a laptop computer. The paper is focused on the rotor design, since it is the most controversial part of IPM design and the most difficult to be modeled due to magnetic saturation. Three different approaches are presented: a fast one, based on 2- objective optimization, a hybrid one, based on 2-objective optimization and 3-objective refinement, and actual 3-objective optimization. The results presented here will be the base for future, more comprehensive optimization.