The paper presents a methodology for the optimization of a Brushless Direct Current motor (BLDC). In particular it is focused on multiobjective optimization using a genetic algorithm (GA) developed in Matlab/Optimization Toolbox coupled with Maxwell from ANSYS. Optimization process was divided into two steps. The aim of the first one was to maximize the RMS torque value and to minimize the mass. The second part of the optimization process was to minimize the cogging torque by selecting proper magnet angle. The paper presents the methodology and capabilities of scripting methods rather than specific optimization results for the applied geometry.

Permanent magnet (PM) excited synchronous machines used in modern drives for electro-mobiles suffer in high speed regions from the limited battery-voltage. The field weakening requires designing machines with reduced power conversion properties or increasing the size of the power converter. A new concept of such a machine features PM excitation, single-tooth winding and an additional circumferential excitation coil fixed on the stator in the axial center of the machine. By the appropriate feeding of this coil, the amplitude of the voltage effective excitation field can be varied from zero to values above those of the conventional PM-machines. The capability of reducing the excitation field to zero is an important safety aspect in case of failing of the feeding convertor.