Design Optimization of Permanent Magnet-Brushless DC Motor using Elitist Genetic Algorithm with Minimum loss and Maximum Power Density

In this paper, design optimization of Permanent Magnet-Brushless DC (PM-BLDC) motor is presented by using Elitist Genetic Algorithm (GA). For this purpose, three objective functions are considered i.e. total loss and power density of the motor and combinations of both. Aim of this paper is to optimize the motor with these three objective functions separately. The first two objective functions are single-objective but for the third case, multi-objective optimization is performed in which total loss and power density that are technically opposite are formulated into one single objective. Seven design variables including stator inner diameter (D), axial length of motor (L), pole pitch (?p), specific magnetic loading (Bav), specific electric loading (ac), stator back-iron length (hbis) and stator slot height (hs) are chosen as optimization variables. Optimization is carried out by Elitist GA which has a better performance in comparison with conventional GA. Optimization results show that multi-objective functions performs much better comparing to single-objective functions because more reliable and realistic design optimization would be carried out by multi-objective functions. At last, Finite Element Method (FEM) is used that its results have well validated the analytical design optimization.