Finite Element Design and Multi-objective Optimization of Four Pole Reluctance Motor Based on NSGA-II Intelligent Algorithm

The design of a four-pole reluctance motor with multiple objectives is discussed in this paper using a finite element design methodology based on multi-objective genetic algorithm. Non-dominated genetic algorithm (NSGA-II) is used because of its high performance and intensification in optimization problems. The global sensitivity chart revealed that the motor’s stator pole embrace and yoke thickness are key parameters for the optimization objectives, while the rotor’s pole embrace should be restrained and closely associated with these two key parameters. According to the optimization and sensitivity analysis results, a final design which is superior to the base design was achieved. There were 15 % and 13.2 % improvement in the optimized model in terms of the average torque and efficiency respectively. Also, the optimized model recorded a reduction in the average torque ripple and total loss by 1.55 % and 30.1 % respectively. This demonstrates the NSGA-II intelligent optimization program is a suitable framework to optimize specified objective functions.