Torque ripple suppression of a new in-wheel motor based on quantum genetic algorithm

Abstract

Quantum genetic algorithm (QGA) was proved better than the traditional genetic algorithm in numerical and combinatorial optimization problems. However, it was seldom applied to optimize In-Wheel motors. For exerting the advantages of QGA adequately, a new In-Wheel motor which is similar to the transverse-flux permanent magnet motor (TFPMM) is optimized based QGA. Firstly, the structure and working principle of this motor are introduced. Secondly, the motor torque ripple rate model is established and it could be found that the motor permanent magnet size and the air gap size have a great relationship with its torque ripple. Finally, QGA is adopted to optimize the relevant motor sizes and to obtain a lower torque ripple ratio. The results induce the torque ripple ratio could be reduced by 7.5% just after 9 evolutions and the population size is only 5. That means QGA has a fast convergence speed and global optimization capability, therefore it is applicable for optimizing the similar motors.