Low spatial back electromotive force harmonics design for fractional-slot winding permanent-magnet motor
Fractional-slot winding permanent magnet motors(PM)are widely applied due to their unique advantages,with the most notable being their high-quality sinusoidal no-load back electromotive force(EMF)waveform.This characteristic not only enhances the control performance of the motor system but also significantly reduces torque ripple and operational losses,thus improving overall motor quality.Firstly,a comparative analysis of the structural character-istics and magnetic circuit features of integer-slot and fractional-slot winding PM motors was conducted.Next,based on winding functions and air-gap flux density functions,a general expression for the no-load back-EMF of permanent magnet motors was derived.From the perspective of spatial harmonics,the intrinsic impact of the fractional-slot win-ding structure on optimizing the no-load back EMF waveform of PM motors is elucidated.To validate the theoretical analysis,an optimized design of a 24-slot 8-pole integer-slot winding PM motor and a 27-slot 8-pole fractional-slot winding PM motor is performed.The winding factor harmonics,air-gap flux density harmonics,no-load back EMF,and torque performance of both motors were comparatively analyzed.The findings indicate that the k-th harmonic of the no-load back EMF is generated by the no-load air gap magnetic density harmonic of the order kpr-th(where pr is the number of PM pole pairs)and the winding factor harmonic of the order v=kpr.The presence of more harmonics of the same order in the winding factor and air-gap flux density increases the harmonic content in the no-load back EMF.Finally,a prototype of the 27-slot 8-pole fractional-slot winding PM motor was developed and partially tested,measuring the motor's no-load back-EMF waveform,phase current,and torque performance.These experimental re-sults were compared with finite element predictions to confirm the accuracy of both the theoretical analysis and the fi-nite element simulations.
国家科技期刊平台
NSTL
万方数据
