磁动力
转矩脉动
电磁线圈
电枢(电气工程)
谐波
反电动势
波形
扭矩
控制理论(社会学)
磁场线圈
电气工程
磁场
谐波
磁路
工程类
物理
计算机科学
声学
电压
直接转矩控制
感应电动机
量子力学
控制(管理)
人工智能
热力学
作者
Shuangchun Xie,Yuefei Zuo,Zaixin Song,Shun Cai,Fawen Shen,Junwei Goh,Boon Siew Han,Chi Cuong Hoang,Christopher H. T. Lee
出处
期刊:IEEE Transactions on Transportation Electrification
日期:2023-03-27
卷期号:10 (1): 2170-2182
被引量:3
标识
DOI:10.1109/tte.2023.3262301
摘要
Magnetic-geared machine (MGM) has been widely investigated as the power split device for hybrid electric vehicles (HEVs) with the inherently high integration of multimechanical ports. It is revealed that the magnetic circuits of armature winding in MGM are essentially asymmetrical for different phases under the magnetic field modulation effect. This inevitably results in unbalanced phase back electromotive force (EMF) waveforms, severe torque ripples, acoustic noise, and difficulty with precise control. To alleviate the asymmetry problem, a generic winding design model is presented for MGMs in this article, which allows for consideration of all potential winding configurations. First, a novel back-EMF harmonic factor is proposed to account for both the harmonic content and asymmetric level in MGM. The genetic algorithm (GA) is further employed to optimize the winding in terms of minimizing the back-EMF harmonic factor, armature magnetomotive force (MMF) harmonic contents, and phase winding resistance. The proposed winding layout exhibits superior filtering capability for three-phase asymmetric magnetic field harmonics. As compared with the conventional integral-slot distributed winding, the MGM with the proposed winding exhibits improved back-EMF waveform symmetry, suppressed torque ripple, and reduced core losses. Finally, an experimental prototype is manufactured to validate the effectiveness of asymmetric magnetic field suppression for MGMs. The results show the investigated winding design method is an effective solution to the asymmetric issue of MGMs, paving the way to research opportunities for further improvements.
科研通智能强力驱动
Strongly Powered by AbleSci AI