CAJ | 학술논문

针对分布式驱动电动汽车主要噪声源-永磁同步电机电磁噪声，通过噪声测试实验测得了电机噪声信号，实验结果表明：电磁噪声具有明显的阶次特征，主要阶次不仅包含了偶数阶、开关频率附近阶次，还包含了大量奇数阶，甚至以往研究中被当做信号毛刺而忽略的分数阶。为解释该实验现象，通过磁势磁导法得到考虑时间谐波电流的气隙磁场分布，应用麦克斯韦张量法获得了电磁径向力波，并沿外转子内表面应用复合柯特斯公式对径向力波积分，求得了具有6阶代数精度的电磁径向集中力，进而应用圆柱壳体理论建立了一种考虑时间谐波电流的永磁同步电机电磁噪声数学预测模型，预测了变频器供电时永磁同步电机电磁噪声的特征阶次：2hi、2hj、2（2μ-1）、（hi ±hj）、（2μ-1±hi ）、（2μ-1±hj）[hi、hj为时间谐波电流阶次，μ∈N*]。当 hi或 hj是分数时，电磁噪声将出现分数阶特征频率。经过与实验结果对比，本模型能精确预测和解释永磁同步电机电磁噪声的特征阶次。
침대분포식구동전동기차주요조성원-영자동보전궤전자조성，통과조성측시실험측득료전궤조성신호，실험결과표명：전자조성구유명현적계차특정，주요계차불부포함료우수계、개관빈솔부근계차，환포함료대량기수계，심지이왕연구중피당주신호모자이홀략적분수계。위해석해실험현상，통과자세자도법득도고필시간해파전류적기극자장분포，응용맥극사위장량법획득료전자경향력파，병연외전자내표면응용복합가특사공식대경향력파적분，구득료구유6계대수정도적전자경향집중력，진이응용원주각체이론건립료일충고필시간해파전류적영자동보전궤전자조성수학예측모형，예측료변빈기공전시영자동보전궤전자조성적특정계차：2hi、2hj、2（2μ-1）、（hi ±hj）、（2μ-1±hi ）、（2μ-1±hj）[hi、hj위시간해파전류계차，μ∈N*]。당 hi혹 hj시분수시，전자조성장출현분수계특정빈솔。경과여실험결과대비，본모형능정학예측화해석영자동보전궤전자조성적특정계차。
Aiming at electromagnetic noise in a permanent magnet synchronous motor (PMSM)as the main noise source of a distributed-drive electric vehicle,firstly,the noise signals of PMSM were measured.The test results showed that the main orders of electromagnetic noise are not only even number orders and the orders in a close vicinity of switching frequency,but also odd number orders and fraction orders ignored as signal burrs in the past.Then,in order to explain the test phenomenon,the air gap magnetic field distribution considering current harmonics was obtained using the way of the magnetomotive multiplied permeance.Radial force wave was obtained with Maxwell’s tensor method and was transformed to a radial concentrate force with the 6 th order algebraic accuracy using the composite Cotes formula.An analytical model of electromagnetic noise in PMSM considering current harmonics was built.Orders and frequencies of electromagnetic noise were predicted.It was shown that they are 2hi ,2hj,2(2μ-1),(hi ±hj),(2μ-1 ±hi )and (2μ-1 ±hj)(hi and hj are orders of current harmonics,μ∈N*);if hi or hj is a fraction,the fraction orders of electromagnetic noise occur;the orders and frequencies of noise of PMSMare accurately predicted with the proposed analytical model.