轴向磁通切换永磁电机精确铁损计算模型的研究

Research on Accurate Iron Loss Calculation Model of Axial Flux Switching Permanent Magnet Machine

摘要: 由于磁场调制效应，轴向磁通切换永磁电机(Axial Field Flux-Switching Permanent Magnet Machine，简称AFFSPMM)中存在丰富的调制谐波，传统的Bertotti损耗分离模型无法满足铁损的计算精度需求。为了降低损耗，提高能源利用率，本文提出了一种精确、快速的AFFSPMM分频分段变系数铁损计算模型。该模型考虑了调制谐波和铁磁材料非线性饱和特性对AFFSPMM铁损的影响，引入磁滞损耗附加磁通密度低阶项和涡流损耗附加磁通密度高阶项。同时，将磁通密度和频率进行分段，在每个磁通密度或频率区间拟合不同的补偿项系数来补偿磁滞和涡流损耗。最后，基于一种考虑风磨损耗的铁损实测方法对分频分段变系数铁损模型进行实验验证，并将该铁损模型分别与有限元模型和Bertotti损耗分离模型的计算速度进行了对比。结果表明所提出的铁损模型计算值与实测值较为接近，且计算速度相比于有限元法有较大提升。

Abstract: Due to the magnetic field modulation effect, there are abundant modulation harmonics in the axial flux-switching permanent magnet motor (AFFSPMM), and the traditional Bertotti loss separation model cannot meet the calculation accuracy requirements of iron loss. In order to reduce losses and improve energy utilization, an accurate and fast AFFSPMM frequency division and segmented variable coefficient iron loss calculation model is proposed in the paper. This model considers the influence of modulation harmonics and nonlinear saturation characteristics of ferromagnetic materials on the iron loss of AFFSPMM, and introduces the hysteresis loss additional magnetic flux density low-order terms and the eddy current loss additional magnetic flux density high-order terms. At the same time, the magnetic flux density and frequency are segmented, and the different compensation term coefficients are fitted in each magnetic flux density or frequency interval to compensate for hysteresis and eddy current losses. Finally, an actual iron loss measurement method with consideration of friction and windage losses is used to experimentally validate the accuracy of the frequency-divided variable coefficient iron loss model. And the calculation speed of the proposed iron loss model is separately compared with the finite element model and the Bertotti loss separation model. The results show that the calculated value of the proposed iron loss model is close to the measured value, and the calculation speed is greatly improved in contrast with the finite element method.