| 834 | 5 | 538 |
| 下载次数 | 被引频次 | 阅读次数 |
磁性元件在磁能传递、存储和滤波中起着关键作用，直接影响功率变换器的体积、质量、损耗及成本。因此，准确预测磁芯损耗至关重要。针对励磁波形对磁芯损耗的显著影响，提出了一种基于集成学习的励磁波形分类策略。首先，采用支持向量机（support vector machine,SVM）、随机森林（random forest,RF）和梯度提升决策树（gradient boosting decision tree,GBDT）作为基分类器，通过将分类结果与原始特征结合构建新的特征集，并使用元分类器进行训练以提升模型的泛化能力；然后，选择XGBoost作为磁芯损耗预测的核心模型；最后，通过遗传算法进行多目标优化，寻找到最小磁芯损耗与最大传输磁能的最佳工况。实验结果表明：提出的集成学习分类模型能够准确分类励磁波形，XGBoost模型相较于传统磁芯损耗预测模型及其他机器学习模型，展现了更高的预测精度和拟合效果。优化后的模型成功实现了磁芯损耗最小化与传输磁能最大化的平衡。
Abstract：Magnetic components play a key role in energy transfer, storage, and filtering, directly affecting the size,weight, loss, and cost of power converters. Therefore, accurate prediction of core loss is essential. To address the significant influence of excitation waveforms on core loss, an ensemble learning-based waveform classification strategy is proposed. Support vector machine(SVM), random forest(RF), and gradient boosting decision tree(GBDT) are used as base classifiers. The classification outputs are combined with original features to construct a new feature set, which is then used to train a meta-classifier to enhance generalization. XGBoost is selected as the core model for core loss prediction. A genetic algorithm is applied for multi-objective optimization to identify the optimal operating condition with minimal core loss and maximal magnetic energy transfer. Experimental results show that the ensemble classification model can accurately classify excitation waveforms. Compared with traditional core loss prediction models and other machine learning methods, the XGBoost model demonstrates higher prediction accuracy and better regression performance. The optimized framework demonstrates the capability to meet both loss reduction and energy efficiency objectives.
[1] SAFAYATULLAH M, ELRAIS M T, GHOSH S, et al. A comprehensive review of power converter topologies and control methods for electric vehicle fast charging applications[J]. IEEE Access, 2022, 10:40753-40793.
[2] HONG W Z, LIU H J, LIU F, et al. Improved calculation of magnetic hysteresis loss of stacked superconducting cable under T-a formulation[J]. IEEE Transactions on Applied Superconductivity, 2022, 32(6):5900905.
[3] JIN J M. Introduction to magnetic resonance imaging[M]//Electromagnetic Analysis and Design in Magnetic Resonance Imaging. London:Routledge, 2018:1-37.
[4] DOGARIU E, LI H R, SERRANO LóPEZ D, et al. Transfer learning methods for magnetic core loss modeling[C]//Proceedings of the 2021 IEEE 22nd Workshop on Control and Modelling of Power Electronics(COMPEL), November2-5, 2021, Cartagena, Colombia. New York:IEEE Xplore,2021:1-6.
[5] VENKATACHALAM K, SULLIVAN C R, ABDALLAH T,et al. Accurate prediction of ferrite core loss with nonsinusoidal waveforms using only Steinmetz parameters[C]//Proceedings of the 2002 IEEE Workshop on Computers in Power Electronics, June 3-4, 2002, Mayaguez, PR, USA.New York:IEEE Xplore, 2002:36-41.
[6] NABIH A, JIN F, GADELRAB R, et al. Characterization and mitigation of dimensional effects on core loss in highpower high-frequency converters[J]. IEEE Transactions on Power Electronics, 2023, 38(11):14017-14036.
[7]汪晶慧，卢志诚，陈为.高低频复合励磁磁心损耗量化的研究[J].中国电机工程学报，2024, 44(10):4091-4103.WANG J H, LU Z C, CHEN W. Research on quantization of core losses of high and low frequency composite excitation[J]. Proceedings of the CSEE, 2024, 44(10):4091-4103.(in Chinese)
[8] DONG X B, YU Z W, CAO W M, et al. A survey on ensemble learning[J]. Frontiers of Computer Science, 2020,14(2):241-258.
[9] PISNER D A, SCHNYER D M. Support vector machine[M]//Machine Learning. Amsterdam:Elsevier, 2020:101-121.
[10] PARMAR A, KATARIYA R, PATEL V. A review on random forest:an ensemble classifier[M]//International Conference on Intelligent Data Communication Technologies and Internet of Things(ICICI)2018. Cham:Springer International Publishing, 2018:758-763.
[11] FENG J, YU Y, ZHOU Z H. Multi-layered gradient boosting decision trees[EB/OL].(2018-05-31)[2024-06-22]. http://doi.org/10.48550/arXiv.1806.00007.
[12] CHEN T Q, GUESTRIN C. XGBoost:a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA. New York:ACM,2016:785-794.
[13] REEVES C R. Genetic algorithms[M]//GENDREAU M,POTVIN J Y. Handbook of Metaheuristics. 2nd ed. New York:Springer, 2010:109-139.
[14]沈春城，严柏平，黄大卓，等.基于波形复杂特性的励磁涌流快速识别算法研究[J].电气工程学报，2024, 19(1):243-253.SHEN C C, YAN B P, HUANG D Z, et al. Research on fast identification algorithm of inrush current based on complex characteristics of waveform[J]. Journal of Electrical Engineering, 2024, 19(1):243-253.(in Chinese)
[15]赵志刚，徐曼，胡鑫剑，等.考虑磁通波形特征的谐波磁损耗计算方法及验证[J].电网技术，2021, 45(2):811-817.ZHAO Z G, XU M, HU X J, et al. Calculation and verification of harmonic magnetic loss considering magnetic flux waveform characteristics[J]. Power System Technology,2021, 45(2):811-817.(in Chinese)
[16]叶建盈，陈为，汪晶慧. PWM波及直流偏磁励磁下磁芯损耗模型研究[J].中国电机工程学报，2015, 35(10):2601-2606.YE J Y, CHEN W, WANG J H. Research on the core loss model under PWM wave and DC bias excitations[J]. Proceedings of the CSEE, 2015, 35(10):2601-2606.(in Chinese)
[17]旷建军，郑力新，卢小芬，等.非正弦励磁下磁芯损耗的计算[J].磁性材料及器件，2009, 40(1):44-46.KUANG J J, ZHENG L X, LU X F, et al. Calculation of core losses under nonsinusoidal excitation[J]. Journal of Magnetic Materials and Devices, 2009, 40(1):44-46.(in Chinese)
[18] DEB K, SINDHYA K, HAKANEN J. Multi-objective optimization[M]//Decision Sciences. Taylor&Francis Group,6000 Broken Sound Parkway NW, Suite 300, Boca Raton, FL 33487-2742:CRC Press, 2016:145-184.
[19]张丽萍，陈为，汪晶慧.速饱和电感磁芯损耗测量与模型研究[J].中国电机工程学报，2018, 38(15):4593-4600.ZHANG L P, CHEN W, WANG J H. Research on the measurement and modeling of the core loss of fast saturable inductor[J]. Proceedings of the CSEE, 2018, 38(15):4593-4600.(in Chinese)
[20] REINERT J, BROCKMEYER A, de DONCKER R W.Calculation of losses in Ferro-and ferrimagnetic materials based on the modified Steinmetz equation[C]//Proceedings of the Conference Record of the 1999 IEEE Industry Applications Conference, Thirty-Forth IAS Annual Meeting,October 3-7, 1999, Phoenix, AZ, USA. New York:IEEE,1999:2087-2092.
[21] MACKIEWICZ A, RATAJCZAK W. Principal components analysis(PCA)[J]. Computers&Geosciences, 1993, 19(3):303-342.
[22]孟瑞，旷建军.最小二乘法在磁芯损耗参数拟合中的应用[J].现代科学仪器，2013(1):91-92.MENG R, KUANG J J. The application of the method of least squares parameter fitting of core loss[J]. Modern Scientific Instruments, 2013(1):91-92.(in Chinese)
基本信息:
DOI：10.12194/j.ntu.20241023001
中图分类号:TP181;TM46
引用信息:
[1]姚启达,平鹏,朱心怡,等.基于机器学习的磁性元件磁芯损耗预测方法[J].南通大学学报(自然科学版),2025,24(02):29-38.DOI:10.12194/j.ntu.20241023001.
基金信息:
国家自然科学基金青年科学基金项目（52202496）