MagNet Challenge for Data-Driven Power Magnetics Modeling
Minjie Chen, Haoran Li, Shukai Wang, Thomas Guillod, Diego Serrano, Nikolas Förster, Wilhelm Kirchgässner, Till Piepenbrock, Oliver Schweins, Oliver Wallscheid, Qiujie Huang, Yang Li, Yu Dou, Bo Li, Sinan Li, Emmanuel Havugimana, Vivek Thomas Chacko, Sritharini Radhakrishnan, Mike K. Ranjram, Bailey Sauter, Skye Reese, Shivangi Sinha, Lizhong Zhang, Tom McKeague, Binyu Cui, Navid Rasekh, Jun Wang, Song Liu, Alfonso García Martínez, Xinyu Liu, Chaoying Mei, Rui Zhao, Gaoyuan Wu, Hao Wu, Rui Zhang, Hao Song, Lie Zhang, Yibo Lu, Lijun Hang, Neha Rajput, Himanshu Bhusan Sandhibigraha, Neeraj Agrawal, Vishnu Mahadeva Iyer, Xiaobing Shen, Fanghao Tian, Qingcheng Sui, Jiaze Kong, Wilmar Martínez, Asier Arruti, Borja Alberdi, Anartz Agote, Iosu Aizpuru, Minmin Zhang, Xia Chen, Yuchen Dong, Duo Wang, T. Shen, Yan Zhou, Yaohua Li, Sicheng Wang, Yue Wu, Yongbin Jiang, Ziheng Xiao, Yi Tang, Yun-Shan Hsieh, Jian-De Li, Lichen Yu, Tzu-Chieh Hsu, Yuchen Liu, Chih‐Hsien Hsia, Chen Chen, Alessio Giuffrida, Nicolò Lombardo, Fabio Marmello, Simone Morra, Marco Pasquale, Luigi Solimene, C. Ragusa, Jacob Reynvaan, Martin Stoiber, Chengbo Li, Wei Qin, Xiang Ma, Boyu Zhang, Zheng Wang, Ming Cheng, Wei Xu, Jiyao Wang, Youkang Hu, Jing Xu, Zhongqi Shi, Dixant Bikal Sapkota, Puskar Neupane, Mecon Joshi, Shahabuddin Khan, Bowen Su, Yunhao Xiao, Min Yang, Kai Sun, Zhengzhao Li
Abstract
This article summarizes the main results and contributions of the MagNet Challenge 2023, an open-source research initiative for data-driven modeling of power magnetic materials. The MagNet Challenge has (1) advanced the state-of-the-art in power magnetics modeling; (2) set up examples for fostering an open-source and transparent research community; (3) developed useful guidelines and practical rules for conducting data-driven research in power electronics; and (4) provided a fair performance benchmark leading to insights on the most promising future research directions. The competition yielded a collection of publicly disclosed software algorithms and tools designed to capture the distinct loss characteristics of power magnetic materials, which are mostly open-sourced. We have attempted to bridge power electronics domain knowledge with state-of-the-art advancements in artificial intelligence, machine learning, pattern recognition, and signal processing. The MagNet Challenge has greatly improved the accuracy and reduced the size of data-driven power magnetic material models. The models and tools created for various materials were meticulously documented and shared within the broader power electronics community.