基于U-net卷积神经网络的电磁场快速计算方法

doi:10.19595/j.cnki.1000-6753.tces.230650

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (14141 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract During the engineering case design session, Finite Element Method (FEM) analysis is essential to address the physical fields of the study object. However, the significant number of degrees of freedom often results in lengthy computation times for FEM, particularly when solving 3D problems involving coupled multi-physical fields. In design-oriented problems like battery structure optimization, finding the optimal solution typically requires thousands of iterations. If the FEM method is employed, this process could take months to complete, significantly delaying project schedules. Furthermore, it is crucial to assess equipment status based on sensor data such as motor digital twin modeling and transformer hot spot temperature real-time calculation. Utilizing the time-consuming finite element method in these scenarios may lead to delayed identification and resolution of failures, potentially causing further harm.
A deep learning-based fast computation method for physical fields is proposed by combining FEM with the U-net convolutional neural network (U-net CNN), studying the effectiveness of this approach in electromagnetic fields. Firstly, we employ the finite element method to model and simulate the research object based on the actual physical model. Secondly, we export the obtained simulation results and transform the model by dividing, rasterizing, and converting it into a point cloud. These modifications are based on the geometric parameters, boundary conditions, and physical field solution results. Thirdly, we train the U-net CNN by optimizing its network parameters and utilize the trained model for swift physical field calculations of the research object. The trained model allows for rapid computation of the physical field.
To assess the practicality of our method, we select a two-dimensional insulator uniform pressure ring model and a three-dimensional transformer model to calculate the electrostatic and magnetic fields. In the two-dimensional model, we employ both single-channel and multi-channel inputs. The single-channel input solely consists of the geometric model parameters, while the multi-channel input includes additional data including the radius and uplift of the size equalizing ring. To minimize memory space, we merge the multi-channel inputs using a Gaussian distribution. The results demonstrate that the single-channel U-net CNN achieves the accuracy of 99.88% for potential and 99.52% for electric field strength. Meanwhile, the FCNs-16 model achieves a potential accuracy of 99.15% and an electric field strength of 98.33%. The potential accuracy attained with the multi-channel input is 99.93%, with an electric field strength of 99.52%. The potential accuracy achieved with the FCNs-16 model is 99.80%, along with an electric field strength of 99.24%. Additionally, the computation time for the U-net CNN is 0.017 s for the single-channel input and 0.016 s for the multi-channel input both significantly faster than the finite element method. Finally, by reducing the size of the dataset, the network's prediction accuracy remains above 90% with 306 groups of data, and even with just 203 groups, it maintains an accuracy of over 85%. In the three-dimensional model, we address the magnetic field of the transformer at t=0.05 s through field-path coupling. The inputs for this model consist of Sampling pointcoordinates and voltage. The results reveal that the U-net CNN achieves the highest accuracy for magnetic induction intensity at 99.26%, while the FCNs-16 model achieves an accuracy of 98.91%. By reducing the dataset size, the model can still maintain a high prediction accuracy, even with only 193 groups of data.
The method possesses the capability to be employed in real-time calculations for equipment condition assessment, as well as in design sessions necessitating multiple iterations.

Key words： Electromagnetic fields convolutional neural network (CNN) fast calculation finite element method (FEM)

Received: 06 May 2023

PACS:

TM15

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Zhang Yujiao
	Zhao Zhitao
	Xu Bin
	Sun Hongda
	Huang Xiongfeng

Cite this article:

Zhang Yujiao,Zhao Zhitao,Xu Bin等. Fast Calculation Method of Electromagnetic Field Based on U-Net Convolutional Neural Network[J]. Transactions of China Electrotechnical Society, 2024, 39(9): 2730-2742.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.230650 OR https://dgjsxb.ces-transaction.com/EN/Y2024/V39/I9/2730

[1] 赵升, 舒亮, 吴自然, 等. 电磁-永磁复合型接触器数值仿真与控制方法[J]. 电工技术学报, 2020, 35(5): 1083-1092.
Zhao Sheng, Shu Liang, Wu Ziran, et al.Numerical simulation and control method of electromagnetic- permanent composite contactor[J]. Transactions of China Electrotechnical Society, 2020, 35(5): 1083-1092.
[2] 阮江军, 张宇, 张宇娇, 等. 电气设备电磁多物理场数值仿真研究与应用[J]. 高电压技术, 2020, 46(03): 739-756.
RuanJiangjun, Zhang Yu, Zhang Yujiao, et al. Numerical simulation research and applications of electromagnetic multiphysical field for electrical equipment[J]. High Voltage Engineering, 2020, 46(3): 739-756.
[3] Liu C, Hu S, Han K, et al.electric-field distribution and insulation status of ±800 kV UHVDC converter valve after implanting full-view micro-sensor detector[J]. IEEE Access, 2019, 201(7): 86534-86544.
[4] 曹龙飞, 范兴纲, 李大伟, 等. 基于快速有限元的永磁电机绕组涡流损耗半解析高效计算[J]. 电工技术学报, 2023, 38(1): 153-165.
Cao Longfei, Fan Xinggang, Li Dawei, et al.Semi analytical and efficient calculation method of eddy current loss in windings of permanent magnet machines based on fast finite element method[J]. Transactions of China Electrotechnical Society, 2023, 38(1): 153-165.
[5] 黄永宁, 余晓, 樊益平, 等. 363 kV多断口快速真空断路器电场仿真分析[J]. 高压电器, 2019, 55(10): 19-24, 32.
Huang Yongning, Yu Xiao, Fan Yiping, et al.Electric Field simulation of the 363 kV multi-break fast vacuum circuit breaker[J]. High Voltage Apparatus, 2019, 55(10): 19-24, 32.
[6] 杜志叶, 田雨, 金硕, 等. 大规模复杂模型电场计算的快速精细化有限元数值建模[J]. 中国电机工程学报, 2018, 38(24): 7166-7173, 7443.
Du Zhiye, Tian Yu, Jin Shuo, et al.A fast finite element numerical modeling method for the calculation of electric field in large scale complex model[J]. Proceedings of the CSEE, 2018, 38(24): 7166-7173, 7443.
[7] 狄冲, 鲍晓华, 潘晋, 等. 基于Elmer开源有限元平台的铁氧体辅助同步磁阻电机的建模和分析[J].电工技术学报, 2022, 37(5): 1136-1144.
Di Chong, Bao Xiaohua, Pan Jin, et al.modelling and analysis of a ferrite assisted synchronous reluctance machine based on the open-source platform Elmer[J]. Transactions of China Electrotechnical Society, 2022, 2022, 37(5): 1136-1144.
[8] 赵玉顺, 戴义贤, 庄加才, 等. 基于热固耦合的中频变压器绝缘材料性能参数优化配合方法[J]. 电工技术学报, 2023, 38(4): 1051-1063.
Zhao Yushun, Dai Yixian, Zhuang Jiacai, et al.optimization of insulation material performance parameters for medium frequency transformers based on thermosolid coupling[J]. Transactions of China Electrotechnical Society, 2023, 38(04): 1051-1063.
[9] 刘素贞, 王淑娟, 张闯, 等. 钢板电磁超声表面波的仿真分析及缺陷定量检测[J]. 电工技术学报, 2020, 35(1): 97-105.
Liu Suzhen, Wang Shujuan, Zhang Chuang, et al.Simulation analysis of electromagnetic acoustic surface wave of steel plate and quantitative defect detection[J]. Transactions of China Electrotechnical Society, 2020, 35(1): 97-105.
[10] 于丰源, 陈昊, 闫文举, 等. 宽窄定子极轴向磁通开关磁阻电机的设计与分析[J]. 电工技术学报, 2023, 38(5): 1261-1274.
Yu Fengyuan, Chen Hao, Yan Wenju, et al.Design and characteristic analysis of a wide-narrow stator poles axial flux switched reluctance machine[J]. Transactions of China Electrotechnical Society, 2023, 38(05): 1261-1274.
[11] 刘利强, 尹彦博, 齐咏生, 等. 一种多模型融合的风电系统永磁同步发电机数字孪生建模方法[J].电机与控制学报, 2023, 27, 17(11): 149-162.
LiuLiqiang, Yin Yanbo, Qi Yongsheng, et al. A digital twin modeling method of wind power system permanent magnet synchronous generator based on multi model fusion[J]. Electric Machines and Control, 2023, 27, 17(11): 149-162.
[12] Li Ning, Liu Yongqian, Li Li, et al.Numerical simulation of wind turbine wake based on extended k-epsilon turbulence model coupling with actuator disc considering nacelle and tower[J]. IET Renewable Power Generation, 2020, 14(1): 3834-3842.
[13] 胡万君, 刘刚, 朱章宸, 等. 油浸式电力变压器绕组稳态温升降阶计算方法研究[J].中国电机工程学报, 2023, 43(16): 6505-6517.
Hu Wanjun, Liu Gang, Zhu Zhangchen.Reduced order calculation method of steady temperature rise of oil immersed power transformer[J]. Proceedings of the CSEE, 2023, 43(16): 6505-6517.
[14] 张翼, 朱永利. 图信号与图卷积网络相结合的局部放电模式识别方法[J]. 中国电机工程学报, 2021, 41(18): 6472-6481.
Zhang Yi, Zhu Yongli.A partial discharge pattern recognition method combining graph signal and graph convolutional network[J]. Proceedings of the CSEE, 2021, 41(18): 6472-6481.
[15] 贾光辉, 于云瑞, 王丹. 卷积神经网络求解有限元单元刚度矩阵[J]. 北京航空航天大学学报, 2020, 46(3): 481-487.
Jia Guanghui, Yu Yunrui, Wang Dan.Solving finite element stiffness matrix based on convolutional neural network[J]. Journal of Beijing University of Aeronautics and Astronautics, 2020, 46(3): 481-487.
[16] Bhaduri A, Gupta A, Graham-Brady L. Stress field prediction in fiber-reinforced composite materials using a deep learning approach[J/OL]. ArXiv, 2021: 2111.05271. https://arxiv.org/abs/2111.05271v1.
[17] Thuerey N, Weissenow K, Prantl L, et al.Deep learning methods for Reynolds-averaged Navier-Stokes simulations of airfoil flows[J]. AIAA Journal, 2020, 58(1): 25-36.
[18] Mendizabal A, Márquez-Neila P, Cotin S. Simulation of hyperelastic materials in real-time using deep learning[J]. Medical Image Analysis, 2020, 59: 101569(1-11).
[19] Kim D J, Kim S I, Kim H S. Thermal simulation trained deep neural networks for fast and accurate prediction of thermal distribution and heat losses of building structures[J]. Applied Thermal Engineering, 2022, 202: 117908(1-11).
[20] Zhang H, Wang Y, Lu K, et al. SAP-Net: deep learning to predict sound absorption performance of metaporous materials[J]. Materials & Design, 2021, 212: 110156(1-7).
[21] 黄巨挺, 高宏力, 戴志坤. 基于编码解码结构的移动端电力线语义分割方法[J]. 计算机应用, 2021, 41(10): 2952-2958.
Huang Junting, Gao Hongli, Dai Zhikun.Semantic segmentation method of power line on mobile terminals based on encoder-decoder structure[J]. Journal of Computer Applications, 2021, 41(10): 2952-2958.
[22] 裴少通, 杨家骏, 马子儒, 等. 基于深度学习的输变电设备紫外放电光斑分割方法[J]. 科学技术与工程, 2022, 22(33): 14759-14766.
Pei Shaotong, Yang Jiajun, Ma Ziru, et al.transmission and distribution equipment ultraviolet discharge spot segmentation method based on deep learning[J]. Science Technology and Engineering, 2022, 22(33): 14759-14766.
[23] Siddique N, Paheding S, Elkin C P, et al.U-Net and its variants for medical image segmentation: a review of theory and applications[J]. IEEE Access, 2021, 9: 82031-82057.
[24] Charles R Q, Su H, Kaichun M, et al.PointNet: deep learning on point sets for 3D classification and segmentation[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA, 2017: 77-85.
[25] Balasundaram S, Prasad S C.Robust twin support vector regression based on Huber loss function[J]. Neural Computing and Applications, 2020, 32(15): 11285-11309.
[26] 王思华, 王军军, 赵磊, 等. 污秽成分对复合绝缘子表面电场的影响[J]. 中国电力, 2021, 54(7): 149-157.
Wang Sihua, Wang Junjun, Zhao Lei, et al.influence of pollution components on surface electric field of composite insulators[J]. Electric Power, 2021, 54(7): 149-157.
[27] 赵建利, 姚顺, 岳永刚, 等. 500 kV SF₆瓷质套管多工况仿真与故障分析[J]. 电工技术学报, 2021, 36(增刊2): 736-745.
Zhao Jianli, Yao Shun, Yue Yonggang, et al.Simulation and failure analysis of 500 kV SF₆ porcelain bushing under complicated working conditions[J]. Transactions of China Electrotechnical Society, 2021, 36(S2): 736-745.
[28] 鲁尧, 咸日常, 王立权, 等. 干式空心串联电抗器绕组匝间短路电磁特性研究[J]. 高压电器, 2021, 57(11): 116-123.
Lu Yao, Xian Richang, Wang Liquan, et al.Study on electromagnetic characteristics of winding inter-turn short circuit of dry-type air-core series reactor[J]. High Voltage Apparatus, 2021, 57(11): 116-123.
[29] 潘超, 米俭, 王格万, 等. 基于场路耦合的变压器绕组匝间短路电磁谐响应分析方法[J]. 电工技术学报, 2019, 34(4): 673-682.
Pan Chao, Mi Jian, Wang Gewan, Cai Guowei, et al.electromagnetic harmonic response analysis method of inter-turn short circuit in transformer winding based on field circuit coupling[J]. Transactions of China Electrotechnical Society, 2019, 34(4): 673-682.
[30] Park Y R, Seol Y J, Kim Y J, et al.Comparison on the deep learning performance of a field of view variable color images of uterine cervix[C]//International Forum on Medical Imaging in Asia, Taipei, 2021: 99-102.