Simulation and Optimization of Permanent Magnet Linear Machine Based on Deep Neural Network

doi:10.16182/j.issn1004731x.joss.23-0211

Abstract

Abstract:

The finite element model (FEM) of permanent magnet linear synchronous machines (PMLSMs) takes a long computing time and cannot directly display the relationship between structural parameters and output thrust, thus failing to guide the structural parameter optimization of the machine. An improved simulation model of PMLSMs based on the subdomain analytical method and deep neural network (DNN) algorithm is proposed. The magnetic flux density, no-load counter electromotive force (EMF), and other data are obtained according to Maxwell's equations. The nonlinear relationship between the structural parameters of the machine and output thrust is fitted by the DNN algorithm. Based on this model, an adaptive genetic algorithm is used to optimize the thrust density of PMLSMs, and the results are compared with the finite element simulation. The results show that the computing speed of the improved simulation model of PMLSMs is 87.1 times that of the FEM. The average error of thrust between these two models is 2.87%, and the optimized thrust density of the machine is increased by 5.7%.

Key words: permanent magnet linear synchronous machine, subdomain analytical method, deep neural network, adaptive genetic algorithm, thrust optimization

CLC Number:

TM351

Yan Shiliang, Wang Yinling, Lu Dandan, Pan Xiaoqin. Simulation and Optimization of Permanent Magnet Linear Machine Based on Deep Neural Network[J]. Journal of System Simulation, 2024, 36(3): 713-725.

Figures/Tables 24

Fig. 1

Table 1

Fig. 2

Fig. 3

Table 2

Fig. 4

Fig. 5

Fig. 6

Fig. 7

Fig. 8

Table 3

Sample data of PMLSM

序号	结构参数				推力性能
序号	τ_s/mm	h_s/mm	h_m/mm	α_p	$F ¯$ /N	$γ$ /%
1	11.4	40.0	4.5	0.74	285.2	19.3
2	11.6	40.0	4.5	0.74	290.6	18.9
3	11.8	40.0	4.5	0.74	296.7	18.6
$⋮$
20 734	13.6	42.2	5.6	0.83	424.5	13.6
20 735	13.6	42.2	5.6	0.84	429.8	13.4
20 736	13.6	42.2	5.6	0.85	435.6	13.1

Table 3

Fig. 9

Table 4

Fig. 10

Fig. 11

Fig. 12

Table 5

Fig. 13

Fig. 14

Table 6

Fig. 15

Table 7

Optimized parameter and thrust performance of machine

结构参数	数值	推力性能	数值
τ_s/mm	12.1	$F ¯$ /N	371.7
h_s/mm	41.6	$γ$ /%	14.6
h_m/mm	5.2	推力密度/(N/mm³)	3.483
α_p	0.76	推力密度/(N/mm³)	3.483

Table 7

Fig. 16

Fig. 17

References 21

1	曹瑞武, 张铮, 金毅, 等. 次级无轭部双边磁通切换永磁直线电机及其控制[J]. 中国电机工程学报, 2017, 37(22): 6585-6593.
	Cao Ruiwu, Zhang Zheng, Jin Yi, et al. Double-sided Linear Flux-Switching Permanent Magnet Motor with Yokeless Secondary and Control System[J]. Proceedings of the CSEE, 2017, 37(22): 6585-6593.
2	寇发荣, 方涛, 任全, 等. 直线电机式主动悬架内外环控制研究[J]. 系统仿真学报, 2018, 30(7): 2615-2621.
	Kou Farong, Fang Tao, Ren Quan, et al. Research on Inner and Outer Loop Control of Active Suspension with Linear Motor[J]. Journal of System Simulation, 2018, 30(7): 2615-2621.
3	夏涛. 用于直驱式波浪发电的圆筒型直线电机优化设计及其控制研究[D]. 南京: 东南大学, 2019.
	Xia Tao. Research on Optimal Design and Control of Tubular Linear Generators for Direct Drive Wave Energy Conversion[D]. Nanjing: Southeast University, 2019.
4	周游. 高推力密度磁场调制永磁直线电机拓扑研究[D]. 武汉: 华中科技大学, 2021.
	Zhou You. Research on Flux-Modulation Linear Permanent Magnet Machine Topology with High Thrust Density[D]. Wuhan: Huazhong University of Science and Technology, 2021.
5	邓涛, 邓凇. 分体式开关磁阻双转子电机设计与电磁仿真[J]. 系统仿真学报, 2021, 33(4): 935-942.
	Deng Tao, Deng Song. Design and Electromagnetic Simulation of Split Switched Reluctance Dual-Rotor Motor[J]. Journal of System Simulation, 2021, 33(4): 935-942.
6	Shi Chaojie, Qu Ronghai, Gao Yuting, et al. Design and Analysis of an Interior Permanent Magnet Linear Vernier Machine[J]. IEEE Transactions on Magnetics, 2018, 54(11): 1-5.
7	牛彦昭. 无铁芯永磁同步直线电机的多目标优化设计与研究[D]. 长春: 吉林大学, 2021.
	Niu Yanzhao. Multi-objective Optimization Design and Research on Coreless Permanent Magnet Linear Synchronous Motor[D]. Changchun: Jilin University, 2021.
8	栾茹, 郭金茹. 基于ANSYS的开关磁阻电机负载振动仿真研究[J]. 系统仿真学报, 2017, 29(11): 2760-2766, 2773.
	Luan Ru, Guo Jinru. Simulation on Switch Reluctance Motor Vibration Under Load Condition Based on ANSYS[J]. Journal of System Simulation, 2017, 29(11): 2760-2766, 2773.
9	王明杰, 徐伟, 杨存祥, 等. 基于精确子域模型的永磁直线同步电机空载磁场解析计算[J]. 电工技术学报, 2020, 35(5): 942-953.
	Wang Mingjie, Xu Wei, Yang Cunxiang, et al. Analytical Calculation of No-Load Magnetic Field in Permanent Magnet Linear Synchronous Motors Based on an Accurate Subdomain Model[J]. Transactions of China Electrotechnical Society, 2020, 35(5): 942-953.
10	Zhu Minchen, Wu Lijian, Fang Youtong, et al. Subdomain Model for Predicting Armature Reaction Field of Dual-stator Consequent-pole PM Machines Accounting for Tooth-tips[J]. CES Transactions on Electrical Machines and Systems, 2019, 3(2): 143-150.
11	杨阳, 赵吉文, 宋俊材, 等. 基于深度神经网络模型的无铁心永磁同步直线电机结构优化研究[J]. 中国电机工程学报, 2019, 39(20): 6085-6094.
	Yang Yang, Zhao Jiwen, Song Juncai, et al. Structural Optimization of Air-Core Permanent Magnet Synchronous Linear Motors Based on Deep Neural Network Models[J]. Proceedings of the CSEE, 2019, 39(20): 6085-6094.
12	周杨, 赵静, 汪伟涛, 等. 基于BP神经网络遗传优化的电机推力品质研究[J]. 传感器与微系统, 2022, 41(1): 58-61.
	Zhou Yang, Zhao Jing, Wang Weitao, et al. Research on Motor Thrust Quality Based on BPNN Genetic Optimization[J]. Transducer and Microsystem Technologies, 2022, 41(1): 58-61.
13	Wu L J, Zhu Z Q, Staton D, et al. An Improved Subdomain Model for Predicting Magnetic Field of Surface-Mounted Permanent Magnet Machines Accounting for Tooth-Tips[J]. IEEE Transactions on Magnetics, 2011, 47(6): 1693-1704.
14	Gysen B L J, Meessen K J, Paulides J J H, et al. General Formulation of the Electromagnetic Field Distribution in Machines and Devices Using Fourier Analysis[J]. IEEE Transactions on Magnetics, 2010, 46(1): 39-52.
15	夏加宽, 沈丽, 彭兵, 等. 齿槽效应对永磁直线伺服电机法向力波动的影响[J]. 中国电机工程学报, 2015, 35(11): 2847-2853, 后插24.XiaJiakuan, ShenLi, PengBing, et al. Influences of Cogging Effects on Normal Force Ripple in Permanent Magnet Linear Servo Motors[J]. Proceedings of the CSEE, 2015, 35(11): 2847-2853, S24.
16	张博远, 龚光红, 王泽, 等. 基于深度网络的蓝军装备模型参数生成方法研究[J]. 系统仿真学报, 2022, 34(12): 2629-2638.
	Zhang Boyuan, Gong Guanghong, Wang Ze, et al. Research on Parameter Construction Method of Blue Army Equipment Model Based on a Deep Network[J]. Journal of System Simulation, 2022, 34(12): 2629-2638.
17	He Liange, Wu Xinyang, Nie Yuanhang, et al. Loss Prediction of Vehicle Permanent Magnet Synchronous Motor Based on Deep Learning[J]. Journal of Electrical Engineering & Technology, 2023, 18(2): 1053-1063.
18	温江涛, 闫常弘, 孙洁娣, 等. 基于压缩采集与深度学习的轴承故障诊断方法[J]. 仪器仪表学报, 2018, 39(1): 171-179.
	Wen Jiangtao, Yan Changhong, Sun Jiedi, et al. Bearing Fault Diagnosis Method Based on Compressed Acquisition and Deep Learning[J]. Chinese Journal of Scientific Instrument, 2018, 39(1): 171-179.
19	Song Juncai, Dong Fei, Zhao Jiwen, et al. A New Regression Modeling Method for PMSLM Design Optimization Based on K-Nearest Neighbor Algorithm[C]//2017 20th International Conference on Electrical Machines and Systems (ICEMS). [S.l.]: IEEE, 2017: 1-5.
20	陈芮莹, 王承涛, 刘振元. 考虑服务水平的IT运维人员调度的智能遗传算法[J]. 系统仿真学报, 2021, 33(3): 732-744.
	Chen Ruiying, Wang Chengtao, Liu Zhenyuan. Intelligent Genetic Algorithm for Workforce Scheduling Considering Service Level in IT Maintenance Service[J]. Journal of System Simulation, 2021, 33(3): 732-744.
21	张聪, 许晓慧, 孙海顺, 等. 基于自适应遗传算法的规模化电动汽车智能充电策略研究[J]. 电力系统保护与控制, 2014, 42(14): 19-24.
	Zhang Cong, Xu Xiaohui, Sun Haishun, et al. Smart Charging Strategy of Large-Scale Electric Vehicles Based on Adaptive Genetic Algorithm[J]. Power System Protection and Control, 2014, 42(14): 19-24.

结构参数	初始值	优化范围
极槽比	6槽5极	—
横向长度L/mm	55	—
次级铁芯高度h_c/mm	12	—
次级极距τ_p/mm	27	—
永磁体高度h_m/mm	5	4.5~5.6
极弧系数α_p	0.8	0.74~0.85
气隙长度g/mm	1.5	—
一个周期解析域长度τ_k/mm	270
初级齿槽宽度τ_s/mm	12.5	11.4~13.6
初级齿槽深度h_s/mm	41	40.0~42.2

子域编号	物理空间	子域编号	物理空间
①	顶端空气域	④	永磁体域
②	初级端部域	⑤	绕组线圈域
③	气隙域

名称	值
周期epoch	95
批大小batch-size	128
学习率α	1e-3
一阶矩指数衰减率β₁	0.9
一阶矩指数衰减率β₂	0.999
ε	1e-8
暂退概率drop rate	0.3、0.5、0.2

模型	R²
DNN	0.96~0.97
KNN	0.90~0.94

[1]	Sijin Zhou, Dicheng Chen, Geng Tu, Dazhi Jiang. Multi-modality Affective Computing Model Based on Personality and Memory Mechanism [J]. Journal of System Simulation, 2022, 34(4): 745-758.
[2]	Zhou Jin. Joint Low Rank and Sparsity-based Channel Estimation for FDD Massive MIMO [J]. Journal of System Simulation, 2021, 33(1): 99-108.
[3]	Xing Zhiwei, Li Biao, Zhu Hui, Luo Qian. Research on Flight Ground Service Time Prediction Based on Deep Neural Network [J]. Journal of System Simulation, 2020, 32(4): 678-686.
[4]	Zhang Yunjing, Tang Guangming, Xu Xiaoyu. Decision-making Method for Formulating Spares Reserve Scheme Based on Deep Neural Network [J]. Journal of System Simulation, 2019, 31(11): 2238-2246.
[5]	Wang Heng, Xu Yaxing, Wang Zhenfeng, Zhou Tianpeng, Tian Dechun. Distribution Routing Optimization of Fresh Agricultural Products Based on Road Conditions [J]. Journal of System Simulation, 2019, 31(1): 126-135.
[6]	Wang Yan, Cheng Lijun. Event Triggered Optimization Method for Dynamic Task Decomposition Mode in Cloud Fusion [J]. Journal of System Simulation, 2018, 30(11): 4029-4042.
[7]	Wang Kun, Zhang Yang, Song Shengbo, Zhuge Jingchang. Infrared Image Segmentation Based on Improved Dimensional OTSU and Genetic Algorithm [J]. Journal of System Simulation, 2017, 29(6): 1229-1236.