基于深度强化学习的PMSM匝间短路故障诊断方法

李方丽; 刘 杰; 王延波; 魏海峰; 李垣江

doi:10.11993/j.issn.2096-3920.2021.06.013

基于深度强化学习的PMSM匝间短路故障诊断方法

doi: 10.11993/j.issn.2096-3920.2021.06.013

1.
江苏科技大学电子信息学院, 江苏镇江, 212003
2.
江苏仕净环保科技股份有限公司, 江苏苏州, 215137

基金项目: 国家自然科学基金项目(51977101); 江苏省研究生科研创新计划项目资助(KYCX21_3479).

详细信息

作者简介:
李方丽(1990-), 女, 硕士, 主要研究方向为深度强化学习、故障诊断.

中图分类号: TM351
计量
- 文章访问数: 171
- HTML全文浏览量: 14
- PDF下载量: 75
- 被引次数: 0
出版历程
- 收稿日期: 2021-01-18
- 修回日期: 2021-02-06
- 刊出日期: 2021-12-31

Fault Diagnosis Method of PMSM Inter Turn Short Circuit Based on Deep Reinforcement Learning

1.
School of Electronic Information, Jiangsu University of Science and Technology, Zhenjiang 212003, China
2.
Jiangsu Shijing Environmental Protection Technology Co., Ltd, Suzhou 215137, China

摘要

摘要: 永磁同步电机(PMSM)凭借其稳定性强、损耗低、效率高、体积小、调速范围宽等优势而广泛应用于水下航行推进等领域。Deep-Q-Net-work (DQN)算法进行故障诊断。首先使用Maxwell软件建立PMSM模型并分析电机不同状态下三相电流、最低磁密度和电磁转矩的相关变化, 采集上述五维特征分量构建特征数据集并组成电机故障样本, 然后使用深度强化学习DQN算法对样本集和测试集进行数据训练与分析, 通过调节神经元节点数、迭代次数、学习率和经验回放数等, 提高算法诊断准确率, 最终得到该算法对PMSM故障诊断的准确率高达99.61%, 从而验证了该算法在PMSM故障诊断方面的有效性。
- 永磁同步电机 /
- 匝间短路 /
- 故障诊断 /
- 深度强化学习 /
- Deep-Q-Network算法
Abstract: Permanent magnet synchronous motors(PMSMs) have been widely used in the field of underwater navigation and propulsion due to their advantages of strong stability, low loss, high efficiency, small size, and wide speed range. In this study, the deep Q network(DQN) algorithm in the deep reinforcement learning method is used to diagnose the inter turn short circuit fault of the PMSM. First, Maxwell software is used to establish the PMSM model and analyze the relevant changes in the three-phase domain current, minimum magnetic density, and electromagnetic torque under different states, and the five dimensional characteristic components are collected to construct the characteristic data set and form the motor fault samples. Then, the DQN algorithm is used to train and analyze the data of the sample and test sets. By adjusting the number of neuron nodes, iteration, learning rate, and number of experience playbacks, the accuracy rate obtained of the algorithm for PMSM fault diagnosis is as high as 99.61%, which verifies the effectiveness of the algorithm in the diagnosis of PMSM faults.
- permanent magnet synchronous motor /
- inter turn short circuit /
- fault diagnosis /
- deep reinforcement learning /
- Deep-Q-Network Algorithm

HTML全文

参考文献(27)

[1]	赵涛, 刘明雍, 周良荣. 自主水下航行器的研究现状与挑战[J]. 火力与指挥控制, 2010, 35(6): 1-6. Zhao Tao, Liu Ming-yong, Zhou Liang-rong. A Survey of Autonomous Underwater Vehicle Recent Advances and Future Challenges[J]. Fire Control ＆ Command Control, 2010, 35(6): 1-6.
[2]	徐海珠, 谢顺依, 连军强, 等. 对转鱼雷永磁同步推进电机直接转矩控制系统设计[J]. 鱼雷技术, 2012, 20(1): 51-55. Xu Hai-zhu, Xie Shun-yi, Lian Jun-qiang, et al. A Direct Torque Control System of Counter-rotating PMSM for Torpedo Prolulsion[J]. Torpedo Technology, 2012, 20(1): 51-55.
[3]	钱鑫, 王崇武, 于琨琨, 等. 基于递推最小二乘法的永磁同步电机参数辨识[J]. 鱼雷技术, 2014, 22(6): 452- 456. Qian Xin, Wang Chong-wu, Yu kun-kun, et al. Parameters Identification of Permanent Magnet Synchronous Motor Based on Recursive Least Squares[J]. Torpedo Technology, 2014, 22(6): 452-456.
[4]	陈进. 机械设备振动监测与故障诊断[M]. 上海: 上海交通大学出版社, 1999.
[5]	徐海珠, 谢顺依, 崔维开, 等. 基于脉振高频电压注入的对转永磁同步电机无位置传感器控制[J]. 鱼雷技术, 2012, 20(4): 290-294. Xu Hai-zhu, Xie Shun-yi, Cui Wei-kai, et al. Non-position-sensor Control for Counter-rotating PMSM Based on High Frequency Voltage Pulse Injection[J]. Torpedo Tech- nology, 2012, 20(4): 290-294.
[6]	Wang Z, Yang J, Ye H, et al. A Review of Permanent Magnet Synchronous Motor fault Diagnosis[C]//IEEE Conference and Expo Transportation Electrification Asia-Pacific(ITEC Asia-Pacific). Beijing: IEEE, 2014.
[7]	Elbouchikhi E, Choqueuse V, Auger F, et al. Motor Current Signal Analysis Based on a Matched Subspace Detector[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(12): 3260-3270.
[8]	Immovilli F, Bianchini C, Cocconcelli M, et al. Bearing Fault Model for Induction Motor with Externally Induced Vibration[J]. IEEE Transactions on Industrial Electronics, 2013, 60(8): 3408-3418.
[9]	李垣江, 张周磊, 李梦含, 等. 采用深度学习的永磁同步电机匝间短路故障诊断方法[J]. 电机与控制学报, 2020, 24(9): 173-180. Li Yuan-jiang, Zhang Zhou-lei, Li Meng-han, et al. Fault Diagnosis of Inter-turn Short Circuit of Permanent Magnet Synchronous Motor Based on Deep Learning[J]. Electric Machines and Control, 2020, 24(9): 173-180.
[10]	罗烨锋. 双绕组永磁同步电机伺服控制技术研究[D]. 杭州: 浙江大学, 2019.
[11]	Kim K, Lim S, Koo D, et al. The Shape Design of Permanent Magnet for Permanent Magnet Synchronous Motor Considering Partial Demagnetization[J]. IEEE Transactions on Magnetics, 2006, 42(10): 3485-3487.
[12]	梁星星, 冯旸赫, 马扬, 等. 多Agent深度强化学习综述[J]. 自动化学报, 2020, 46(12): 2537-2557. Liang Xing-xing, Feng Yang-he, Ma Yang, et al. Deep Multi-Agent Reinforcement Learning: a Survey[J]. Acta Automatica Sinca, 2020, 46(12): 2537-2557.
[13]	Ding Y, Ma L, Ma J, et al. Intelligent Fault Diagnosis for Ro-tating Machinery Using Deep Q-network Based Health State Classification: A Deep Reinforcement Learning Approach[J]. Advanced Engineering Informatics, 2019, 42: 100977.
[14]	Wang Q, Zhao X, Huang J, et al. Addressing Complexities of Machine Learning in Big Data: Principles, Trends and Challenges from Systematical Perspectives[EB/OL]. [2017-10-17]. https://www.preprints.org/manuscript/2017 10.0076/v1.
[15]	Mnih V, Kavukcuoglu K, Silver D, et al. Human-level Control through Deep Reinforcement Learning[J]. Nature, 2015, 518: 529-533.
[16]	赵冬斌, 邵坤, 朱圆恒, 等. 深度强化学习综述: 兼论计算机围棋的发展[J]. 控制理论与应用, 2016, 33(6): 701-717. Zhao Dong-bin, Shao Kun, Zhu Yuan-heng, et al. Review of Deep Reinforcement Learning and Discussions on the Development of Computer Go[J]. Control Theory & Applications, 2016, 33(6): 701-717.
[17]	Silver D, Schrittwieser J, Simonyan K, et al. Mastering the Game of Go Without Human Knowledge[J]. Nature, 2017, 550: 354-359.
[18]	Graves A, Wayne G, Reynolds M, et al. Hybrid Computing Using A Neural Network with Dynamic External Memory[J]. Nature, 2016, 538: 471-476. [19] Wang Z, Schaul T, Hessel M, et al. Dueling Network Architectures for Deep Reinforcement Learning[C]// Pro-ceedings of The 33rd International Conference on Machine Learning. New York, JMLR: W&CP, 2016.
[19]	[20] Sutton R, Barto A. Introduction to Reinforcement Learn-ing[J]. Machine Learning, 2005, 16(1): 285-286.
[20]	[21] Peuget R, Courtine S, Rognon J. Fault Detection and Iso-lation on A PWM Inverter by Knowledge-based Model[J]. IEEE Transactions on Industry Applications, 1998, 34(6): 1318-1326.
[21]	曹卫东, 施卫东. 退磁曲线法和磁导法的磁力泵耦合器转矩分析[J]. 农业机械学报, 2004, 32(2): 145-149. Cao Wei-dong, Shi Wei-dong. Analysis on Torque of Magnetic-force Pump Coupler Based on Demagnetization Curve and Magnet Conductivity Methodology[J]. Transactions of the Chinese Society for Agricultural Machinery, 2004, 32(2): 145-149.
[22]	伍召莉. 永磁同步电机失磁分析及在线监测[D]. 北京: 北京交通大学, 2012.
[23]	[24] Guo H, Xu J. Fault Tolerant Control with Torque Limitation Based on Fault Mode for Tenphase Permanent Magnet Synchronous Motor[J]. Chinese Journal of Aeronautics, 2015, 28(5): 1464-1475.
[24]	史丽萍, 汤家升, 王攀攀, 等. 采用最优小波树和改进BP神经网络的感应电动机定子故障诊断[J]. 电工技术学报, 2015, 30(24): 38-45. Shi Li-ping, Tang Jia-sheng, Wang Pan-pan, et al. Stator Fault Diagnosis of Induction Motors Using the Optimal Wavelet Tree and Improved BP Neural Network[J]. Transactions of China Electrotechnical Society, 2015, 30(24): 38-45.
[25]	[26] Gritli Y, Rossi C, Casadei D, et al. Demagnetizations Di-agnosis for Permanent Magnet Synchronous Motors Based on Advanced Wavelet Analysis[C]//2012 International Conference on Electrical Machines. Marseille: IEEE, 2012: 2397-2403.
[26]	孙文珺, 邵思羽, 严如强. 基于稀疏自动编码深度神经网络的感应电动机故障诊断[J]. 机械工程学报, 2016, 52(9): 65-71. Sun Wen-jun, Shao Si-yu, Yan Ru-qiang. Induction Motor Fault Diagnosis Based on Deep Neural Network of Sparse Auto-encoder[J]. Journal of Mechanical Engineering, 2016, 52(9): 65-71.
[27]	[28] Wang L, Xie Y, Zhang Y, et al. A Fault Diagnosis Method for Asynchronous Motor using Deep Learning[J]．Journal of Xi’an Jiaotong University, 2017, 51(10): 128-134．