基于OC-SVM与DNN相结合的ZPW-2000R轨道电路故障诊断研究

doi:10.3969/j.issn.1007-7375.2023.04.018

工业工程 ›› 2023, Vol. 26 ›› Issue (4): 154-163.doi: 10.3969/j.issn.1007-7375.2023.04.018

• 系统建模与优化算法 • 上一篇

基于OC-SVM与DNN相结合的ZPW-2000R轨道电路故障诊断研究

谢本凯¹, 蔡水涌¹, 黄春雷², 禹建丽¹, 陈广智³, 王国保¹

1. 郑州航空工业管理学院管理工程学院，河南郑州 450046;
2. 黑龙江瑞兴科技股份有限公司客户技术服务中心，黑龙江哈尔滨 150030;
3. 郑州航空工业管理学院智能工程学院，河南郑州 450046

收稿日期:2023-02-05 发布日期:2023-09-08
作者简介:谢本凯(1987-),男,河南省人,讲师,博士,研究方向为复杂系统建模与仿真
基金资助:
河南省软科学研究计划资助项目 (212400410099)；河南省高等教育教学改革研究与实践项目 (2021SJGLX470)；河南省自然科学基金资助项目 (222300410367)；河南省高等学校重点科研资助项目 (22B520041)

Fault Diagnosis of ZPW-2000R Track Circuits Based on OC-SVM and DNN

XIE Benkai¹, CAI Shuiyong¹, HUANG Chunlei², YU Jianli¹, CHEN Guangzhi³, WANG Guobao¹

1. School of Management Engineering, Zhengzhou University of Aeronautics, Zhengzhou 450046, China;
2. Customer Technical Service Center, Heilongjiang Ruixing Technology Co., Ltd., Harbin 150030, China;
3. School of Intelligent Engineering, Zhengzhou University of Aeronautics, Zhengzhou 450046, China

Received:2023-02-05 Published:2023-09-08

摘要/Abstract

摘要： 针对轨道电路故障诊断准确率低且高质量故障数据难以收集等问题，提出一种基于单分类支持向量机 (OC-SVM) 与深度神经网络 (DNN) 相结合的故障诊断方法。该方法使用OC-SVM模型对数据进行单分类识别，将正样本数据输入到DNN模型进行训练和预测，为负样本数据添加标签并收集。利用ZPW-2000R轨道电路信号数据进行大量实验，结果表明OC-SVM模型能精确地识别出正负样本数据，DNN模型能准确高效地诊断出15种数据类型，且准确率高达99%。与粒子群算法优化支持向量机、卷积神经网络、堆叠自编码器3种故障诊断方法相比，该组合方法的准确率更高，诊断效果更稳定。

关键词: 故障诊断, 深度神经网络, 单分类支持向量机, ZPW-2000R轨道电路

Abstract: To address the problems of low accuracy for fault diagnosis for track circuits and difficulty in collecting high-quality fault data, a fault diagnosis method based on the combination of one-class support vector machine (OC-SVM) and deep neural network (DNN) is proposed. This method uses OC-SVM model to perform single-class recognition of data. The positive sample data are input into DNN model for training and prediction to label and collect the negative sample data. A large number of experiments are conducted using ZPW-2000R track circuit signal data. Results show that the OC-SVM model can accurately identify the positive and negative sample data, and DNN model can accurately and efficiently diagnose 15 types of data with an accuracy rate of 99%. Compared with the three fault diagnosis methods of particle swarm algorithm optimized support vector machine, convolutional neural network and stacked self-encoder, this combined method has a higher accuracy and more stable diagnosis effectiveness.

Key words: fault diagnosis, deep neural network, one-class support vector machine, ZPW-2000R track circuit

中图分类号:

U284.92

谢本凯, 蔡水涌, 黄春雷, 禹建丽, 陈广智, 王国保. 基于OC-SVM与DNN相结合的ZPW-2000R轨道电路故障诊断研究[J]. 工业工程, 2023, 26(4): 154-163.

XIE Benkai, CAI Shuiyong, HUANG Chunlei, YU Jianli, CHEN Guangzhi, WANG Guobao. Fault Diagnosis of ZPW-2000R Track Circuits Based on OC-SVM and DNN[J]. Industrial Engineering Journal, 2023, 26(4): 154-163.

参考文献

[1] 董昱, 陈星. 基于粗糙集和模糊认知图的ZPW-2000轨道电路故障诊断[J]. 铁道学报, 2018, 40(6): 83-9
DONG Yu, CHEN Xing. Fault diagnosis of ZPW-2000 rail circuit based on rough set and fuzzy cognitive graph[J]. Journal of Railways, 2018, 40(6): 83-9
[2] 李长春, 赵林海. 基于LightGBM算法的25Hz相敏轨道电路故障诊断方法优化[J]. 铁道学报, 2022, 44(8): 68-77
LI Changchun, ZHAO Linhai. Optimization of 25Hz phase-sensitive rail circuit fault diagnosis method based on LightGBM algorithm[J]. Journal of Railway, 2022, 44(8): 68-77
[3] SUN S, ZHAO H. Fault diagnosis in railway track circuits using support vector machines[C]//2013 12th International Conference on Machine Learning and Applications. IEEE, 2013, 2: 345-350.
[4] MA Q, HE Y, ZHOU F. A new decision tree approach of support vector machine for analog circuit fault diagnosis[J]. Analog Integrated Circuits and Signal Processing, 2016, 88(3): 455-463
[5] YUAN X, LIU Z, MIAO Z, et al. Fault diagnosis of analog circuits based on IH-PSO optimized support vector machine[J]. IEEE Access, 2019, 7: 137945-137958
[6] 林俊亭, 王帅. 基于DBN-MPA-LSSVM的无绝缘轨道电路故障诊断研究[J]. 电子测量与仪器学报, 2022, 36(9): 37-44
LIN Junting, WANG Shuai. Research on fault diagnosis of uninsulated rail circuits based on DBN-MPA-LSSVM[J]. Journal of Electronic Measurement and Instrumentation, 2022, 36(9): 37-44
[7] 卢皎, 禹建丽, 黄春雷, 等. 基于卷积神经网络的ZPW-2000R轨道电路智能故障诊断方法[J]. 工业工程, 2021, 24(4): 127-33
LU Jiao, YU Jianli, HUANG Chunlei, et al. Intelligent fault diagnosis method of ZPW-2000R rail circuit based on convolutional neural network[J]. Industrial Engineering Journal, 2021, 24(4): 127-33
[8] 田粉霞, 杨世武, 崔勇, 等. 基于改进卷积神经网络的无绝缘轨道电路调谐区故障诊断[J]. 铁路计算机应用, 2020, 29(6): 58-63+74
TIAN Fenxia, YANG Shiwu, CUI Yong, et al. Improved convolutional neural network-based fault diagnosis in tuned areas of uninsulated rail circuits[J]. Railroad Computer Applications, 2020, 29(6): 58-63+74
[9] 朱文博, 王小敏. 基于组合决策树的无绝缘轨道电路故障诊断方法研究[J]. 铁道学报, 2018, 40(7): 74-9
ZHU Wenbo, WANG Xiaomin. Research on fault diagnosis method of uninsulated rail circuit based on combined decision tree[J]. Journal of Railways, 2018, 40(7): 74-9
[10] 宋威, 林建维, 周方泽, 等. 基于改进降噪自编码器的风机轴承故障诊断方法[J]. 电力系统保护与控制, 2022, 50(10): 61-68
SONG Wei, LIN Jianwei, ZHOU Fangze, et al. Fault diagnosis method for wind turbine bearings based on improved noise-reducing self-encoder[J]. Power System Protection and Control, 2022, 50(10): 61-68
[11] 张洪亮, 余其源, 王锐. 跳连接变分自编码器与CNN相结合的滚动轴承故障诊断方法[J/OL]. 机械科学与技术: 1-8 (2022-10-14)[2023-3-03-27]. https://doi.org/10.13433/j.cnki.1003-8728.20220273.
ZHANG Hongliang, YU Qiyuan, WANG Rui. A rolling bearing fault diagnosis method combining jump-connected variational self-encoder and CNN [J/OL]. Mechanical Science and Technology: 1-8 (2022-10-14) [2023-03-27]. https://doi.org/10.13433/j.cnki.1003-8728.20220273.
[12] 王秋实, 王小敏. 基于FTA与改进神经网络的轨道电路红光带诊断方法[J]. 铁道标准设计, 2017, 61(4): 147-153
WANG Qiushi, WANG Xiaomin. A red light band diagnosis method for rail circuit based on FTA and improved neural network[J]. Railway Standard Design, 2017, 61(4): 147-153
[13] STEINWART I, CHRISTMANN A. Support vector machines [M]. Berlin: Springer Science & Business Media, 2008.
[14] 彭彬森, 夏虹, 王志超, 等. 深度神经网络在滚动轴承故障诊断中的应用[J]. 哈尔滨工业大学学报, 2021, 53(6): 155-162
PENG Binsen, XIA Hong, WANG Zhichao, et al. Application of deep neural network in rolling bearing fault diagnosis[J]. Journal of Harbin Institute of Technology, 2021, 53(6): 155-162
[15] ZHENG Z, DAI S, XIE X. Research on fault detection for ZPW-2000A jointless track circuit based on deep belief network optimized by improved particle swarm optimization algorithm[J]. IEEE Access, 2020, 8: 175981-175997
[16] KHAN M A, ABU-KHADRAH A, SIDDIQUI S Y, et al. Support-Vector-Machine-based Adaptive Scheduling in Mode 4 Communication[J]. Computers, Materials & Continua, 2022, 73(2): 3319-3331
[17] ROY A, CHAKRABORTY S. Support Vector Machine in Structural Reliability Analysis: A Review[J]. Reliability Engineering & System Safety, 2023, 233: 109119-109126
[18] SHIM K, DO T N, NGUYEN T-V, et al. Enhancing PHY-Security of FD-Enabled NOMA Systems Using Jamming and User Selection: Performance Analysis and DNN Evaluation[J]. IEEE Internet of Things Journal, 2021, 8(24): 17476-17494
[19] 孟庆成, 万达, 吴浩杰, 等. 基于卷积神经网络的混凝土裂缝图像识别方法[J]. 沈阳建筑大学学报 (自然科学版), 2021, 37(5): 832-840
MENG Qingcheng, WAN Da, WU Haojie, et al. Convolutional neural network based concrete crack image recognition method[J]. Journal of Shenyang University of Architecture (Natural Science Edition), 2021, 37(5): 832-840

基于OC-SVM与DNN相结合的ZPW-2000R轨道电路故障诊断研究

Fault Diagnosis of ZPW-2000R Track Circuits Based on OC-SVM and DNN

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 8

编辑推荐

Metrics

本文评价

[1]	陈光林, 于丽娅, 张成龙, 周鹏, 李笑瑜. 基于麻雀算法优化宽度学习系统的轴承故障诊断[J]. 工业工程, 2023, 26(3): 151-158.
[2]	杨静宗, 施春朝, 杨天晴, 吴丽玫. 基于LMD和ABC优化KELM的故障诊断方法[J]. 工业工程, 2022, 25(3): 124-131.
[3]	叶祎旎, 李艳婷. 基于CNN-集成学习的多风电机组故障诊断[J]. 工业工程, 2022, 25(1): 136-143.
[4]	张炎亮, 颜健勇. 基于G-DPSO算法的决策树轴承故障诊断方法[J]. 工业工程, 2021, 24(6): 41-47.
[5]	卢皎, 禹建丽, 黄春雷, 陈洪根. 基于卷积神经网络的ZPW-2000R轨道电路智能故障诊断方法[J]. 工业工程, 2021, 24(4): 127-133.
[6]	王宇凡，张淑娟, 梁工谦. 非线性复合FSVM在设备故障诊断中的应用[J]. 工业工程, 2014, 17(4): 139-144.
[7]	朱清香1，焦朋沙1，刘晶2，郝红红1. 矩阵加权关联规则在故障诊断系统中的应用[J]. 工业工程, 2013, 16(2): 87-91.
[8]	刘晶1,2, 季海鹏3. 基于神经网络的关联规则在故障诊断中的应用[J]. 工业工程, 2011, 14(2): 118-121.