李小平,孙龙,王建功,曾佳焘

• 1:兰州交通大学自动化与电气工程学院
• 2:兰州交通大学机电工程学院
• 3:中国铁路兰州局集团有限公司

摘要(Abstract)：

转辙机健康状态评估对保证轨道交通运营安全至关重要。针对ZYJ7型电液转辙机运行过程中可能出现的退化与故障问题，提出一种融合多源信号与BiGRU的电液转辙机健康状态评估方法。首先采集转辙机电气、液压、机械、环境等子系统的13类原始信号，在此基础上提取38维多域物理特征；其次构建一种基于核主成分分析(KPCA)与马氏距离(MD)的健康指数HI计算方法，通过核主成分分析(KPCA)实现38维多域非线性物理特征降维，利用马氏距离计算样本与健康状态之间的偏离度；最后设计一种融合注意力机制的双向门控循环单元(BiGRU-Attention)双任务评估模型，完成对HI回归预测与健康状态分类的双任务输出。实验结果表明，基于多源信号与BiGRU的电液转辙机健康状态评估方法的性能均优于CNN-GRU、TCN和Transformer等主流深度学习模型，评估准确率达到95.8%,比CNN-GRU、TCN和Transformer等方法分别提升了1.3%、1.7%和3.8%。该方法为转辙机健康状态评估提供了一种有效的技术框架，助力转辙机维修从传统定期维修向智能运维的转型升级。

关键词(KeyWords)： 电液转辙机;健康状态评估;多源信息融合;健康指数;BiGRU-Attention

Abstract:

Keywords:

基金项目(Foundation): 2025年甘肃省联合科研基金重点项目(25JRRA1072)

作者(Author): 李小平,孙龙,王建功,曾佳焘

参考文献(References)：

• [1] Lao Z P,He D Q,Wei Z X,et al.Intelligent fault diagnosis for rail transit switch machine based on adaptive feature selection and improved LightGBM [J].Engineering Failure Analysis,2023,148:107219.
• [2] Wang Z S,Sun D,Zhou J,et al.Health monitoring analysis of an urban rail transit switch machine [J].Sustainability,2024,16(9):3527.
• [3] 尹航，梁玉琦，王成龙.基于深度门控循环单元网络的转辙机健康状态评估 [J].铁道学报，2021,43(11):88-96.Yin Hang,Liang Yuqi,Wang Chenglong.Health condition assessment of point machine based on a deep GRU model [J].Journal of the China Railway Society,2021,43(11):88-96.
• [4] Wu X C,Wen X.Research on health stage division of switch machine based on bray-Curtis distance and fisher optimal segmentation method [J].Chinese Journal of Electronics,2023,32(5):955-962.
• [5] 米根锁，窦媛媛.基于CEEMDAN与改进核极限学习机的S700K转辙机健康状态诊断 [J].电子测量与仪器学报，2023,37(6):232-239.Mi Gensuo,Dou Yuanyuan.Health state diagnosis of S700K switch machine based on CEEMDAN and improved kernel based extreme learning machine [J].Journal of Electronic Measurement and Instrumentation,2023,37(6):232-239.
• [6] 刘文涛，黄子辉，潘庆球，等.基于人工智能的道岔健康度评估方法与评估系统建设路径研究 [J].城市轨道交通研究，2020,23(10):149-153.Liu Wentao,Huang Zihui,Pan Qingqiu,et al.Evaluation method of turnout health and construction path of evaluation system based on artificial intelligence [J].Urban Mass Transit,2020,23(10):149-153.
• [7] 张振海，李哲远，鲍盛成，等.基于数字孪生的转辙机健康状态智能分析方法 [J].铁道科学与工程学报，2025,22(10):4697-4711.Zhang Zhenhai,Li Zheyuan,Bao Shengcheng,et al.Intelligent analysis method for the health status of switch machines based on digital twin technology [J].Journal of Railway Science and Engineering,2025,22(10):4697-4711.
• [8] Yang J,Sun Y K,Cao Y,et al.Predictive maintenance for switch machine based on digital twins [J].Information,2021,12(11):485.
• [9] 张轩赫，梁志国，张宏扬，等.基于联锁运行数据的转辙机健康状态智能分析方法 [J].北京交通大学学报，2023,47(2):137-146.Zhang Xuanhe,Liang Zhiguo,Zhang Hongyang,et al.An intelligent analysis method for health status of point machine based on interlocking operation data [J].Journal of Beijing Jiaotong University,2023,47(2):137-146.
• [10] Chen L,Zhang Y,Zheng Y,et al.Remaining useful life prediction of lithium-ion battery with optimal input sequence selection and error compensation [J].Neurocomputing,2020,414:245-254.
• [11] 罗家祥，罗丹，胡跃明.带权重变化和决策融合的ELM在线故障检测 [J].控制与决策，2018,33(6):1033-1040.Luo Jiaxiang,Luo Dan,Hu Yueming.A new online extreme learning machine with varying weights and decision level fusion for fault detection [J].Control and Decision,2018,33(6):1033-1040.
• [12] Kumar A,Gandhi C P,Zhou Y Q,et al.Improved CNN for the diagnosis of engine defects of 2-wheeler vehicle using wavelet synchro-squeezed transform (WSST) [J].Knowledge-Based Systems,2020,208:106453.
• [13] Xiang S,Qin Y,Luo J,et al.Multicellular LSTM-based deep learning model for aero-engine remaining useful life prediction [J].Reliability Engineering & System Safety,2021,216:107927.
• [14] Zhang Y,Chen L,Li Y,et al.A hybrid approach for remaining useful life prediction of lithium-ion battery with Adaptive Levy Flight optimized Particle Filter and Long Short-Term Memory network [J].Journal of Energy Storage,2021,44:103245.
• [15] Xiao L,Liu Z X,Zhang Y,et al.Degradation assessment of bearings with trend-reconstruct-based features selection and gated recurrent unit network [J].Measurement,2020,165:108064.
• [16] Ma G J,Xu S P,Cheng C.Fault detection of lithium-ion battery packs with a graph-based method [J].Journal of Energy Storage,2021,43:103209.
• [17] Peng W W,Ye Z S,Chen N.Bayesian deep-learning-based health prognostics toward prognostics uncertainty [J].IEEE Transactions on Industrial Electronics,2020,67(3):2283-2293.
• [18] Mao W T,Chen J X,Chen Y J,et al.Construction of health indicators for rotating machinery using deep transfer learning with multiscale feature representation [J].IEEE Transactions on Instrumentation and Measurement,2021,70:1-13.
• [19] Wang F,Chen G X,Liu K Y,et al.Health assessment of an electro-hydraulic servo pump control system for servomotor based on LGA deep neural network [J].Heliyon,2024,10(5):e26936.
• [20] Zheng Y S,Chen W M,Zhang Y N,et al.Prediction of the remaining useful life of a switch machine,based on multi-source data [J].Sustainability,2022,14(21):14517.
• [21] 单增海，李志远，张旭，等.基于多传感器信息融合和多粒度级联森林模型的液压泵健康状态评估 [J].中国机械工程，2021,32(19):2374-2382.Shan Zenghai,Li Zhiyuan,Zhang Xu,et al.Health status assessment of hydraulic pumps based on multi-sensor information fusion and multi-grained cascade forest model [J].China Mechanical Engineering,2021,32(19):2374-2382.
• [22] 黄晨涛.一种转辙机运行参数综合监测系统 [J].铁路通信信号工程技术，2023,20(12):37-42.Huang Chentao.Comprehensive monitoring system for operating parameters of switch machines [J].Railway Signalling & Communication Engineering,2023,20(12):37-42.
• [23] Meng Q Z,Wen W G,Bai Y H,et al.A fault detection method for electrohydraulic switch machine based on oil-pressure-signal-sectionalized feature extraction [J].Entropy,2022,24(7):848.
• [24] Chen C,Xu Z W,Mei M,et al.Fault diagnosis scheme for railway switch machine using multi-sensor fusion tensor machine [J].Computers,Materials & Continua,2024,79(3):4533-4549.
• [25] Yan R Q,Gao R X,Chen X F.Wavelets for fault diagnosis of rotary machines:a review with applications [J].Signal Processing,2014,96:1-15.
• [26] Liu R N,Yang B Y,Zio E,et al.Artificial intelligence for fault diagnosis of rotating machinery:a review [J].Mechanical Systems and Signal Processing,2018,108:33-47.
• [27] Zhao R,Yan R Q,Chen Z H,et al.Deep learning and its applications to machine health monitoring [J].Mechanical Systems and Signal Processing,2019,115:213-237.
• [28] 上官伟，师泽斌，彭聪，等.基于加权马氏距离的BUCK电路健康度评估与预测[J].铁道学报，2023,45(12):103-111.Shangguan Wei,Shi Zebin,Peng Cong,et al.Evaluation and prediction of health of BUCK circuit based on weighted mahalanobis distance[J].Journal of the China Railway Society,2023,45(12):103-111.
• [29] 欧旭鹏，任涛，王玉鹏，等.基于改进麻雀搜索算法优化深度学习网络超参数的短期风电功率预测[J].智慧电力，2023,51(3):31-38,52.Ou Xupeng,Ren Tao,Wang Yupeng,et al.Short-term wind power prediction by optimizing deep learning network hyper-parameters based on ISSA[J].Smart Power,2023,51(3):31-38,52.
• [30] Jafari S,Byun Y C.A CNN-GRU approach to the accurate prediction of batteries’ remaining useful life from charging profiles [J].Computers,2023,12(11):219.
• [31] Liu M P,Sun X H,Wang Q N,et al.Short-term load forecasting using EMD with feature selection and TCN-based deep learning model [J].Energies,2022,15(19):7170.
• [32] Arkin E,Yadikar N,Xu X B,et al.A survey:object detection methods from CNN to transformer [J].Multimedia Tools and Applications,2023,82(14):21353-21383.