高建敏,裴治民,陈耀威,朱胜阳

• 1:西南交通大学轨道交通运载系统全国重点实验室

摘要(Abstract)：

轨道结构服役状态在高速、重载列车安全、平稳运营方面起着至关重要的作用。随着传感器技术和计算机技术的提升，以及机器学习、深度学习理论的发展，将人工智能新方法运用到轨道结构损伤与劣化病害智能诊断识别领域成为新的研究热点和趋势。为此，在铁路有砟轨道结构病害形式及其检测技术总结分析基础上，综述应用车辆振动信号的有砟轨道结构劣化病害识别研究现状，包括应用传统信号时频分析方法的研究和应用机器学习(深度学习)方法的研究，总结不同方法的数据输入形式、识别功能及其准确率、验证方式等。结合当前技术未来发展趋势，给出基于深度学习理论方法的有砟轨道结构损伤与劣化病害智能诊断识别未来的发展方向和进一步研究建议，主要包括多源异构数据深入融合与挖掘分析、无监督或自监督学习模式应用、小样本或零样本迁移学习与模型泛化能力研究以及模型轻量化与复杂环境适应性研究。相关成果可为铁路有砟轨道结构状态车载智能化检监测技术研究提供参考。

关键词(KeyWords)： 有砟轨道;结构病害;识别方法;时频分析;深度学习;数据融合分析;迁移学习

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金(52372435,52388102);; 2026年轨道交通运载系统全国重点实验室自主课题(2026RVL-T05)

作者(Author): 高建敏,裴治民,陈耀威,朱胜阳

参考文献(References)：

• [1] 国家铁路局.“十四五”铁路科技创新规划[J].铁道技术监督，2022,50(1):9-15,20.National Railway Administration of the People’s Republic of China.Railway science and technology innovation plan in the 14th five-year plan[J].Railway Quality Control,2022,50(1):9-15,20.
• [2] 王照伟，刘传帅，赵文祥，等.多尺度多任务注意力卷积神经网络滚动轴承故障诊断方法 [J].电机与控制学报，2024,28(7):65-76.Wang Zhaowei,Liu Chuanshuai,Zhao Wenxiang,et al.Rolling bearing fault diagnosis with multi-scale multi-task attention convolutional neural network [J].Electric Machines and Control,2024,28(7):65-76.
• [3] Liu G Z,Wu L F.Incremental bearing fault diagnosis method under imbalanced sample conditions [J].Computers & Industrial Engineering,2024,192:110203.
• [4] Wang Z Y,Luo Q S,Chen H,et al.A high-accuracy intelligent fault diagnosis method for aero-engine bearings with limited samples [J].Computers in Industry,2024,159/160:104099.
• [5] 张庆铼.基于卷积神经网络的浮置板轨道钢弹簧损伤检测方法研究 [D].成都：西南交通大学，2021.Zhang Qinglai.Research on damage detection method of floating slab track steel spring based on convolutional neural network [D].Chengdu:Southwest Jiaotong University,2021.
• [6] Oregui M.Vertical railway track dynamics:from measurements to numerical modeling[D].Delft:TU Delft,2015.
• [7] 李维刚，梅洋，樊响，等.基于车载激光点云的铁路轨道检测 [J].中国激光，2022,49(4):162-173.Li Weigang,Mei Yang,Fan Xiang,et al.Railway track detection based on vehicle laser point cloud [J].Chinese Journal of Lasers,2022,49(4):162-173.
• [8] Kumar A,Harsha S P.A systematic literature review of defect detection in railways using machine vision-based inspection methods [J].International Journal of Transportation Science and Technology,2025,18:207-226.
• [9] 王石磊，赵志刚，刘宏江，等.铁路碎石道床雷达检测技术研究与应用进展 [J].铁道学报，2023,45(2):116-126.Wang Shilei,Zhao Zhigang,Liu Hongjiang,et al.Research and application progress of radar detection technology for railway ballast bed [J].Journal of the China Railway Society,2023,45(2):116-126.
• [10] 井国庆，卜俊杰，彭湛.基于探地雷达单道信号的碎石道床病害智能识别 [J].铁道学报，2025,47(6):170-178.Jing Guoqing,Bu Junjie,Peng Zhan.Intelligent recognition of ballast track bed defects based on A-scan GPR data [J].Journal of the China Railway Society,2025,47(6):170-178.
• [11] Bombarda D,Vitetta G M,Ferrante G.Rail diagnostics based on ultrasonic guided waves:an overview [J].Applied Sciences,2021,11(3):1071.
• [12] 胡文博，邱实，许馨月，等.基于深度学习的钢轨伤损超声检测与分类 [J].铁道学报，2021,43(4):108-116.Hu Wenbo,Qiu Shi,Xu Xinyue,et al.Ultrasonic detection and classification for internal defect of rail based on deep learning [J].Journal of the China Railway Society,2021,43(4):108-116.
• [13] 牛留斌，刘金朝.基于车辆动态响应分析的轨道状态检测技术应用现状及展望 [J].中国铁道科学，2021,42(6):130-142.Niu Liubin,Liu Jinzhao.Application status and prospect of track state detection technology based on vehicle dynamic response analysis [J].China Railway Science,2021,42(6):130-142.
• [14] 缪炳荣，刘俊利，张盈，等.轨道车辆结构振动损伤识别技术综述 [J].交通运输工程学报，2021,21(1):338-357.Miao Bingrong,Liu Junli,Zhang Ying,et al.Review on structural vibration damage identification technology for railway vehicles [J].Journal of Traffic and Transportation Engineering,2021,21(1):338-357.
• [15] De Donato L,Flammini F,Marrone S,et al.A survey on audio-video based defect detection through deep learning in railway maintenance [J].IEEE Access,2022,10:65376-65400.
• [16] Molodova M,Li Z L,Dollevoet R.Axle box acceleration:measurement and simulation for detection of short track defects [J].Wear,2011,271(1/2):349-356.
• [17] Oregui M,Li Z L,Dollevoet R.Identification of characteristic frequencies of damaged railway tracks using field hammer test measurements [J].Mechanical Systems and Signal Processing,2015,54/55:224-242.
• [18] Cantero D,Basu B.Railway infrastructure damage detection using wavelet transformed acceleration response of traversing vehicle:railway infrastructure damage detection using vehicle acceleration [J].Structural Control and Health Monitoring,2015,22(1):62-70.
• [19] Chudzikiewicz A,Bogacz R,Kostrzewski M,et al.Condition monitoring of railway track systems by using acceleration signals on wheelset axle-boxes [J].Transport,2018,33(2):555-566.
• [20] Cho H,Park J.Study of rail squat characteristics through analysis of train axle box acceleration frequency [J].Applied Sciences,2021,11(15):7022.
• [21] Molodova M,Li Z L,Núňez A,et al.Monitoring the railway infrastructure:detection of surface defects using wavelets [C]//16th International IEEE Conference on Intelligent Transportation Systems (ITSC 2013).Piscataway:IEEE,2013:1316-1321.
• [22] Wei X K,Liu Y X,Yin X X.Detection of rail squats based on Hilbert-Huang transform by using bogie acceleration measurement [C]//2016 Chinese Control and Decision Conference (CCDC).Piscataway:IEEE,2016:2983-2988.
• [23] 张愿明.基于振动响应特征的连续多点钢轨压溃识别研究 [D].石家庄：石家庄铁道大学，2022.Zhang Yuanming.Research on continuous multipoint rail crushing identification based on vibration response characteristics [D].Shijiazhuang:Shijiazhuang Tiedao University,2022.
• [24] Liu C,Zhang L H,Ren T Q,et al.Detection of rail fastener conditions using time-frequency entropy based on orthogonal empirical mode decomposition [J].Applied Mechanics and Materials,2013,333/334/335:1708-1712.
• [25] Wang L Q,Zhang Y,Lie S T.Detection of damaged supports under railway track based on frequency shift [J].Journal of Sound and Vibration,2017,392:142-153.
• [26] 周文静，刘冲，韦佳宏，等.基于小波包能量谱的钢轨扣件松脱检测研究 [J].传感器与微系统，2017,36(5):40-43.Zhou Wenjing,Liu Chong,Wei Jiahong,et al.Study on rail fastener loosening identification based on wavelet packet energy spectrum [J].Transducer and Microsystem Technologies,2017,36(5):40-43.
• [27] 张丽华.钢轨扣件松脱检测信号的HHT分析方法 [D].大连：大连理工大学，2014.Zhang Lihua.HHT analysis application in rail fastener loosening detection [D].Dalian:Dalian University of Technology,2014.
• [28] Yang C,Kaynardag K,Salamone S.Missing rail fastener detection based on laser Doppler vibrometer measurements [J].Journal of Nondestructive Evaluation,2023,42(3):1-18.
• [29] Wang S H,Tang L H,Dou Y L,et al.Enhancement of track damage identification by data fusion of vibration-based image representation [J].Journal of Nondestructive Evaluation,2023,43(1):1-13.
• [30] 侯博文，李佳静，高亮，等.基于脉冲激励下钢轨振动响应的扣件失效识别方法 [J].工程力学，2021,38(2):122-133.Hou Bowen,Li Jiajing,Gao Liang,et al.Failure identification method of fastening system based on rail response under pulse excitation [J].Engineering Mechanics,2021,38(2):122-133.
• [31] Rui L,Zappa E,Collina A.Vision-based measurement of crack generation and evolution during static testing of concrete sleepers [J].Engineering Fracture Mechanics,2020,224:106715.
• [32] Aela P,Cai J F,Jing G Q,et al.Vision-based monitoring of railway superstructure:a review [J].Construction and Building Materials,2024,442:137385.
• [33] Clark A,Kaewunruen S,Janeliukstis R,et al.Damage detection in railway prestressed concrete sleepers using acoustic emission [J].IOP Conference Series:Materials Science and Engineering,2017,251:012068.
• [34] 陶智亮.铁路路基车载GPR剖面翻浆冒泥识别技术 [D].成都：西南交通大学，2018.Tao Zhiliang.A technique of recognizing mud pumping in railway subgrade on train-mounted GPR profiles [D].Chengdu:Southwest Jiaotong University,2018.
• [35] 刘恒柏，江波，瞿起明，等.基于车载探地雷达技术的碎石道床检测数据的应用探讨 [J].铁道建筑，2021,61(8):134-137,158.Liu Hengbai,Jiang Bo,Qu Qiming,et al.Discussion of application of detection data of track bed based on vehicle-mounted ground penetrating radar technology [J].Railway Engineering,2021,61(8):134-137,158.
• [36] 王石磊，彭湛，李明，等.基于高频雷达的铁路碎石道床状态表征指标研究 [J].中国铁道科学，2022,43(3):8-16.Wang Shilei,Peng Zhan,Li Ming,et al.Research on state characterization index of railway ballast bed based on high frequency radar [J].China Railway Science,2022,43(3):8-16.
• [37] 杨爱红，王石磊，冯乾旭.运营铁路有砟道床车载式检测技术综述 [J].铁道建筑，2022,62(7):47-51.Yang Aihong,Wang Shilei,Feng Qianxu.Review on vehicle mounted detection technology of railway ballasted bed in operation [J].Railway Engineering,2022,62(7):47-51.
• [38] Teng J,Long X J,Yang Q,et al.Semi-Conv-DETR:a railway ballast bed defect detection model integrating convolutional augmentation and semi-supervised DETR [J].Transportation Geotechnics,2024,48:101334.
• [39] 邵志慧.基于振动响应的轨枕服役状态诊断方法研究 [D].上海：上海工程技术大学，2021.Shao Zhihui.Research on diagnosis method of sleepers service state based on vibaration response [D].Shanghai:Shanghai University of Engineering Science,2021.
• [40] Kang J,Ullah Z,Gwak J.MRI-based brain tumor classification using ensemble of deep features and machine learning classifiers [J].Sensors,2021,21(6):2222.
• [41] Manikandan S,Duraivelu K.Fault diagnosis of various rotating equipment using machine learning approaches-A review [J].Proceedings of the Institution of Mechanical Engineers,Part E:Journal of Process Mechanical Engineering,2021,235(2):629-642.
• [42] Mahadevkar S V,Khemani B,Patil S,et al.A review on machine learning styles in computer vision:techniques and future directions [J].IEEE Access,2022,10:107293-107329.
• [43] Ji A,Woo W L,Wong E W L,et al.Rail track condition monitoring:a review on deep learning approaches [J].Intelligence & Robotics,2021,1(2):151-75.
• [44] Yuan Z D,Zhu S Y,Chang C,et al.An unsupervised method based on convolutional variational auto-encoder and anomaly detection algorithms for light rail squat localization [J].Construction and Building Materials,2021,313:125563.
• [45] Phusakulkajorn W,Hendriks J,Li Z L,et al.Spiking neural network with time-varying weights for rail squat detection [J].Applied Soft Computing,2025,184:113689.
• [46] Xie Q L,Tao G Q,He B,et al.Rail corrugation detection using one-dimensional convolution neural network and data-driven method [J].Measurement,2022,200:111624.
• [47] Wang Y,Xiao H,Nadakatti M M,et al.A metro rail corrugation detection framework based on car body vibration signals and unsupervised learning [J].Engineering Applications of Artificial Intelligence,2025,153:110976.
• [48] Ouyang R,Luo J,Zhu S Y,et al.Surrogate model construction and improved detection of rail corrugation in high-speed railways based on a hybrid neural network [J].Engineering Structures,2025,342:120949.
• [49] Yuan Z D,Zhu S Y,Zhai W M.Damage detection of rail fastening system through deep learning and vehicle-track coupled dynamics[C]//Advances in Dynamics of Vehicles on Roads and Tracks.Cham:Springer International Publishing,2020:148-153.
• [50] Yuan Z D,Zhu S Y,Yuan X C,et al.Vibration-based damage detection of rail fastener clip using convolutional neural network:experiment and simulation [J].Engineering Failure Analysis,2021,119:104906.
• [51] Chen M,Zhai W M,Zhu S Y,et al.Vibration-based damage detection of rail fastener using fully convolutional networks [J].Vehicle System Dynamics,2022,60(7):2191-2210.
• [52] 吴送英，刘林芽，万强华，等.基于小波包能量谱与改进BP神经网络的铁路扣件松脱检测算法研究 [J].华东交通大学学报，2023,40(6):69-78.Wu Songying,Liu Linya,Wan Qianghua,et al.Research on detection algorithm of railway fastener looseness based on wavelet packet energy spectrum and improved BP neural network [J].Journal of East China Jiaotong University,2023,40(6):69-78.
• [53] Yin X X,Wei X K,Zheng H C.Railway track vibration analysis and intelligent recognition of fastener defects [J].Advanced Theory and Simulations,2022,5(10):2200027.
• [54] 罗锟，余崇彬，李泳辉，等.应用导纳法与注意力卷积神经网络相结合的扣件失效识别方法研究[J/OL].机械科学与技术，2025.https://doi.org/10.13433/j.cnki.1003-8728.20240173.Luo Kun,Yu Chongbin,Li Yonghui,et al.Research on a fastener failure indentification method combining the admittance approach and attention convolutional neural network[J/OL].Mechanical Science and Technology for Aerospace Engineering,2025.https://doi.org/10.13433/j.cnki.1003-8728.20240173.
• [55] Wang S H,Kim M,Li G,et al.Simulation-derived semantic features enabling zero-shot domain adaptation for rail fastener failure detection in unseen scenarios [J].Engineering Failure Analysis,2025,182:110186.
• [56] 华莉，杨俭，袁天辰，等.基于VMD和BP神经网络的轨道病害诊断方法 [J].电子科技，2022,35(4):40-46.Hua Li,Yang Jian,Yuan Tianchen,et al.Track disease diagnosis method based on VMD and BP neural network [J].Electronic Science and Technology,2022,35(4):40-46.
• [57] 华莉，孔尧尧，陈永逸，等.基于GA-BP神经网络的轨道结构病害诊断方法研究 [J].智能计算机与应用，2022,12(7):123-128.Hua Li,Kong Yaoyao,Chen Yongyi,et al.Research on track structure disease diagnosis method based on GA-BP neural network [J].Intelligent Computer and Applications,2022,12(7):123-128.
• [58] 丁仪.基于车致振动响应的轨下支承状态识别研究 [D].石家庄：石家庄铁道大学，2022.Ding Yi.Research on identification of support state under rail based on vehicle-induced track vibration response [D].Shijiazhuang:Shijiazhuang Tiedao University,2022.
• [59] Sysyn M,Przybylowicz M,Nabochenko O,et al.Identification of sleeper support conditions using mechanical model supported data-driven approach [J].Sensors,2021,21(11):3609.
• [60] Zhang D W,Xu P J,Tian Y Y,et al.Ballasted track behaviour induced by absent sleeper support and its detection based on a convolutional neural network using track data [J].Urban Rail Transit,2023,9(2):92-109.
• [61] Sresakoolchai J,Kaewunruen S.Prognostics of unsupported railway sleepers and their severity diagnostics using machine learning [J].Scientific Reports,2022,12:6064.
• [62] 崔艺泉.重载铁路道床板结车载检测算法研究 [D].北京：北京交通大学，2024.Cui Yiquan.Research on heavy haul railway ballast on-board detection algorithm [D].Beijing:Beijing Jiaotong University,2024.
• [63] 陈耀威.基于卷积块注意力机制神经网络的轨枕异常支撑状态检测方法研究[D].成都：西南交通大学，2025.Chen Yaowei.Detection method of abnormal support conditions of railway sleepers based on convolutional block attention module neural network[D].Chengdu:Southwest Jiaotong University,2025.
• [64] 段礼祥，李涛，唐瑜，等.基于多源异构信息融合的机械故障诊断方法[J].石油机械，2021,49(2):60-67,80.Duan Lixiang,Li Tao,Tang Yu,et al.Mechanical fault diagnosis method based on multi-source heterogeneous information fusion[J].China Petroleum Machinery,2021,49(2):60-67,80.
• [65] 张广远，王哲，王保宪，等.基于多源特征融合的无砟轨道砂浆层脱空病害检测方法 [J].现代电子技术，2020,43(22):62-66.Zhang Guangyuan,Wang Zhe,Wang Baoxian,et al.Method of ballastless track mortar layer void disease detection based on multi-source feature fusion [J].Modern Electronics Technique,2020,43(22):62-66.
• [66] Shen Y F,Zhong Q W,Zheng S B,et al.A multi-modal approach to rail surface condition analysis:the MFDF-net [J].IEEE Access,2024,12:132480-132494.
• [67] Miao S J,Gao L,Nian F J,et al.Detection of ballastless track interlayer gap based on vehicle’s multivariate dynamic response and deep learning [J].Mechanical Systems and Signal Processing,2025,223:111830.
• [68] 韩宏洋，李明，郄录朝，等.基于多模态融合的有砟道床脏污评估方法 [J].铁道学报，2026,48(4):187-194.Han Hongyang,Li Ming,Qie Luchao,et al.Multimodal fusion method for fouling assessment of ballasted trackbeds [J].Journal of the China Railway Society,2026,48(4):187-194.
• [69] 王玉欣，张长伦.深度时间序列异常检测在轨道状态检测中的应用研究 [J].图像与信号处理，2025(4):422-428.Wang Yuxin,Zhang Changlun.Deep time series anomaly detection and its application in track anomaly detection [J].Journal of Image and Signal Processing,2025(4):422-428.
• [70] 陈菲，杨峥，张志成，等.面向无标签数据的旋转机械故障诊断方法 [J].吉林大学学报(工学版),2022,52(11):2514-2522.Chen Fei,Yang Zheng,Zhang Zhicheng,et al.Fault diagnosis method of rotating machinery for unlabeled data [J].Journal of Jilin University (Engineering and Technology Edition),2022,52(11):2514-2522.
• [71] 邵海东，肖一鸣，颜深.仿真数据驱动的改进无监督域适应轴承故障诊断 [J].机械工程学报，2023,59(3):76-85.Shao Haidong,Xiao Yiming,Yan Shen.Simulation data-driven enhanced unsupervised domain adaptation for bearing fault diagnosis [J].Journal of Mechanical Engineering,2023,59(3):76-85.
• [72] Popov K,De Bold R,Chai H K,et al.Data-driven track geometry fault localisation using unsupervised machine learning [J].Construction and Building Materials,2023,377:131141.
• [73] Origlia A,Di Martino S,Battista E.Rail anomalies detection:a comparative analysis of three self-supervised models on real data [J].Computers in Industry,2023,148:103909.
• [74] 马超智，王阳，张淑芳，等.基于车厢内部噪声的地铁钢轨波磨自监督快速识别方法 [J].振动与冲击，2025,44(6):282-290.Ma Chaozhi,Wang Yang,Zhang Shufang,et al.Self-supervised fast identification method of metro rail corrugation based on carriage inner noise [J].Journal of Vibration and Shock,2025,44(6):282-290.
• [75] Zhuang F Z,Qi Z Y,Duan K Y,et al.A comprehensive survey on transfer learning [J].Proceedings of the IEEE,2021,109(1):43-76.
• [76] 蒋玲莉，李书慧，李学军，等.基于迁移学习和卷积神经网络的牵引电机轴承健康评估方法 [J].交通运输工程学报，2023,23(3):162-172.Jiang Lingli,Li Shuhui,Li Xuejun,et al.Health assessment method of traction motor bearing based on transfer learning and convolutional neural network [J].Journal of Traffic and Transportation Engineering,2023,23(3):162-172.
• [77] 陈超.基于迁移学习的旋转机械故障诊断方法研究 [D].南京：东南大学，2020.Chen Chao.Research on fault diagnosis method of rotating machinery based on transfer learning [D].Nanjing:Southeast University,2020.
• [78] 邱实，陈斌，胡文博，等.基于深度风格迁移合成数据的扣件异常状态检测 [J].铁道学报，2024,46(10):122-131.Qiu Shi,Chen Bin,Hu Wenbo,et al.Detection of fastener abnormal state based on deep style migration synthetic data [J].Journal of the China Railway Society,2024,46(10):122-131.
• [79] Shi P M,Zhao Y,Xu X F,et al.A small-sample cross-domain bearing fault diagnosis method based on knowledge-enhanced domain adversarial learning [J].Neurocomputing,2025,658:131699.
• [80] 王阳，肖宏，张智海，等.基于一维卷积神经网络的钢轨波磨迁移诊断方法 [J].铁道学报，2025,47(4):115-123.Wang Yang,Xiao Hong,Zhang Zhihai,et al.Rail corrugation migration diagnosis method based one-dimensional convolutional neural network [J].Journal of the China Railway Society,2025,47(4):115-123.
• [81] 戴宝锐，沈汝康，徐幼麟，等.板式无砟轨道层间病害识别技术综述 [J].铁道学报，2025,47(12):150-166.Dai Baorui,Shen Rukang,Xu Youlin,et al.Review of interlayer defect identification for ballastless slab tracks [J].Journal of the China Railway Society,2025,47(12):150-166.
• [82] Jin T T,Yan C L,Chen C H,et al.Light neural network with fewer parameters based on CNN for fault diagnosis of rotating machinery [J].Measurement,2021,181:109639.
• [83] Zhang X R,Wen X,Lin L F,et al.Robust GSR array-assisted fast intelligent diagnosis for multi-harmonic fault characteristics under variable-speed conditions [J].Mechanical Systems and Signal Processing,2025,241:113445.
• [84] 李世运，于重重，秦勇，等.基于FANF-Net的实时轻量轨道异常检测 [J].铁道学报，2026,48(3):92-98.Li Shiyun,Yu Chongchong,Qin Yong,et al.Real-time lightweight track anomaly detection based on FANF-net [J].Journal of the China Railway Society,2026,48(3):92-98.