• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2026, Vol. 62 ›› Issue (16): 343-354.doi: 10.3901/JME.260712

• 特邀专辑:轨道运载装备动态设计理论与运维技术 • 上一篇    

扫码分享

高速列车故障诊断与健康管理技术研究综述

王海明1,2, 刘永强1,2, 杨绍普2, 魏沐雨1, 魏海石1, 李昊1, 王艳丽3   

  1. 1. 石家庄铁道大学机械工程学院 石家庄 050043;
    2. 石家庄铁道大学省部共建交通结构力学行为与系统安全国家重点实验室 石家庄 050043;
    3. 河北经贸大学统计与数学学院 石家庄 050061
  • 收稿日期:2025-08-15 修回日期:2026-02-07 发布日期:2026-10-14
  • 作者简介:王海明,男,1993年出生,博士,讲师,硕士研究生导师。主要研究方向为智能运维与异常识别。E-mail:wanghm@stdu.edu.cn;刘永强(通信作者),男,1983年出生,博士,教授,博士研究生导师。主要研究方向为车辆动力学与智能运维。E-mail:liuyq@stdu.edu.cn;杨绍普,男,1962年出生,博士,教授,博士研究生导师。主要研究方向为非线性动力学与控制、机械系统智能诊断。E-mail:yangsp@stdu.edu.cn
  • 基金资助:
    国家自然科学基金(12393783,12502017,U24A2003)、河北省自然科学基金(A2025210032)、河北省高等学校科学技术研究(BJ2026090)和中央引导地方科技发展资金(246Z2206G)资助项目。

Research Review on Fault Diagnosis and Health Management Technologies for High-speed Trains

WANG Haiming1,2, LIU Yongqiang1,2, YANG Shaopu2, WEI Muyu1, WEI Haishi1, LI Hao1, WANG Yanli3   

  1. 1. School of Mechanical Engineering, Shijiazhuang Tiedao University, Shijiazhuang 050043;
    2. State Key Laboratory of Mechanical Behavior and System Safety of Traffic Engineering Structures, Shijiazhuang Tiedao University, Shijiazhuang 050043;
    3. College of Statistics & Mathematics, Hebei University of Economics and Business, Shijiazhuang 050061
  • Received:2025-08-15 Revised:2026-02-07 Published:2026-10-14

摘要: 为应对高速列车轮对系统故障随机性强、演化速度快带来的运维挑战,本研究对高速列车动力学行为与故障智能监测技术进行系统性的研究与综述。在动力学建模方面,阐述从车辆-轨道独立模型到集成轴承、齿轮箱等关键旋转部件的多系统耦合模型的发展,揭示复杂轮轨系统动态响应机理。在故障诊断方面,基于传统信号处理、机器学习、深度学习的智能诊断方法,实现车轮、轴承、齿轮等旋转部件的故障识别与预警。结论指出,融合动力学机理与数据驱动的智能诊断是实现列车故障预测与健康管理的有效途径。应进一步发展高精度耦合模型与可解释、强泛化的智能算法,推动动力学机理与运维体系协同创新,构建覆盖“感知-诊断-预测-决策”的智能运维系统。

关键词: 高速列车, 动力学建模, 故障诊断, 旋转部件, 健康管理

Abstract: To address the operational challenges posed by the high randomness and rapid evolution of faults in the high-speed train wheelset system, the high-speed train dynamic behavior and intelligent fault monitoring technology are systematically researched and summarized. In terms of dynamic modeling, the development of multi-system coupling models from independent vehicle-track models to integrated models of bearings and gearboxes is elaborated, and the complex wheel-rail system dynamic response mechanism is revealed. Furthermore, intelligent diagnosis methods based on traditional signal processing, machine learning, and deep learning have achieved fault identification and early warning for rotating components such as wheels, bearings, and gears. The conclusion indicates that integrating dynamic mechanisms with data-driven intelligent diagnosis is an effective approach for achieving train fault prognostics and health management. Further development of high-precision coupling models as well as interpretable and highly generalizable intelligent algorithms is required to promote the collaborative innovation of dynamics mechanisms and maintenance systems, thereby constructing an intelligent operation and maintenance system covering perception- diagnosis-prognosis-decision.

Key words: high-speed train, dynamic modeling, fault diagnosis, rotating parts, health management

中图分类号: