基于声发射技术的轨道车辆车轴疲劳裂纹在线监测和风险评估

In light of recent accidents in the rail industry, the assessment of the mechanical integrity of railway vehicle is of vital importance. Although structural damage can happen to any structural component of railway vehicle, one of the most common types of damages is axle cyclic fatigue damage on high speed passenger trains and heavy haul freight trains. Axle are experience cyclic load conditions, are difficult to access for maintenance and are vulnerable to fatigue cracks nucleation and growth. Whilst Existing techniques for testing and inspection of railway vehicle axle such as ultrasound and magnetic particle test which are scheduled maintenances can be employed for crack detection,several defects have clearly been missed. Predicting and preventing the crack phenomenon has attracted the attention of researchers and has continued to provide a large incentive for the use of condition monitoring techniques to detect the earliest stages of cracks..This project presents an experimental study on the applicability of Acoustic Emissions (AE) for the detection of active fatigue cracks in railway vehicle axle in real-time. An experimental test-rig will be employed for this programme for generating natural degradation on a railway vehicle axle. A set of laboratory experiments will be carried out on the scaled test rig to characterise the continuous AE generated by a railway vehicle axle and, secondarily, to find the characteristic index by using not only the traditional AE parameters but also the recently developed signal processing techniques such as Hilbert-Huang transform (HHT) and so on as indications of crack initiation and/or propagation. Furthermore, we will use risk forecasting model commonly used in economic fields such as vector autoregression(VAR) model to study the relationship between the effect factors, e.g. speed and the characteristic index.And, this project will try to assess supply the crack risks of railway vehicle axles in a value-at-risk (VaR) model..The project presented is part of an first step to develop the non-destructive technique of AE for assessing the surface integrity of railway vehicle axle. The AE technique,including the HHT method,the VAR and the VaR model are not new but the application in this particular instance are unique. It is concluded that the study will offer a potential complementary tool for railway vehicle axle detection and will also offer a new application area for Acoustic Emission (AE) technology.

近期铁路行业故障频发,高速旅客列车和重载货物列车的车轴疲劳破坏是常见故障。目前检测采用定期维修的超声波探伤技术与磁粉探伤技术，有漏检。本项目应用声发射技术(AE)对轨道车辆车轴疲劳裂纹扩展开展在线监测的实验研究。将设计一套实验装置，模拟轨道车辆车轴产生疲劳裂纹的状态。然后开展AE实验，采用参数法及先进信号处理技术如希尔伯特-黄变换（HHT）等方法找到可标明裂纹形成及扩展的特征指标。还开展轨道车辆车轴风险预测评估研究，将经济学领域常用的风险预测评估模型如向量自回归模型（VAR）和风险价值模型(VaR)引入本研究，研究裂纹特征指标与影响因素如车速等的关系，并评估车轴裂纹风险。本项目提出将已在各自领域广泛应用的AE技术，HHT方法，VAR和VaR模型引入本项目研究中将具有创新性。因此，本项目的研究将为轨道车辆车轴检测提供一种有潜力的补充工具，同时也为声发射技术开辟一个新的应用领域。

近期铁路行业故障频发,高速旅客列车和重载货物列车的车轴疲劳破坏是常见故障。目前检测采用定期维修的超声波探伤技术与磁粉探伤技术，有漏检。本项目应用声发射技术(AE)对轨道车辆车轴疲劳裂纹扩展开展在线监测的实验研究。设计了一套实验装置，模拟轨道车辆车轴产生疲劳裂纹的状态。然后开展了AE实验，采用参数法及先进信号处理技术如局部均值分解（LMD）、FIXED-EMD–ENN、数据流、滑动窗口、提升小波熵等方法找到可标明裂纹形成及扩展的特征指标。还开展轨道车辆车轴风险预测评估研究，将经济学领域常用的风险预测评估模型如向量自回归模型（VAR）、灰色聚类等引入本研究，评估车轴裂纹风险。本项目提出将已在各自领域广泛应用的AE技术，LMD、EMD–ENN、数据流、滑动窗口、提升小波熵，VAR模型、灰色聚类等引入本项目研究中具有创新性。因此，本项目的研究将为轨道车辆车轴检测提供一种有潜力的补充工具，同时也为声发射技术开辟一个新的应用领域。