程岳梅 李小波 田世贺 陆朱剑

摘 要： 牵引系统身为地铁车辆的关键系统之一，提升其可靠性对于改善整个地铁车辆的可靠运行尤为重要。但是，牵引系统的结构复杂，且失效模式多样，采用传统的可靠性分析方法较为困难。针对这一问题，提出了一种基于贝叶斯网络的系统可靠性评估方法，并结合故障树法，对地铁车辆牵引系统的可靠性进行建模和分析。以某地铁牵引系统为例，利用贝叶斯网络模型计算出牵引系统的可靠度;通过贝叶斯网络的双向推理能力，找出对牵引系统可靠性影响最大的组件，即牵引系统的薄弱环节（IGBT反馈故障）。该方法的评估结果合理、准确，可为地铁车辆牵引系统的可靠性评估、故障诊断提供一定的理论参考。

关键词： 地铁车辆;牵引系统;可靠性分析;故障树;贝叶斯网络

文章编号： 2095-2163（2021）03-0056-05 中图分类号：U270.38;TP202+.1 文献标志码：A

【Abstract】The traction system is one of the key systems of subway vehicles， and improving its reliability is particularly important for improving the reliable operation of the entire subway vehicle. However， the structure of the traction system is complex and the failure modes are diverse， so it is difficult to adopt traditional reliability analysis methods. Aiming at this problem， a system reliability assessment method based on Bayesian network is proposed， combined with fault tree method， to model and analyze the reliability of metro vehicle traction system is modeled and analyzed. Taking a subway traction system as an example， use the Bayesian network model to calculate the reliability of the traction system; use the two-way reasoning ability of the Bayesian network to find the component that has the greatest impact on the reliability of the traction system， that is， the weak link of the traction system （IGBT feedback failure）. The evaluation results of this method are reasonable and accurate， which can provide a certain theoretical reference for the reliability evaluation and fault diagnosis of the traction system of subway vehicles.

【Key words】 subway car; traction system; reliability analysis; fault tree; Bayesian network

0 引 言

目前，地鐵车辆凭借自身的准时、运载量大、安全便捷、独立的运行环境等优点，已成为了大中型城市缓解交通拥堵的主要途径。地铁车辆是以人为载运对象的一种交通工具，因此，保障地铁车辆安全可靠地运行是重要前提。牵引系统作为地铁车辆的动力来源系统，在车辆运行期间，倘若出现由于牵引系统故障引发的各种故障形式，都有可能引发严重的交通安全事故，同时危及着出行市民的安全。因此，有必要评估地铁车辆牵引系统的可靠性，找到影响该系统可靠运行的薄弱环节，对指导牵引系统的维护检修以及进一步改善该系统的可靠性具有重大意义，从而保障地铁车辆的安全运行。……