• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2017, Vol. 53 ›› Issue (10): 116-124.doi: 10.3901/JME.2017.10.116

• 运载工程 • 上一篇    下一篇

滚动轴承安全域建模方法及其在高速列车异常检测中的应用

刘志亮1,2, 刘仕林1, 李兴林2, 康金龙1, 彭兴1   

  1. 1. 电子科技大学机械电子工程学院 成都 611731;
    2. 杭州轴承试验研究中心有限公司 杭州 310022
  • 出版日期:2017-05-15 发布日期:2017-05-15
  • 作者简介:

    刘志亮(通信作者),男,1984年出生,博士,副教授,格拉斯哥大学(英国)荣誉讲师、阿尔伯塔大学(加拿大)访问学者。主要研究方向为旋转机械预测与健康管理、工业大数据挖掘。

    E-mail:Zhiliang_Liu@uestc.edu.cn

  • 基金资助:
    * 国家重点研发计划(2016YF131200401)、国家自然科学基金(51505066)、中央高校基本科研业务经费(ZYGX2015J081)、中国博士后科学基金(2016T90842)和浙江省博士后科研项目择优(B5H1502002)资助项目; 20160604收到初稿,20170115收到修改稿;

Safety Domain Modelling of Rolling Bearings and Its Application to Anomaly Detection for High-speed Rail Vehicles

LIU Zhiliang1,2, LIU Shilin1, LI Xinglin2, KANG Jinlong1, PENG Xing1   

  1. 1. School of Mechatronics Engineering, University of Electronic Science and Technology of China, Chengdu 611731;
    2. State Testing Laboratory of Hangzhou Bearing Test and Research Center, Hangzhou 310022
  • Online:2017-05-15 Published:2017-05-15

摘要:

安全域是一种从域的角度描述轴承安全状态的模型。它从轴承的运行状态角度考虑,在状态特征变量确定的空间内,用于辨识其运行状态是否在正常区域内。然而,高速列车运行状态数据通常是符合期望的正常样本,故障样本由于获取代价高昂,使得对其故障行为知之甚少,甚至一无所知,这意味着传统基于正常和故障样本的安全域建模方法很难应用在高速列车故障诊断上。支持矢量数据描述(Support vector data description, SVDD)虽然能够仅利用正常样本在运行状态空间上构建安全域,但是基于SVDD的安全域模型易受到惩罚参数取值的影响,特别是在大数据的背景下,合理快速地选择惩罚参数对提高安全域模型的边界估计能力及异常检测具有重要意义。鉴于此,提出了一种基于核空间距离熵的安全域惩罚参数选择算法。该算法依据核空间上样本点的位置分布,计算每个样本点距离核中心的距离,在大量基准数据集的基础上,寻找最优惩罚参数与距离熵之间的对应关系,最终获得最优惩罚参数的经验表达式。试验结果表明,相比基于网格搜索和遗传算法的参数选取方法,该方法物理意义明确,能够在保证分类准确率的基础上显著提高安全域模型的优化效率。所提出的基于安全域模型的异常检测方法应用于圆锥滚子轴承的安全状态辨识中。

关键词: 惩罚参数, 滚动轴承, 异常检测, 支持矢量数据描述, 安全域

Abstract:

Safety domain can be used to describe bearing health condition from the perspective of the concept of domain. It is used to identify whether a bearing is normal in a feature space established by the operation condition. However, high-speed rail vehicles usually operate in normal condition, that is, the collected data are usually normal samples rather than faulty ones. Actually, it is costly to get faulty samples for high-speed rail vehicles. Therefore, we know very little or even nothing about their fault behaviors. It means that the conventional methods, based on both normal and fault samples, cannot be applied to fault diagnosis of high speed rail vehicles. Support vector data description (SVDD) is a tool to use only normal samples in the feature space to build the so-called safety domain; however, the SVDD based models are easily influenced by the penalty parameter. Particularly in the context of big data, it is crucial to select the penalty parameter in a reasonable and quick way so that it helps to improve accuracy of both the safety domain boundary and the anomaly detection. For this reason, a penalty parameter selection method based on the distance entropy in the kernel space of safety domain is proposed. Based on the position distribution of samples in the kernel space, the method calculates the distance between every sample and the kernel center, computes the distance entropy of the training data, finds the relationship between the optimal penalty parameter and the distance entropy on the basis of a large number of reference data points, and finally provides an empirical formula to determine the optimal penalty parameter. Experimental results show that, comparing with grid search methods and genetic algorithm, the proposed method can be easily interpreted and can also significantly improve efficiency of safety domain modeling without much accuracy loss. The proposed approach is applied to anomaly detection of a wheel set bearing for high-speed rail vehicles.

Key words: anomaly detection, penalty parameter, rolling bearing, support vector data description, safety domain