多域特征提取结合AdaBoost的含未知故障提速道岔故障诊断方法  

Multi-domain Feature Extraction Combined with AdaBoost for Fault Diagnosis of Speed-up Turnout with Unknown Fault

作　　者：郑云水[1] 张亚宁 ZHENG Yunshui;ZHANG Yaning(School of Automation and Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学自动化与电气工程学院,兰州730070

出　　处：《机械科学与技术》2024年第8期1350-1358,共9页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金地区科学基金项目(61763023);甘肃省科技厅计划(20YF8GA037)。

摘　　要：针对提速道岔未知新故障误判影响列车安全运行及道岔检修效率的问题,提出一种基于多域特征提取和自适应提升算法(Adaptive boosting, adaboost)的信号分析及故障诊断模型。首先,为了深入挖掘道岔的故障特征,分别从时域、频域及时频域提取故障特征,构造原始特征集;然后根据AdaBoost模型获得的特征重要度排序构造不同特征数量的分类模型,并利用模型分类精度进一步获得最佳特征子集;最后将最佳特征子集输入含判定机制的AdaBoost故障诊断模型,完成对提速道岔含未知故障类型的诊断,同时,通过对模型的再训练,实现了对现有故障诊断模型的自适应更新。结果表明:本文方法在有效提取故障特征,提高道岔已知类故障诊断精度的同时,可以有效地识别出道岔之前未出现的新故障。In view of the problem that the unknown new faults of speed-increasing turnouts are misjudged,which will affect the operation safety of trains and the turnouts maintenance efficiency.A signal analysis and fault diagnosis model based on multi-domain feature extraction and adaptive boosting algorithm(AdaBoost)is proposed.Firstly,in order to deeply extract the fault features of turnouts,the original feature set is constructed by extracting fault features from time domain,frequency domain,and time-frequency domain respectively.Secondly,the classification models with different numbers of features are constructed based on the feature importance ranking obtained by the AdaBoost model,and the best feature subset is further obtained by using the classification accuracy of the model.Finally,the best feature subset is input into the AdaBoost fault diagnosis model with a decision mechanism to complete the diagnosis of unknown faults on the speed-increasing turnout.Meanwhile,through the retraining of the model,the adaptive update of the existing fault diagnosis model is realized.Research indicates:the algorithm in this paper can effectively extract the fault features and improve the diagnosis accuracy of the known faults of the turnout,at the same time,the method can effectively identify new faults that have not occurred ago.

关 键 词：特征提取 ADABOOST 未知故障 提速道岔 故障诊断 

分 类 号：U284.92[交通运输工程—交通信息工程及控制]