基于长短期记忆网络的动车组轴箱轴承故障诊断预测模型研究
刘冠男常振臣高明亮赵明高珊
Fault Diagnosis Model for EMU Bogie Bearing Based on LSTM
LIU GuannanCHANG ZhenchenGAO MingliangZHAO MingGAO Shan
-
作者信息:中车长春轨道客车股份有限公司国家轨道客车工程研究中心，130062， 长春
-
Affiliation:National Engineering Research Center of Railway Vehicles, CRRC Changchun Railway Vehicles Co., Ltd., 130062, Changchun, China
-
关键词:
-
Key words:
-
DOI:10.16037/j.1007-869x.2022.02.022
-
中图分类号/CLCN:U266.2; U270.331+2
-
栏目/Col:研究报告
摘要:
动车组轴箱轴承是动车组转向架的关键部件，其运行品质直接影响动车组的运营安全。以深度学习算法为基础，利用轴承振动信号时间序列的特点和LSTM(长短期记忆网络)擅长处理时间序列的优势，通过构建LSTM模型对轴承的故障状态进行识别，开发了基于深度学习的轴承故障诊断预测软件，实现了轴承故障早期的分类与诊断。模型的仿真和试验表明，该诊断模型能有效地提高故障诊断的辨识精度，模型拟合优度可达到90%，辨识准确率最高可达到98%。
Abstracts:
Bogie axle box bearing is a key component of EMU bogie, its operation quality directly affects the EMU operation safety. Based on deep learning algorithm, the characteristics of bearing vibration signal time series and the advantages of LSTM (long-short term memory) in dealing with time series are used, to identify the fault state of bearing by way of constructing an LSTM model. Then, a prediction software based on deep learning is developed, so as to realize the early classification and diagnosis of bearing faults. The simulation and test results show that the diagnostic model can effectively improve the identification accuracy of fault diagnosis, the goodness of model fitting and the identification accuracy rate could reach 90% and 98% respectively.
引文 / Ref:
刘冠男, 常振臣, 高明亮, 等. 基于长短期记忆网络的动车组轴箱轴承故障诊断预测模型研究. 城市轨道交通研究, 2022, 25(2): 86.
LIU Guannan, CHANG Zhenchen, GAO Mingliang, et al. Fault diagnosis model for EMU bogie bearing based on LSTM. Urban Mass Transit, 2022, 25(2): 86.
LIU Guannan, CHANG Zhenchen, GAO Mingliang, et al. Fault diagnosis model for EMU bogie bearing based on LSTM. Urban Mass Transit, 2022, 25(2): 86.
过刊浏览
资料下载
友情链接