期刊文献+

基于双层长短时记忆网络的齿轮故障诊断方法 被引量:8

Gear Fault Diagnosis Based on Binary Long Short Term Memory Network
原文传递
导出
摘要 为了提高齿轮故障诊断准确率,解决齿轮故障诊断中数据量大、提取特征困难等问题,构建了齿轮故障诊断系统,采用深度学习方法建立了齿轮故障诊断模型,提出一种基于双层长短时记忆(Binary Long Short Term Memory,Bi LSTM)网络的故障诊断方法,并对该方法进行了性能分析和对比实验。结果表明:采用Bi LSTM网络方法进行齿轮故障诊断的准确率达到99.76%,分类效果优于支持向量机、Xg Boost、卷积神经网络和长短时记忆(LSTM)网络等方法,有效地提高了故障诊断精度。 In order to improve the accuracy of gear fault diagnosis and solve the problems of large amount of data and difficult feature extraction, this paper constructs a gear fault diagnosis system and uses deep learning methods to establish a gear fault diagnosis model. A gear fault diagnosis method based on Binary Long Short Term Memory (BiLSTM) network is proposed, and the performance analysis and the compari- son experiment of this method are carried out. The experimental results show that the accuracy of gear fault diagnosis using BiLSTM network method improves effectively and reaches 99.76%. The classifica- tion effect of BiLSTM network method is superior to Support Vector Machine ( SVM), XgBoost, Convolu- tional Neural Network (CNN) and LSTM network.
作者 王维锋 邱雪欢 孙剑桥 张惠民 WANG Wei-feng;QIU Xue-huan;SUN Jian-qiao;ZHANG Hui-min(Information and Communication Department,Army Academy of Armored Forces,Beijing 100072,China;Equipment Support and Remanufacturing Department,Army Academy of Armored Forces,Beijing 100072,China)
出处 《装甲兵工程学院学报》 2018年第2期81-85,共5页 Journal of Academy of Armored Force Engineering
基金 军队科研计划项目
关键词 齿轮 故障诊断 双层长短时记忆(BiLSTM)网络 深度学习 gear fault diagnosis Binary long Short Term Memory (BiLSTM) network deep learning
  • 相关文献

参考文献10

二级参考文献55

共引文献220

同被引文献54

引证文献8

二级引证文献30

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部