张昌凡,刘佳峰,何静,刘建华.基于改进卷积双向门控循环网络的轴承故障诊断*[J].电子测量与仪器学报,2021,35(11):61-67
基于改进卷积双向门控循环网络的轴承故障诊断*
Improved CNN BiGRU method for bearing fault diagnosis
  
DOI:
中文关键词:  卷积神经网络  双向门控循环网络  通道注意力机制  轴承故障诊断
英文关键词:convolutional neural network  bidirectional gated recurrent neural network  channel attention mechanism  bearing fault diagnosis
基金项目:国家自然科学基金(61733004,52172403,62173137)、湖南省自然科学基金(2021JJ30217,2021JJ50001)、湖南省教育厅(19A137)项目资助
作者单位
张昌凡 湖南工业大学电气与信息工程学院株洲412007 
刘佳峰 湖南工业大学电气与信息工程学院株洲412007 
何静 湖南工业大学电气与信息工程学院株洲412007 
刘建华 湖南工业大学电气与信息工程学院株洲412007 
AuthorInstitution
Zhang Changfan Technology College of Electrical and Information Engineering, Hunan University of Technology, Zhuzhou 412007, China 
Liu Jiafeng Technology College of Electrical and Information Engineering, Hunan University of Technology, Zhuzhou 412007, China 
He Jing Technology College of Electrical and Information Engineering, Hunan University of Technology, Zhuzhou 412007, China 
Liu Jianhua Technology College of Electrical and Information Engineering, Hunan University of Technology, Zhuzhou 412007, China 
摘要点击次数: 410
全文下载次数: 0
中文摘要:
      针对传统深度学习方法没有充分利用轴承信号的时序特点,以及难以处理动态数据的问题,提出一种基于改进卷积双向门控循环神经网络的轴承故障智能诊断方法。采用卷积神经网络从输入信号中提取代表性特征,引入双向门控循环神经网络挖掘故障数据在时间维度上的语义信息,通过注意力机制自适应地对特征图通道赋予不同权值,从而实现高精度的轴承故障诊断。在公开轴承数据集上进行实验,实验结果表明,该方法能够正确地将轴承故障分类,分类精度可达996%。
英文摘要:
      Aiming at the problems that traditional deep learning methods do not make full use of the timing characteristics of bearing signals, and are difficult to process dynamic data, an improved CNN BiGRU intelligent diagnosis method for bearing faults is proposed. The convolutional neural network is used to extract representative features from the input signal, and the bidirectional gated recurrent neural network is introduced to mine the semantic information in the time dimension of the fault data, and the attention mechanism is used to adaptively assign different weights to the feature map to achieve high precision diagnosis of bearing faults. Experiments on public bearing data sets show that the method can correctly classify bearing faults with a classification accuracy of 996%。
查看全文  查看/发表评论  下载PDF阅读器