期刊文献+

基于压缩激励残差神经网络的轴承损伤诊断

A bearing damage diagnosis method based on compressed excitation residual neural network
下载PDF
导出
摘要 滚动轴承是旋转机械的重要零件之一,文中针对滚动轴承损伤类型有效识别问题,提出了一种基于压缩激励残差神经网络的滚动轴承损伤诊断方法。本方法对轴承原始振动信号使用连续小波变换提取特征,形成二维时频样本,再利用样本对压缩激励残差神经网络进行训练,最后在全连接层使用Softmax分类器实现对轴承损伤的分类。用QPZZ-Ⅱ旋转机械振动损伤实验平台数据验证模型性能。实验结果表明:该方法对不同负载下滚动轴承损伤识别的准确率达99.95%,具有良好的泛化性和鲁棒性。 A rolling bearing is one of the important parts of rotating machinery.In order to effectively identify the damage types of rolling bearings,a damage diagnosis method of rolling bearings based on squeeze and excitation residual neural network is proposed.In this method,continuous wavelet transform is used to extract the characteristics of the original vibration signal of bearings to collect two-dimensional time-frequency samples which are used to train the squeeze and excitation residual neural network.Finally,at the fully connected layer,the bearing damage is classiified by using a Softmax classifier.The QPZZ-Ⅱrotating machinery vibration damage experimental platform is used to verify the model performance with data.Results show that the accuracy rate of damage identification of rolling bearings under different loads is as high as 99.95%,and this method can also bring a good generalization and robustness.
作者 韩元政 谷艳玲 陈长征 田淼 孙鲜明 Han Yuanzheng;Gu Yanling;Chen Changzheng;Tian Miao;Sun Xianming
出处 《起重运输机械》 2022年第5期30-34,70,共6页 Hoisting and Conveying Machinery
基金 国家自然科学基金项目(51675350)。
关键词 滚动轴承 小波变换 压缩激励残差神经网络 损伤诊断 rolling bearing wavelet transform squeeze and excitation residual neural network damage diagnosis
  • 相关文献

参考文献6

二级参考文献47

  • 1周小勇,叶银忠.基于Mallat塔式算法小波变换的多故障诊断方法[J].控制与决策,2004,19(5):592-594. 被引量:14
  • 2Wiggins S.Introduction to apphed nonlinear dynamical systems and chaos[M].New York:Springer-Vedag,1990.
  • 3Marcin S, Piotr W. Neuro-wavelet classifiters for EEG signals based on rough set methods[J]. Neurocomputing, 2001,36(1) : 103-122.
  • 4Mallat S.A theory for multiresolution signal decomposition:the wavelet representation[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1989,11(7): 674-693.
  • 5Mallat S , Hwang W L. Singularity detection and processing with wavelets[J]. IEEE Trans on Information Theory, 1992,38(2) :617-643.
  • 6Daubechies I. The wavelet transform, time-frequency localization and signal analysis[J]. IEEE Trans Inform Theory, 1990,36:961-1005.
  • 7Canny J.A wavelet transform and edge detection[J] .IEEE Trans Pattern Anal Machine Intell, 1986,8: 679-698.
  • 8Mallat S,Zhong S. Characterization of signals from multi-scale edges[J]. IEEE Trans Pattern Analysis and Machine Intelligence, 1992,14(7) :710-732.
  • 9Satish L. Short-time Fourier and Wavalet transforms for fault detection in power transformers during impulse tests[J]. In:IEEE Proc Sco Meas Technol, 1998,145(2) :77-84.
  • 10Donoho D L. De-noising via soft thresholdingIJ]. IEEE Transon Information Theory, 1995,41(5) :613-627.

共引文献146

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