电子测量与仪器学报

唐波,陈慎慎.基于深度卷积神经网络的轴承故障诊断方法[J].电子测量与仪器学报,2020,34(3):88-93

基于深度卷积神经网络的轴承故障诊断方法

Method of bearing fault diagnosis based on deep convolutional neural network

DOI：

英文关键词:rolling bearing fault diagnosis convolutional neural network

基金项目:国家自然科学基金(11872061)、国家市场监督管理总局科技计划(2019MK103)资助项目

作者	单位
唐波	1.中国计量大学计量测试工程学院，2.浙江大学浙江省先进制造技术重点实验室
陈慎慎	1.中国计量大学计量测试工程学院

Author	Institution
Tang Bo	1.College of Metrology and Measurement Engineering, China Jiliang University，2. Zhejiang Province Key Laboratory of Advanced Manufacturing Technology, Zhejiang University
Chen Shenshen	1.College of Metrology and Measurement Engineering, China Jiliang University

摘要点击次数: 512

全文下载次数: 859

中文摘要:

针对传统浅层轴承故障诊断方法依赖于人工特征提取和诊断专业知识从而缺乏自适应性问题，结合卷积神经网络善于识别二维形状的特点，提出一种基于深度卷积神经网络的故障诊断方法（DCNN）。首先，为充分展现滚动轴承故障特征信息，利用短时傅里叶变换得到滚动轴承振动时间序列的二维时频谱；其次，通过卷积神经网络自适应提取时频谱中不同故障特征；最后，将提取的轴承故障特征利用Softmax分类器输出诊断结果，实现轴承故障诊断。通过实测故障轴承数据对该方法进行验证，结果表明DCNN在多故障、变负载的轴承故障诊断准确率高达999%，证明了所提方法具有良好的泛化性能和可行性。

英文摘要:

Aiming at the lack of adaptability as the traditional shallow bearing fault diagnosis method relying on artificial feature extraction and diagnosis expertise, a fault diagnosis method based on deep convolutional neural network is proposed to recognize two dimensional shapes. Firstly, in order to fully display the fault characteristic information of rolling bearing, the two dimensional time spectrum of rolling bearing vibration time series is obtained by using the short time Fourier transform. Secondly, different fault features are extracted by convolutional neural network adaptively. Finally, Softmax classifier is used to output the diagnosis results to realize bearing fault diagnosis. The results show that the accuracy of the measured bearing fault diagnosis is up to 999%, proves that the proposed method has a good generalization performance and feasibility.

查看全文查看/发表评论下载PDF阅读器