基于深度神经网络的滚动轴承故障诊断

发布时间：2018-01-18 05:09

本文关键词：基于深度神经网络的滚动轴承故障诊断　出处：《组合机床与自动化加工技术》2017年11期 　论文类型：期刊论文

【摘要】：针对浅层神经网络方法依靠专家经验和信号处理技术进行人工提取大量滚动轴承数据的特征变得越来越困难,而且神经网络浅层结构限制了神经网络学习复杂非线性关系的功能。结合深度学习在特征提取和处理大数据等优势,研究一种基于深度神经网络的滚动轴承故障诊断方法。该方法直接从原始数据中提取出有用特征,所提取的特征作为BP神经网络(BPNN)的输入识别滚动轴承的故障类别。通过对滚动轴承正常状态,内圈故障,外圈故障和滚珠故障四种状态以及各个状态的不同故障严重程度的分析,实验结果表明所研究的的方法不仅仅能够从原始信号中挖掘出有用的故障特征,还可以诊断出故障的严重程度,和BPNN相比具有更高的诊断准确率。
[Abstract]:It is becoming more and more difficult to extract a large number of rolling bearing data manually based on expert experience and signal processing technology in shallow neural network. Moreover, the shallow structure of neural network limits the function of neural network in learning complex nonlinear relations, which combines the advantages of deep learning in feature extraction and processing big data. A fault diagnosis method for rolling bearing based on depth neural network is studied, in which useful features are extracted directly from the original data. The extracted feature is used as the input of BP neural network (BPNN) to identify the fault category of rolling bearing. The four states of outer ring fault and ball fault and the different fault severity of each state are analyzed. The experimental results show that the proposed method can not only extract useful fault features from the original signal. The degree of fault severity can also be diagnosed, and the accuracy of diagnosis is higher than that of BPNN.
【作者单位】：中北大学计算机与控制工程学院;中北大学机械与动力工程学院;
【分类号】：TH133.33;TP183
【正文快照】： 0引言滚动轴承在旋转机械中应用极为广泛,是传递力矩及动力的重要零部件,但因其工作环境恶劣,也是最容易损坏的零部件之一[1]。因此,对轴承进行状态监测与故障诊断具有重要实践意义。为了全面的检测滚动轴承的健康状况,用状态监测系统采集实时数据,在机器长时间的运转中获取大

【相似文献】