BP神经网络在滚动轴承故障诊断中的应用研究  

Research on the application of BP neural wetwork in fault diagnosis of rolling bearings

作　　者：王尉旭 周豪[1] 洪朝银 Wang Weixu;Zhou Hao;Hong Chaoyin(Chongqing Jiaotong University,Chongqing 400074,China)

机构地区：[1]重庆交通大学,重庆400074

出　　处：《无线互联科技》2024年第5期74-76,共3页Wireless Internet Technology

摘　　要：反向传播神经网络(Back Propagation Neural Network,BPNN)是一种深度学习模型,在各个领域都有重要应用。文章以滚动轴承故障诊断为例,探讨了BP神经网络在其中的应用。文章通过运用及优化BP神经网络,对凯斯西储大学提供的轴承故障数据加窗后进行离散傅里叶变换处理,再进行峰值特征提取,然后利用该数据进行神经网络模型的学习和预测,构建了一个能够准确预测轴承故障类型的网络模型。该模型能够提高轴承故障诊断的效率和准确性,具有重要的实用价值。The back propagation neural network(BPNN)is an important deep learning model,which has important applications and advantages in various fields.This article takes the bearing fault diagnosis as an example to mainly discuss the application of BP neural network.In this article,by using and optimizing BP neural network,the bearing fault data provided by Case Western Reserve University is processed by windowing and discrete Fourier transform,and then peak feature extraction is carried out.Then,the neural network model is learned and predicted using this data,and a network model that can accurately predict the bearing fault type is constructed.This model can improve the efficiency and accuracy of bearing fault diagnosis,and has important practical value.

关 键 词：BP神经网络 故障诊断 滚动轴承 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP277[自动化与计算机技术—控制科学与工程]