基于ResNet50-CBAM模型的滚动轴承故障诊断研究

打开文本图片集

中图分类号：TP18 文献标志码：A 文章编号：2095-2945（2025）19-0001-04

Abstract：Aimingattheshortcomingsintraditionalrollingbearingfaultsignalfeatureextraction，arollingbearingfault diagnosismethodbasedonConvolutionalBlockAtentionModule（CBAM）andresidualnetwork（ResNet5O）isproposed.Thefault signalsintheCaseWesternReserveUniversitydatasetwererandomlyandlocalloverlappedsampled，andthebearingfault signalswereconvertedintotwo-dimensionaltime-frequencydomainimagesusingICEEMDANandHilbert.Thetime-frequency domainimagesweretheninputintotheResNet5O-CBAMnetworkmodel.，trainingandtestingtheaccuracyofthemodel. Convolutionalneuralnetworksandtransferlearningareaddedtothenetworkmodeltosolvetheproblemsofdificultyindata acquisitionandlongtraining time.ExperimentshaveprovedthatResNet5O-CBAMhasstrong faultfeatureextractioncapablities. Compared with other network models，the accuracy rate is 8%～15% higher. Finally，rolling bearing signals are collected on a servosystemexperimentalsimulationplatform，andtheimprovednetworkmodelisusedfordiagnosis.Theresultsprovethatthis diagnosis method has high accuracy in rolling bearing fault diagnosis.

Keywords： rolling bearing; fault diagnosis; ResNet5O-CBAM; network model; data

滚动轴承作为设备随动系统中重要的组件，主要承担支撑和旋转的功能，可降低设备传动轴与支撑部位的摩擦力，同时为传动轴的轴承提供支撑。（剩余4739字）