As a pivotal component of electric motors,bearings primarily function to support and guide shafts,reduce equipment friction,and facilitate the connection of different devices.Accurate identification of bearing failure types and assessment of their health status are of significant importance for the rational scheduling of equipment maintenance.This study devises a real-time fault diagnosis and performance degradation assessment system for electric motor bearings based on the LabVIEW virtual instrument platform.Firstly,by leveraging the feature extraction capability of convolutional neural networks(CNN),the system autonomously learns fault features from raw vibration signals,constructs a fault diagnostic model on the LabVIEW platform,and achieves real-time diagnosis of bearing operating conditions.Secondly,by applying wavelet denoising to the raw vibration signal and extracting their time-domain features,a comprehensive indicator representing bearing performance degradation is obtained using principal component analysis(PCA).The software for the fault diagnosis and performance degradation assessment system for electric motor bearings is developed on the LabVIEW platform.Experimental results of online fault diagnosis and performance assessment validate the real-time effectiveness and efficiency of the proposed system.
关键词
电动机轴承/故障诊断/性能退化评估/卷积神经网络
Key words
electric motor bearing/fault diagnosis/performance degradation assessment/convolutional neural network
维普
万方数据