Bearing Fault Diagnosis Based on Graph Formulation and Graph Convolutional Network

作     者：Xin Wang Wenjin Zhou Xiaodong Li 

作者机构：Research and Development Center of Smart Information and Communications TechnologiesShanghai Advanced Research InstituteChinese Academy of SciencesNo.99 Haike RoadShanghai 201210China Key Laboratory of Noise and Vibration ResearchInstitute of AcousticsChinese Academy 

出 版 物：《Journal of Dynamics, Monitoring and Diagnostics》 (动力学、监测与诊断学报（英文）)

年 卷 期：2023年第2卷第4期

页      面：252-261页

学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学] 

主　　题：bearing fault diagnosis deep learning graph convolutional network 

摘      要：Bearing fault diagnosis stands as a critical component in the maintenance of rotating *** prevalent deep learning techniques are tailored to Euclidean datasets such as audio,image,and ***,these methods falter when confronting non-Euclidean datasets,notably graph *** response,here we introduce an innovative approach harnessing the graph convolutional network(GCN)to analyze graph data derived from vibration signals related to bearing *** enhances the precision and reliability of fault *** methodology initiates by deriving a periodogram from the unprocessed vibration ***,this periodogram is mapped into a graph format,upon which the GCN is engaged for classification *** substantiate the efficacy of our approach through rigorous experimental assessments conducted on a collection of ten bearing *** these experiments,an accelerometer chronicles vibration signals across varying load *** probe into the diagnostic accuracy rates across diverse loads and signal-to-noise ***,a comparative evaluation of our method against several established algorithms delineated in this study is *** observations confirm that our GCN-based strategy registers an elevated diagnostic accuracy quotient.