Research On Fault Diagnosis Method Of Rolling Bearing Based On LCD And Fuzzy Clustering

As a key component of mechanical system,the state of rolling bearing has a decisive influence on whether the mechanical system can function properly,so it is important to diagnose and analyze the faults in the bearing.The focus of diagnosis is on the extraction of real information and pattern recognition.The former usually manifests itself as state characteristic parameters,while the latter is essentially a process of contrasting and classifying each other.The paper takes the vibration signal of rolling bearing as the research object,and studies and analyzes the fault diagnosis of rolling bearing from the three aspects of signal processing,feature extraction and pattern recognition.The main research work is:Firstly,in view of the problem of fault feature frequency extraction,a bearing fault diagnosis method based on maximum correlation cliff anticonvolution(MCKD)and local feature scale decomposition(LCD)is proposed.By analyzing the advantages of signal processing,the combination of the two is applied to the measured signal,and the results show that the method can extract a richer frequency of fault characteristics.Then,the arrangement entropy and fuzzy clustering algorithm are analyzed for information extraction and pattern recognition in fault diagnosis.The operation of the arrangement of entropy is fast and can measure the complexity of the object directly.By studying the evolutionary model and random signal,it is proved that it is effective in detecting signal mutation and characterization characteristic information,and the applicable parameters are analyzed,and fuzzy clustering,as an unsupervised pattern recognition method,can divide the data sets that differ from each other.The Gath-Geve cluster is not bound by the shape direction distribution and adapts well to the nonlinear type of data,which is used as a means of identification of fault type,and combines these two aspects to pave the way for pattern recognition.Finally,the experimental analysis is carried out with the single fault and compound fault of rolling bearing in the relevant test bench,and the validity and superiority of fuzzy cluster recognition of different fault conditions are verified.After MCKD-LCD processing of fault vibration signal,it is used to construct the feature information vector and input it into the clusterer for fault identification.Compared with other methods,the corresponding evaluation index is used to prove the superiority of the method proposed in the article,and the characteristic extraction and fault identification of the rolling bearings under different operating conditions are realized.The experimental results show that the fault diagnosis method based on fuzzy clustering can effectively identify bearing faults in different states and provide new concepts for the fault diagnosis of related equipment.