Research On Improved Fuzzy C-means Clustering Algorithm In Equipment Fault Diagnosis

With the rapid development of the information age,equipment systems tend to be integrated and complicated.It is of great research significance to find faults from a large number of data states and deal with them in time to ensure that the equipment can operate efficiently and reliably during work.In the face of large-scale equipment,it is particularly important to diagnose faults efficiently.However,due to the diversity and complexity of actual equipment faults,it is very difficult to identify and diagnose faults.At the same time,there is a certain degree of overlap in the response data of different fault types,which makes it difficult to accurately classify and judge such faults.The research of fault diagnosis in the equipment is to locate the fault on the highly integrated circuit board.Due to the large number of components and the existence of tolerance,and the uncertainty of the fault,fuzzy theory and cluster analysis can be used to solve this problem.Fuzzy C-means clustering(FCM)uses clustering degree to avoid clustering.FCM can be seen as a common product of fuzzy theory and cluster analysis.Since most equipment systems are composed of analog signals,the most widely used analog signals are analog circuits,and certain research results have been achieved.This article will use the analog circuit as the research object to conduct fault diagnosis in the equipment research,locate the fault on the highly integrated circuit board,and use the two-stage four op amp high-pass filter and Sallen-key low-pass filter circuit in the analog circuit respectively Perform verification and diagnosis.Firstly,the wavelet theory is used as the basis for discussion,and the wavelet packet decomposition is used for feature extraction,and then the dimensionality reduction process is combined with the Local Linear Embedding(LLE)based on manifold learning,and finally the features after dimensionality reduction The data is used as the input of fuzzy C-means clustering(FCM)for fault diagnosis.For the local optimal situation of fuzzy C-means clustering diagnosis,animproved fuzzy C-means clustering fault diagnosis is proposed.This algorithm performs secondary FCM diagnosis for the case where the diagnosis rate is particularly low,and performs cluster diagnosis on the fault again to diagnose The correct rate is the evaluation standard,and the proposed diagnosis scheme is verified and evaluated.The diagnosis effect has been significantly improved.The experimental results of different circuits have verified that the fault diagnosis framework proposed in this paper has certain adaptability in analog circuit diagnosis.The diagnosis results show that the improved fuzzy C-means clustering algorithm is good in equipment fault diagnosis.