Research On Fault Diagnosis Of Machine Based On Vibration Signal Of GIS

In recent years,the rapid development of the country's science and technology process,rising of the corresponding voltage level in system,also will increasing the reliability of the power quality required.The SF6 gas-insulated metal-enclosed switchgear widely used in the power grid system because of occupancy less area and high reliability.But in recent years,the power grid system failure rate has been high.GIS is an important part of the power grid.Once,failure occurs will give the power grid system huge irreparable loss by the large-scale blackout.So,the grid system to ensure the safe operation of GIS benefits the nation and the people.This article is committed to research the two common GIS machinery fault,fault diagnosis has three steps,acquiring the vibration signal,feature extraction,working condition recognition.The second step is so crucial that was deeply studied in this paper.The fundamental principle and algorithm of Variational Mode Decomposition(VMD)was studied,and the Chaotic Cuckoo optimization algorithm was used to solve the problem that parameters of VMD were evaluated difficulty.Then,aimed at test functions,a comparison of the standard Cuckoo optimization algorithm and the Chaotic Cuckoo optimization algorithm were processed by numerical simulation.The search efficiency of the Chaotic Cuckoo optimization algorithm was higher,which was verified.Fault simulator GIS was the experiment setup,two common GIS failure like screws loosening and particle vibration are built.In this paper,a fault diagnosis method for the GIS machinery fault based on the Chaotic Cuckoo optimization algorithm was proposed to optimize VMD parameter.Firstly,the Chaotic Cuckoo optimization algorithm was took advantage of to confirm parameters of VMD.Secondly,three kinds of GIS vibration signals were processed by VMD,the effective components were selected to reconstruct through the Kullback-Leibler divergence.Thirdly,feature vector sets of training samples were clustered by Linear Decreasing Weight PSO K-Means,clustering centers of different signal types were obtained,the principle of the minimum Euclidean distance was adopted as the recognition method of feature vector sets of testing samples,realizing the diagnosis of GIS mechanical fault.Based on simulation result,a GIS mechanical fault on-line monitoring system based on abnormal vibration signal was designed in this article based on the VMD.Its hardware part mainly includes gathering module,output module and analysis module.The gathering module aims at gathering the GIS vibration signals;output module is data acquisition card as the main body,and completes the function of A/D conversion;analysis module is a part whose data is imported in microcomputer processing through data acquisition card,finally realizing the signal feature extraction.The major part of software is the VMD algorithm.Otherwise this paper adds data memory module,historical data retrieval module,which enrich the functions of software greatly.This paper completes all unit testing of GIS mechanical fault software and hardware system with the recording of running GIS as the vibration sources.As a result,when GIS occurs the abnormal mechanical vibration,VMD entropy of diagnostic system often deviates from its nominal values,and varies significantly,which proves the effect diagnostic system.