Research On Fault Diagnosis Of High Voltage Circuit Breakers Based On Vibration Signal Analysis

High-voltage circuit breakers are important electrical switching devices and also have a decisive role in the safe and stable operation of power systems.If the circuit breaker fails,it will affect the safe and stable operation of the power system,thus affecting the lives and production of the residents and causing major economic losses.Therefore,the timely detection and identification of high-voltage circuit breaker faults is a problem worthy of study.This thesis studies the fault identification method of high voltage circuit breakers.Firstly,the basic structure,working principle and classification characteristics of the high voltage circuit breaker are introduced in detail.The common fault types of the high voltage circuit breaker are summarized.Then,the characteristics of the high voltage circuit breaker vibration signal are analyzed,and the vibration signal acquisition is given Method;Then,using the Hilbert-Huang Transform method to process the vibration signal waveforms,respectively,three failures of the high-voltage circuit breaker normal working condition,the closing core stuck,the base screw loose and the abnormal state of the lock bolt Vibration signal waveform.The graph is processed to obtain the EEMD decomposition maps and their respective marginal maps in each state.Thus,the corresponding frequency eigenvalues are calculated,the corresponding eigenvalue database is established,and then the actual vibrations are processed.The signal waveforms are compared with the eigenvalues in the established eigenvalue database to identify the corresponding fault type.Finally,using the first four-order eigenmode function(IMF)component envelope information of the EEMD decomposition graph,the energy entropy eigenvalues of the signal are extracted through equal time segmentation,and then the failure classification is performed using a hierarchical support vector machine method.Realizing the fault identification of high voltage circuit breakers.Through simulation experiments,it is found that this method has a higher accuracy than the traditional BP neural network method and the one-to-one support vector machine principle method.Therefore,the fault diagnosis method based on the feature vector of EEMD-IMF energy entropy and combined with the hierarchical support vector machine can effectively identify various fault types of the high-voltage circuit breaker and has certain practical value.