Research On Mechanical Fault Analysis Technology Of High Voltage Circuit Breaker Based On Acoustic And Vibration

High voltage circuit breaker is one of the most important power system switchgear,which bears a role of protection and control in power system.The acoustic and vibration signa ls generated by the circuit breaker during operation contain a lot of characteristic information representing the operation status of the circuit breaker.Therefore,the operation state of the circuit breaker can be judged by analyzing the acoustic and vibration signals generated during the operation of the circuit breaker.This thesis takes ZN65 high voltage circuit breaker as the research object,and analyzes the mechanical fa ult of the high voltage circuit breaker on the basis of acoustic and vibration signa ls.According to the collected acoustic and vibration signals,this thesis studies four operation states of the circuit breaker:normal operation,spring connecting rod falling off,refus ing closing and base loosening.Firstly,in view of the problem of noise in the acoustic and vibration signa ls of the circuit breaker,this thesis proposes a noise reduction method that combines ensemble empirical mode decomposition and improved wavelet threshold function.The method uses ensemble empirical mode deco mposit ion to decompose the acoustic and vibration signals,and determines the appropria te number of modes according to the correlation coefficient and spectrum.Compared with the vibration signal,the acoustic signal is more easily affected by the surrounding environment,so only the intrins ic modal function of the acoustic signal is further reconstructed by the improved wavelet threshold function.The experimental results show that the proposed method can effectively reduce the noise in the signal.Then,aiming at the problem that the local search ability of flower pollination algorithm is weak and easy to fall into local extreme in the process of parameter search of support vector machine,this thesis proposes an improved flower pollination algorithm.The algorithm can jump o ut of the local extreme and obtain the global optimal solution by incorporating the frog leaping idea and hybrid mutation strategy.The energy entropy fusion of acoustic and vibration signals is used as the feature vector,which is input into the support vector machine optimized by the improved flower pollination algorithm.The experimental results show that the algorithm is effective.Finally,this thesis designs and imp lements a high voltage circuit breaker mechanical fault analysis system.The test shows that the system can complete the functions of signal acquisition,data processing and fault analysis of circuit breaker.