Research On High Voltage Circuit Breaker Status Based On Sound And Vibration Signal

A circuit breaker is one of the most important power system switchgear,which bears the role of protection and control in power system.In order to ensure the normal operation of the power system,fault detection and diagnosis of the circuit breaker it is very necessary.According to statistics,the main failure of the circuit breaker due to its complex mechanical structure,sound and vibration signals generated in the operation belong to the non-stationary,non-deterministic signals,can reflect the complexity of the circuit breaker mechanical operation of the process,based on in-depth study of this article high voltage circuit breaker testing signal processing methods and more data fusion.It's based on the design of multi-sensor circuit breaker test system,focusing on a variety of test signal fusion and fault diagnosis methods.Firstly,the sound signal and vibration signal generated during the operation are selected as the representative quantities of the operating state of the high-voltage circuit breaker,and a field test system of the circuit breaker is built up by sensors,high-speed acquisition card and industrial control computer.Secondly,according to the characteristics of the signal generated during the closing and closing process of the high voltage circuit breaker,the vibration and sound signals of the multi-sensor are decomposed by using wavelet packet,then the entropy input SVM is extracted.In the process of classifying the typical faults of the circuit breaker,the basic probability distribution of each fault is obtained by using the voting rule inside the SVM,and the highest classification accuracy of each support vector machine is obtained.Then,an adaptive weight breaker status identification method is proposed,and the basic probability distribution is merged by evidence fusion theory to get the current state of circuit breaker.The classification results of SVM are used as an evaluation of sensor reliability and algorithm performance.According to the combination of weighted theory and evidence theory,the aim of improving state recognition rate can be further achieved.Finally,the validity of the data fusion method in the fault diagnosis of the circuit breaker is verified by the typical failure test of the circuit breaker,the loosening of the screw and the refusal of the circuit breaker.