Vacuum Circuit Breaker Interrupter Vacuum Degree Detection And Vacuum Life Prediction Method Research

As a control and protection device,vacuum circuit breaker has been widely used in power system for half a century,especially in the field of medium and high voltage,its utilization rate reaches more than 95%,its reliable operation is directly related to the safe operation of the power system.The vacuum degree of interrupter as a key factor to measure the life of vacuum circuit breaker,the vacuum degree will deteriorate with the increase of operation and storage time,if the failure of the vacuum degree failure time is not predicted in advance,it will cause great impact on the power system,the interrupter vacuum degree detection and vacuum life prediction of comprehensive research is important to ensure the safe and stable operation of the power system.The specific research content of this paper is as follows:(1)The theoretical basis of vacuum degree and vacuum life is introduced,and the current research status of vacuum degree detection and remaining life prediction methods is briefly described,which lays the foundation for the subsequent research on vacuum degree detection and vacuum life methods.(2)The existing vacuum detection methods and their respective advantages and disadvantages are analyzed,and the main detection method studied in this paper,the pulsed magnetron discharge method,is determined,and its detection principle and related theoretical basis are introduced.(3)Based on the pulsed magnetron discharge method,a collection scheme is proposed to collect ion current signals to characterize the vacuum degree of interrupter chamber,and a vacuum degree detection system of interrupter chamber is designed and implemented.The feasibility of the vacuum detection system based on the pulsed magnetron discharge method is verified by building a vacuum simulation system,and the development of this vacuum detection system provides more accurate measurement data for subsequent vacuum life prediction.(4)Considering that the current technology for vacuum life prediction is still immature,a vacuum life prediction model based on GA-SVM algorithm is proposed.Genetic algorithm based on K-CV method is used to optimize the parameters,and the final search parameters are applied to the support vector machine to achieve a high fit between the model prediction and the actual interrupter vacuum value,which provides the vacuum degree actual measurement data for the next chapter to realize the test and optimization of the vacuum life prediction model.(5)The vacuum life prediction model based on GA-SVM algorithm proposed in Chapter 4 is tested and optimized with the measured data of the vacuum level detection system based on the pulsed magnetron discharge method,and the superiority of the prediction algorithm proposed in this paper is verified by comparing different kernel functions and prediction algorithms.Finally,the practical value of the prediction algorithm is confirmed by the actual measurement data of a power plant.