Research On Fault Monitoring And Diagnosis Method For Switchgear Based On PHM Technology

As the critical equipment for electrical lines opening and closing, electrical equipment controlling and protecting, SwitchGear is widely used in power generation, transmission and distribution. It contributes lot to the reliability of power supply. If there is no warning before a fault occurs, it will cause mass loss in manpower and resources, and it causes inconvenient for power recover. Therefore, it is very important to realize the real-time monitoring, fault prediction and diagnosis of switchgear. Combining with the actual demand of Qunli Electric Co., Ltd., Zhejiang, this paper develops an intelligent monitoring system platform, which mainly has switchgear fault monitoring and diagnosis. The main contents are as follows:① Analyze the common type of Switchgear, and summarize the fault characteristic quantities to constitute the original features. According to sub-chamber structure and original features, determine the real-time monitoring program of fault characteristics, and design the monitoring structure and monitoring points of fault characteristic quantities.② This paper develops an intelligent fault monitoring and diagnosis method of switchgear based on the Prognostics and Health Management, and uses the principle of minimum redundancy maximum correlation to realize the dimension reduction of original features, and obtain the optimal feature subset; quantify the similarity of fault characteristic and standard samples by Mahalanobis distance, and determine the threshold of Mahalanobis distance to achieve the switchgear fault diagnosis.③ Design and develop an intelligent status monitoring and fault diagnosis of switchgear system platform. The Hardware is used for the acquisition of fault characteristics, real-time display, data storage and recall capabilities; The software is used to process and analysis the fault characteristics, display the real-time status, judge the fault auxiliary source, achieve the switchgear fault diagnosis by a algorithm, and provide the maintenance recommendations.