Research On Vacuum Lifetime Assessment Technology Based On Pulse Magnetic Control Discharge Method

Vacuum switch is an important control and protection equipment in the power system, its working reliability is a key role to the safe operation of power grids. The vacuum degree of vacuum interrupter is a main indicator of its performance. However, there is a limited vacuum life problem, which is caused by vacuum degradation along with the increase of operation and storage time of vacuum interrupter. At present, the research on vacuum degree is mainly focused on the improvement of detection method, less considering the comprehensive evaluation of vacuum detection and life prediction of vacuum, and the existing of vacuum life prediction methods are rare. Therefore, the subject which considering the vacuum degree detection and the limited life of vacuum at the same time, especially in the background of the CBM(condition Based Maintenance), is of great importance in theory and practice.This paper firstly studies on the methods of vacuum detection at home and abroad, including offline testing and online detection, and comes to a conclusion through methods comparison:magnetic control discharge method is currently one of the best vacuum detection method which is high measurement precision and good repeatability test data characteristics; then does an in-depth research on the theory of the pulse magnetic control discharge method, and designs a set of vacuum degree detection system, which includes the pulse high voltage circuit module, the magnetic field circuit module, the field coil voltage control module without dismantling, signal acquisition detection module, driver module and the software design, etc; the research verifies each module function through a large number of experiments and the analysis of historical data, obtaining the relationship curve between vacuum degree and the ion current, creating a database of vacuum degree and realizing the intelligent vacuum measurement at last.In order to solve this problem that the numbers of vacuum life prediction methods are rare at home and abroad, this paper deeply studies the feasibility that the grey GM(1,1) model, BP neural network model and gray BP neural network combination model predict vacuum life, and constructs corresponding three kinds of new methods to vacuum lifetime prediction; selecting the GM(1,1) and BP neural network combined model as the method to the evaluation system of vacuum lifetime through prediction accuracy and comparative test. Finally realizing the vacuum trend prediction based on pulse magnetic control discharge method and vacuum life prediction based on GM(1,1) and BP neural network combined model, providing a new method for the research of vacuum degree testing and life assessment.