Alternating current contactor life cycle segmentation method based on multivariate time series

Abstract

The invention discloses an alternating current contactor life cycle segmentation method based on a multivariate time sequence, and the method comprises the following steps: obtaining the test data of an alternating current contactor through an Alternating current contactor full-life test platform system, carrying out the data processing of the test data, and obtaining a plurality of characteristic parameters related to the life of the alternating current contactor; performing variable selection on the characteristic parameters through a Naive-RreliefF algorithm, and removing redundant characteristic parameters and irrelevant characteristic parameters to obtain optimal characteristic parameters; and segmenting a time sequence corresponding to the full life cycle of the alternating current contactor by combining the optimal characteristic parameter to obtain a time sequence segment containing the optimal characteristic parameter. By adopting a Relief-F-based Naive variable selection method and a Bellman optimization principle-based multivariate time sequence segmentation method, the method has good accuracy and robustness, gives consideration to the dynamic characteristics of the time sequence, has very important practical value, realizes effective segmentation of the life cycle of the alternating current contactor, and makes preparation for accurate evaluation of the full-life state.

Description

technical field [0001] The invention relates to the technical field of electric equipment detection, in particular to a method for dividing the life cycle of an AC contactor based on a multivariate time series. Background technique [0002] There is bound to be performance degradation of contact appliances during their use. The smart grid proposes that electrical products themselves should have the functions of perception, judgment and early warning, hoping to display effective status information during the process of electrical performance degradation. At present, through the in-depth analysis of electrical switch theory at home and abroad, the statistics of a large number of test data, and the study of the failure mechanism of switching electrical appliances will help to speed up the research process of intelligent electrical equipment. [0003] The contact will bounce due to mechanical vibration during the closing process. The bounce will cause a metal bridge and a short...

AI Technical Summary

Problems solved by technology

In the existing mathematical research methods for the life prediction model of switching electrical contacts, it is impossible to give clear characteristic parameters through the electrical contact theory, and it is also impossible to clearly point out the difference between the life states of adjacent contacts, and it is impossible to predict and compare the life stages. division, and it is impossible to further determine the weight of each characteristic parameter in the life degradation process

Images