GIS vibration signal characteristic quantity extraction method based on variational mode decomposition

Abstract

The invention discloses a GIS vibration signal characteristic quantity extraction method based on variational mode decomposition, and solves the problems that useful characteristic information cannotbe accurately obtained under the condition that GIS vibration signals are different in source and characteristics of the GIS vibration signals are different, and information loss caused by excessive decomposition exists. According to the method, a gray wolf predation algorithm is used for carrying out parameter optimization on a variational mode decomposition algorithm; the GIS mechanical vibration signal is decomposed by using an optimized variational mode decomposition algorithm to effectively extracting an optimal IMF component bearing rich feature information; the sum of the maximum amplitude, the root-mean-square value, the kurtosis, the 100Hz proportion, the 50Hz and odd frequency multiplication proportion thereof and the 100Hz frequency multiplication proportion thereof of the optimal IMF component are extracted as characteristic quantities, and a characteristic vector is constructed. By comparing the feature vectors in the normal state with the feature vectors in the fault state, GIS mechanical fault diagnosis is realized.

Description

technical field [0001] The invention relates to the technical field of GIS fault detection, in particular to a method for extracting characteristic quantities of GIS vibration signals based on variational mode decomposition. Background technique [0002] Sulfur hexafluoride closed combined electrical equipment, known internationally as "Gas Insulated Switchgear" (Gas Insulated Switchgear) or GIS for short, combines primary equipment in a substation other than transformers, including circuit breakers, disconnectors, grounding switches, and voltage mutual inductance Devices, current transformers, lightning arresters, busbars, cable terminals, inlet and outlet bushings, etc., are organically combined into a whole through optimized design. Sulfur hexafluoride fully-enclosed gas-insulated combined electrical appliances (GIS) have been successfully used in the world since its inception due to its small footprint, high reliability, strong safety, short installation period, and smal...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is: conventional empirical mode decomposition (EMD) and improved empirical mode decomposition (EEMD) methods have certain disadvantages when dealing with GIS vibration fault signals. The bandwidths are all uncertain, so if the frequency band carrying the fault signal is within the frequency band of the first decomposed component, it may introduce too much interference because the frequency band of the component is too wide. In the frequency band, important feature information may be missed because the component frequency band is too narrow, useful feature information cannot be accurately obtained, and there are problems such as information loss caused by over-decomposition. The present invention provides a variational model-based The GIS vibration signal feature extraction method based on state decomposition, using the gray wolf predation algorithm to optimize the parameters of the variational mode decomposition algorithm, using the optimized variational mode decomposition algorithm to reasonably decompose the GIS mechanical vibration signal, and effectively extract the The best IMF feature components that carry rich feature information and construct feature vectors

Images