GIS partial discharge type identification method based on deep residual network

Abstract

The invention provides a GIS partial discharge type identification method based on a deep residual network, and the method comprises the following steps: setting an ultrahigh-frequency sensor at detected GIS electrical equipment, receiving and processing a discharge signal collected by the ultrahigh-frequency sensor through a partial discharge detection tool, and generating a PRPD spectrogram through a partial discharge analysis tool; summarizing the PRPD spectrograms of the tested GIS electrical equipment, classifying the PRPD spectrograms according to the characteristics of the partial discharge defect spectrograms, and setting labels to generate a training set; building a deep residual neural network classification model according to the size of the PRPD spectrogram in the training set,and determining model parameters; training the deep residual neural network classification model through the PRPD spectrogram of the training set and the corresponding category label, calculating thetest accuracy, and adjusting the weight of each model parameter according to the test accuracy; and judging the PRPD spectrogram of an unknown type by using the trained deep residual neural network classification model.

Description

technical field [0001] The invention belongs to the technical field of electric equipment insulation state evaluation, and in particular relates to a GIS partial discharge type identification method based on a deep residual network. Background technique [0002] GIS, the abbreviation of Gas Insulated Switchgear, gas insulated switchgear, is the primary equipment other than the transformer in the substation. [0003] PRPD is the abbreviation of Phase Resolved Partial Discharge. Phase resolved partial discharge is to display each partial discharge pulse with a phase mark according to the phase. The discharge information has no time information and belongs to the superposition of PRPS information within a period of time. [0004] GIS is widely used in power systems due to its advantages of good insulation, high reliability and small footprint. However, in the complex environment of high temperature and high pressure in the power system, or in the process of GIS manufacturing, ...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the above-mentioned preliminary processing of PRPD spectrograms in the prior art, some important information is ignored while focusing on extracting effective information, and the identification task cannot be completed objectively and comprehensively. The present invention provides a GIS partial discharge based on deep residual network. Type identification method to solve the above technical problems

Images