A GIS insulation defect early warning method, system, device and medium

By fusing multi-source heterogeneous data and using the Transformer neural network model, the problems of weak anti-interference ability and lack of hierarchical early warning in the partial discharge detection of GIS equipment are solved, realizing accurate identification and early warning of GIS insulation defects, and improving the accuracy of detection and the scientific nature of decision-making.

CN121878400BActive Publication Date: 2026-06-12HEFEI UNIV OF TECH
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Patents(China)
Current Assignee / Owner
HEFEI UNIV OF TECH
Filing Date
2026-03-18
Publication Date
2026-06-12

Smart Images

  • Figure CN121878400B_ABST
    Figure CN121878400B_ABST
Patent Text Reader

Abstract

The application discloses a GIS insulation defect early warning method, system, device and medium, the method comprises the following steps: collecting multi-source heterogeneous signals of GIS equipment, and constructing a multi-modal feature dataset after signal data preprocessing; a Transformer neural network model based on multi-modal fusion is constructed; the multi-modal feature dataset is input into the neural network model for training and real-time calculation of the loss function value based on a set of weight parameter optimization algorithms to determine the optimal parameters of the model and update the network structure parameters; electromagnetic wave spectrum data and real-time acoustic waveform data are collected by UHF sensors and ultrasonic sensors respectively, and after the multi-modal input data are constructed, the trained Transformer neural network model is input for deep reasoning, and a hierarchical warning operation is performed based on the reasoning result of the multi-modal data. The application can realize accurate identification and timely warning of GIS weak insulation defects.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • High-voltage switch cabinet partial discharge fault identification method

    CN121522447A

  • Partial discharge type identification method and device, computer equipment, readable storage medium and program product

    CN121561577A

  • Partial discharge on-line monitoring method and system based on multi-modal fusion and adaptive noise reduction

    CN121636975A