Line external damage prevention early warning method and system based on network fuzzy neural algorithm

By fusing multimodal data using a network fuzzy neural algorithm to obtain the confidence level of external damage categories and risk assessment weights, the problem of insufficient accuracy and precision in power line external damage early warning is solved, and efficient hierarchical early warning is achieved.

CN122416628APending Publication Date: 2026-07-17STATE GRID ZHEJIANG ELECTRIC POWER COMPANY TAIZHOU POWER SUPPLY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
STATE GRID ZHEJIANG ELECTRIC POWER COMPANY TAIZHOU POWER SUPPLY
Filing Date
2026-03-25
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies for early warning of external damage to power lines lack accuracy in determining early warning events and in determining risk scores, making it difficult to achieve reliable and effective tiered early warning.

Method used

An early warning method based on network fuzzy neural algorithm is adopted. The confidence level of external damage category is obtained by multimodal data fusion, and risk assessment is carried out by combining short-term energy and spatial and temporal influence weights. This includes the application of knowledge distillation and network fuzzy neural algorithm.

Benefits of technology

It improves the accuracy of early warning event identification and risk scoring, reduces the deployment difficulty of lightweight external damage classification models, and achieves reliable hierarchical early warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122416628A_ABST
    Figure CN122416628A_ABST
Patent Text Reader

Abstract

The application discloses a line external damage early warning method and system based on a network fuzzy neural algorithm, and belongs to the technical field of power line safety protection, and comprises the following steps: obtaining an external damage classification model based on knowledge distillation, performing feature coding and weighted fusion on a plurality of groups of collected multi-modal data through the external damage classification model to obtain external damage category confidence, extracting short-time energy of vibration data in the plurality of groups of multi-modal data, obtaining an early warning event according to the short-time energy and the external damage category confidence, fusing the multi-modal data corresponding to the early warning event and the external damage category confidence to obtain a multi-dimensional feature vector, inputting the multi-dimensional feature vector into a network fuzzy neural algorithm to obtain spatial influence weight and time influence weight of risk assessment, and obtaining a risk score according to the spatial influence weight and the time influence weight, and performing graded early warning according to the risk score. The technical problem that the prior art is difficult to improve the accuracy of early warning event determination and the accuracy of early warning event risk score is solved.
Need to check novelty before this filing date? Find Prior Art