A method for early warning of electric cable winding device failure

By constructing a fault judgment model for an electric cable retrieval device using Granger causality analysis and graph attention networks, the problem of failing to identify cooperative deviations in existing technologies is solved. This enables the identification of system-level cooperative state deviations of the electric cable retrieval device, improving the accuracy and timeliness of fault early warning.

CN122448573APending Publication Date: 2026-07-24武汉中天嘉业机电工程有限公司
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610547299.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-23
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing fault warning methods for electric cable winding devices cannot capture the coordinated deviation between modules, resulting in insufficient timeliness and accuracy of warnings, and an inability to identify progressive and coordinated faults.

Method used

A follow-up cooperative link is constructed using Granger causality analysis algorithm. Combined with graph attention network and spatial attention mechanism, a fault judgment model for electric cable winding device is constructed. By using causal analysis and graph attention network adaptive learning module importance, deviations in cooperative state are identified and fault warning results are generated.

Benefits of technology

It enables system-level collaborative state deviation identification of electric cable winding devices, improves the pertinence and reliability of fault diagnosis, reduces the risk of misjudgment, and significantly enhances the accuracy and timeliness of fault early warning.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122448573A_ABST
    Figure CN122448573A_ABST
Patent Text Reader

Abstract

The application discloses a kind of electric cable winding device fault early warning methods, it is related to fault early warning and processing technology.The method includes: collecting electric cable winding device state information, the cause and effect analysis is carried out to electric cable winding device state information, and constructs follow-up collaborative link;Based on follow-up collaborative link acquisition historical data, and construct qualified stage fingerprint set;Based on follow-up collaborative link acquisition real-time data, obtain current stage fingerprint;Build electric cable winding device fault judgment model, use qualified stage fingerprint set to train, input current stage fingerprint into trained electric cable winding device fault judgment model, output fault judgment result;Based on fault judgment result and follow-up collaborative link, generate potential failure chain, carry out double threshold determination to potential failure chain, obtain electric cable winding device fault early warning result.The application generates electric cable winding device fault early warning result by graph attention network, realizes the fault early warning and processing of electric cable winding device.
Need to check novelty before this filing date? Find Prior Art