Deep learning power distribution network fault classification detection method and system based on attention mechanism

By using deep learning methods based on attention mechanisms, and leveraging the LSTM-Attention model and unsupervised learning, fault labels are automatically generated and data weights are dynamically assigned. This solves the problems of low solution efficiency and poor portability in traditional methods, and achieves efficient and accurate fault detection in power distribution networks.

CN119179970BActive Publication Date: 2026-06-09HUAZHONG UNIV OF SCI & TECH
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
HUAZHONG UNIV OF SCI & TECH
Filing Date
2024-08-28
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

In existing technologies, traditional distribution network fault detection methods have low solution efficiency in large-scale distribution network systems, making it difficult to provide answers within a reasonable time. Furthermore, their design relies on human experience, requiring redesign when the power grid topology changes, increasing costs and resulting in poor portability.

Method used

A deep learning method based on attention mechanism is adopted. By using the LSTM-Attention model, combined with unsupervised learning and clustering algorithms, fault labels are automatically generated, data weights are dynamically assigned, and the temporal features and key information of the power grid operation status are extracted for fault detection.

Benefits of technology

It improves the accuracy and real-time performance of fault detection, reduces costs, enhances the portability and adaptability of the model, and enables efficient fault diagnosis in different scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119179970B_ABST
    Figure CN119179970B_ABST
Patent Text Reader

Abstract

The application belongs to the technical field of power grid fault classification detection, and discloses a deep learning power distribution network fault classification detection method based on an attention mechanism, which is based on real-time monitoring technology of power distribution network power data, combined with a computer deep learning algorithm model, and through the attention mechanism, useless information is inhibited as much as possible, important data is highlighted, and the accuracy of fault detection is effectively improved.The method uses the attention mechanism and performs excellently in processing time series data, can effectively capture the dependency relationship and change trend in the time series, is very helpful for analyzing the change of the power grid state before and after the fault occurs, and is very helpful for accurate fault positioning and diagnosis.In different data sets and application scenarios, it has higher generalization ability.Compared with the traditional mathematical model method, it needs to be adjusted and calibrated in new scenarios.The algorithm model has high portability, is more convenient for fault diagnosis in different scenarios, reduces cost and improves efficiency.
Need to check novelty before this filing date? Find Prior Art