Power grid defect detection method based on entity recognition, electronic equipment and program product

By extending the sample data of the grid entity labeling standard and generating synthetic text for entity labeling, the problem of low accuracy of the grid entity recognition model in the absence of training data is solved, and the accuracy of knowledge graph and grid equipment defect detection is improved.

CN119691154AActive Publication Date: 2025-03-25STATE GRID SHANXI ELECTRIC POWER COMPANY TAIYUAN POWER SUPPLY COMPANY
5 Cites 1 Cited by

Patent Information

Application Number
CN202510196303.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-02-21
Publication Date
2025-03-25
Estimated Expiration
2045-02-21

AI Technical Summary

Technical Problem

The grid entity recognition requires a large amount of labeled data training. In the absence of training data, there is a problem that the entity recognition model recognizes entities is low, resulting in low accuracy of the knowledge graph, which leads to low accuracy of the grid equipment defect detection accuracy.

Method used

By extending the sample data of the entity annotation standard, the required amount of training data is quickly obtained, and the entity recognition model is trained based on this, the sample data is expanded using a large language model, and synthetic text is generated for entity annotation, and training data for training entity recognition model is generated.

Benefits of technology

Effectively train the entity model based on a small amount of accurate standard sample data, ensuring the training efficiency and accuracy of entity recognition model, thereby ensuring the accuracy of creating knowledge graphs based on entities, and improving the accuracy of power grid equipment defect detection based on knowledge graphs.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of power grid detection, and particularly provides a power grid defect detection method based on entity recognition, electronic equipment and a program product. The method comprises the steps of obtaining power grid data; the power grid data are input into a trained entity recognition model, the entity recognition model outputs a corresponding entity, and the entity recognition model is obtained by obtaining training data after sample data are expanded through a large language model and training the entity recognition model through the training data; creating a corresponding power grid equipment defect knowledge graph according to the entity; and detecting the power grid data according to the knowledge graph, and determining whether power grid equipment defects exist or not. The method and the device are used for solving the problems that power grid entity recognition in related technologies needs to depend on a large amount of annotation data training, under the condition of lack of training data, the accuracy of entity recognition of an entity recognition model is low, the accuracy of a knowledge graph is low, and then the defect detection accuracy of power grid equipment is low.
Need to check novelty before this filing date? Find Prior Art