Quality improvement method for poor-quality power grid equipment defect text

Abstract

The invention provides a quality improvement method for a poor-quality power grid equipment defect text. The method comprises the following steps: firstly, correcting a text with poor quality in historical defect texts by utilizing a latent Dirichlet allocation model in Chinese text similarity calculation, and combining with the power transmission and transformation primary equipment defect classification standard of the State Grid Corporation of China for improving quality; then, giving a quality problem prompt for a newly input text by utilizing a text quality detection method and giving a correction advice for the newly input text by utilizing a word vector mapping method for ensuring the quality of the newly input defect text; and finally, carrying out quality comparison on the defecttext before correction and after correction by combining with a living example and classifying the defect text before correction and after correction by utilizing classification methods in machine learning and deep learning according to defect levels for verifying the effectiveness of a method for improving the quality of the text with the poor quality. By use of the method, from the origin, the defect text is specified, the quality of the defect text is guaranteed, and reliable and accurate text data is provided for defect text mining.

Description

technical field [0001] The invention belongs to the field of power systems, in particular to a method for improving the quality of defect texts of grid equipment with poor quality. Background technique [0002] With the in-depth advancement of smart grid construction, a large amount of multi-source heterogeneous data has been generated in various links of the power system, and unstructured data represented by text, audio, and images has grown the fastest. Among them, the text describing the defects of power grid equipment contains the information most closely related to the safety of equipment and power grids, and has attracted the attention of technical and management personnel. Due to manual classification and statistics of defect texts, the workload is large and the efficiency is low, and the results depend on artificial subjective experience. How to improve the efficiency of defect text mining is a problem that needs to be solved. [0003] At present, natural language p...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention is to propose a method for improving the quality of power grid equipment defect

Images