Text Mining of Power Secondary Equipment Based on BiLSTM-Attention

A large number of idle defect and fault texts are accumulated in the process of long-term operation, maintenance and repair of power secondary equipment. In order to realize the effective utilization and deep mining of defect texts, a method of defect text classification of secondary equipment based on BiLSTM-Attention is proposed. Firstly, the text is cleaned according to the natural language characteristics of the historical defect texts of the secondary equipment. Secondly, the text content is vectorized through Word2Vec model, and the semantic information of the text is extracted in depth through BiLSTM-Attention to realize the accurate classification of the defect degree of the equipment. Finally, an example is built through the historical defect texts of a power company. The result shows that the proposed model can accurately classify the defect text of power secondary equipment and help operation and maintenance personnel to judge the defect degree quickly and accurately.