Application of Deep Learning in Intelligentization of Power System Vulnerability Knowledge Graph

Abstract

With the expansion of power grid, the amount of knowledge in power system is exploding. In order to organize, manage and utilize massive knowledge effectively, knowledge graph technology is introduced into the field of power system. In order to further improve the application of knowledge graph technology in power monitoring system. Firstly, this paper analyzes the power system knowledge graph and its advantages in power system knowledge management. Then, the construction method of power system knowledge graph is designed, focusing on the comprehensive analysis of the relationship between defect level and risk level, and drawing heat map. Combined with the characteristics of power system knowledge graph, the typical application scenarios of knowledge graph technology in the field of power system can be intelligently expanded by applying the mature and stable graph database method in the industry and adding the deep learning Convolutional Neural Network (CNN) method innovatively. Finally, on the basis of analyzing the current research hot-spots, the key problems in the application of knowledge graph in power system and the possible research directions in the future are pointed out.