Construction of Knowledge Graph of Substation Main Equipment Based on LLM
Xiaomeng Li, Jingrong Zheng, Zhantao Su, Yuxin Chen, Yifeng Wang, Chen Chen, Yueling Zhao, Kun Wang · 2024
At present, the digital transformation and upgrading of power grid, and the rapid development of intelligent health management technology of power substation main equipment, there are problems of weak information correlation and low efficiency of decision generation in the operation and maintenance process. This paper designs the construction method of fault knowledge graph of substation main equipment, introduces the traditional construction method of knowledge graph and points out the difficulties, and puts forward the problem of using LLM to complete some tasks in knowledge graph construction to improve the cost and workload. A knowledge graph construction idea based on LLM is conceived (the steps include data cleaning, knowledge acquisition and knowledge refining), a knowledge graph of the main equipment of substation is constructed, and some query scenarios are demonstrated. The feasibility and efficiency of the new method is proved through internal testing and evaluation, and the possible application direction of the graph is prospeced. The system can realize the goal of intelligent fault diagnosis and decision making, so as to improve the efficiency of daily operation, maintenance and management of power grid.