面向电力调度的事件知识图谱研究现状及发展

A Review of Research Methods on Event Knowledge Graph for Power Dispatching

事件知识图谱(EKG)是一种可学习事件演化规律的特殊知识图谱,具有推理、预测等功能。针对电力调度业务数据量大、模态多、交互耦合等特点,该文详述了面向电力调度的事件知识图谱的数据集构建、主流方法、技术架构、评价指标、适用场景等,重点分析各场景的可行性,并在应用流程、输入输出、技术架构等方面给出方案,最后对其在电力调度业务长期发展面临的难点和可能的研究方向进行了展望。该文研究为研究电力调度领域特点、事件知识图谱优势和两者结合提供了参考,并为事件知识图谱在电力调度领域中的应用方向提供了指导性思路。

Event Knowledge Graph(EKG)is a special knowledge graph that can learn the evolution laws of events,which has the functions of reasoning and prediction.In view of the characteristics of large amount of data,multiple modes and interactive coupling of power dispatching business,this paper describes in detail the dataset construction,mainstream methods,technical architecture,evaluation indexes,and applicable scenarios of the event knowledge graph for power dispatching,focuses on the feasibility of each scenario,and gives solutions in terms of application process,input and output,technical architecture,etc.,and finally looks forward to the difficulties and possible research directions faced in the long-term development of power dispatching business.This paper provides a reference for the study of the characteristics of the field of power dispatching,the advantages of event knowledge graph and the combination of the two,and provides a guiding idea for the application direction of event knowledge graph in the field of power dispatching.