In order to meet the requirements of power grid operation accident detection and plan generation,and to pro-vide a comprehensive plan knowledge graph and achieve simple and effective accident handling methods,an intelligent gener-ation algorithm of power grid operation accident plan knowledge graph based on abnormal characteristics and power flow cal-culation model is proposed.The abnormal features of power grid operation including volatility and variability are extracted,and the optimal feature combination selection is achieved through the abnormal feature analysis of relevance vector machine,and then the power grid operation accidents are detected with convolutional neural network.After detecting the occurrence of operational faults,a power flow calculation model is established,and the node power is calculated based on the current infor-mation between nodes.Based on this,a power grid operation accident recovery model is constructed with the goal of priori-tizing the recovery of important loads and minimizing the number of switch operations.The optimal accident recovery and disposal plan is obtained through particle swarm optimization algorithm solving.The entity database and ontology model of the accident plan are established,and entities and relationships are extracted through the ontology model to generate the knowledge graph of the power grid operation accident plan to assist in the decision-making of power grid operation accidents.The experimental results show that the algorithm can detect power grid operation accidents,and the generated knowledge graph of power grid operation accident plan contains extensive content,and the accident handling method is simple and effec-tive.
国家科技期刊平台
NSTL
万方数据
