基于数据挖掘的主配网一体化协同监视方法  

Integrated collaborative monitoring method of main distribution network based on data mining algorithm

作　　者：吕磊炎 杜浩良 钱肖 马翔 崔建业 王宁 LV Lei-yan;DU Hao-liang;QIAN Xiao;MA Xiang;CUI Jian-ye;WANG Ning(State Grid Jinhua Power Supply Company,Jinhua 321000,Zhejiang Province,China)

机构地区：[1]国网浙江省电力有限公司金华供电公司,浙江金华321000

出　　处：《信息技术》2023年第4期173-178,共6页Information Technology

摘　　要：为了实现主网和配网之间线路运行状态以及相关设备运行状态的协同决策、协同监管,研究基于数据挖掘的主配网一体化协同监视方法。将采集的数据进行格式统一化处理,传送至数据管理层,数据管理层完成数据清洗及校验后,分类存储;协同监视层通过主配网历史数据模型、实时估计以及态势预判感知存储数据,采用随机森林算法构建协同决策规则库,将感知结果与协同决策规则库进行匹配,获取故障控制决策。测试结果表明,该方法的感知性能良好,可快速完成故障和控制策略的匹配,满足主配网一体化协同监视需求。In order to realize the collaborative decision-making and collaborative supervision of the line operation status between the main network and the distribution network,as well as the operation status of related equipment,an integrated collaborative monitoring method of the main distribution network based on data mining is studied.The collected data is processed in a unified format and transmitted to the data management layer.After the data management layer completes the data cleaning and verification,it would be classified and stored.The collaborative monitoring layer stores the perceived data through the historical data model of the main distribution network,real-time estimation and situational prediction awareness.Besides,the random forest algorithm is adopted to build the collaborative decision rule base to match the perceived results with the collaborative decision rule base,which could obtain fault control decisions.The test results show that this method has good perceptual performance,and could quickly complete the fault and its control strategy matching,which meets the requirements of integrated collaborative monitoring of main distribution network.

关 键 词：数据挖掘算法 主配网 协同监视 态势感知 

分 类 号：TM721[电气工程—电力系统及自动化]