基于AHP-GRA分析与KNN算法的低压配电网拓扑校验  

作　　者：王爱军 王亚飞 李哲明 彭明生 吴廷勇 宋钰 

机构地区：[1]国网江苏省电力有限公司射阳县供电分公司,江苏盐城224300

出　　处：《科技创新与应用》2023年第18期31-36,共6页Technology Innovation and Application

基　　金：国家电网射阳县供电公司科技项目(2022SJYCZC153)。

摘　　要：针对低压配电网接线错综复杂、线路的拓扑关系以及特征呈现出多样化,且在对配电网进行人工拓扑结构校验时所需的成本较高、实时性不强,因此该文研究一种基于AHP-GRA分析与KNN算法的低压配电网拓扑校验方法。首先对一天内的时间段划分为凌晨、上午、下午、夜晚4个阶段,由于4个阶段的电压波动范围不同,因此通过AHP算法进行权值确定以反映其波动程度。然后,通过GRA分析法分别确定各个时间段的电压关联度,并通过权值与对应阶段的关联度相结合,从而确定用户之间的关联程度,由此找出异常用户,然后对异常用户进行再次校验以确定其所属台区,对于异常用户通过GIS系统的数据和《配电网规划设计技术导则》剪辑生成用户样本集,运用KNN算法对样本集进行分析,从而确定异常用户的所属台区。经过验证,该方法能够自动识别异常用户及其所属台区,可以有效降低出错率,具有一定的实用性。In view of the complexity of low-voltage distribution network wiring,the diversity of line topology and characteristics,the cost of manual topology verification of distribution network is high,and the timeliness is not strong.Therefore,this paper studies a low-voltage distribution network topology verification method based on AHP-GRA and KNN algorithm.First of all,the time period of the day is divided into four stages:early morning,morning,afternoon and night.Because the voltage fluctuation ranges of the four stages are different,the weights are determined by AHP algorithm to reflect the degree of fluctuation.Then the voltage correlation degree of each time period is determined by GRA analysis,and the correlation degree between users is determined by combining the weight with the correlation degree of the corresponding stage,so as to find out the abnormal users.Then the abnormal users are checked again to determine the station area to which they belong.For abnormal users,the user sample set is generated by editing the data of GIS system and"Technical Guidelines for Distribution Network Planning and Design",and the KNN algorithm is used to analyze the sample set to determine the station area of abnormal users.After verification,this method can automatically identify abnormal users and their stations,with low error rate and certain practicality.

关 键 词：低压拓扑校验 GRA分析 AHP算法 KNN算法 配电网 

分 类 号：TM71[电气工程—电力系统及自动化]