基于SPSS Modeler的高损耗线路模式识别体系的研究  

SPSS Modeler Based Pattern Recognition System for High-loss Line

作　　者：李坤 邵方冰 张瑞曦 王恩 李博 杨振明 唐标 杨成涛 LI Kun;SHAO Fangbing;ZHANG Ruixi;WANG En;LI Bo;YANG Zhenming;TANG Biao;YANG Chengtao(Ruili Power Supply Bureau,Yunnan Power Grid Co.,Ltd.,Ruili 678400,China;School of Electrical Engineering,Southeast University,Nanjing 210000,China;Yunnan Electric Power Research Institute,Yunnan Power Grid Co.,Ltd.,Kunming 650271,China)

机构地区：[1]云南电网有限责任公司瑞丽供电局,云南瑞丽678400 [2]东南大学电气工程学院,江苏南京210000 [3]云南电网有限责任公司电力科学研究院,云南昆明650217

出　　处：《智慧电力》2019年第11期92-96,103,共6页Smart Power

基　　金：国家自然科学基金资助项目(51877044)~~

摘　　要：在工程实践中,需要在计算条件成熟的线路上计算线损,分析线路的特征量,建立高损耗线路模式识别体系,从而对计算条件不成熟线路的线损率指标进行预测。从理论线损的计算方法出发,分析不同电压等级下的计算条件和适用的计算方法;提出了通过K-Means聚类算法对线路特征进行初步分类,在各个类中使用C&R分类回归树的方法,最终形成模式识别的判别树。分析结果表明,该方法能有效跟踪某一线路高损耗产生的原因,并对线损进行预测。In project practice,it is necessary to calculate line loss meeting its calculation requirement,analyze the characteristic of the line,and establishes the pattern recognition system of high-loss line to predict the indicators of the line loss which does not meet the calculation requirement.The calculation conditions and applicable calculation methods under different voltage levels are analyzed based on theoretical line loss calculation.The K-means clustering algorithm is proposed for the preliminary classification of the line features,and the classification and regression tree method is used to predict the line loss.The analysis results show that this method can effectively track the causes of high loss of a certain line and predict the line loss.

关 键 词：理论线损计算 模式识别 K-MEANS聚类算法 C&R分类回归树 线损预测 

分 类 号：TM711[电气工程—电力系统及自动化]