摘要
针对10 kV配电网网架结构复杂、故障频率高的现状,为提高配电网抢修效率和全面提升服务质量,首先,以武汉市汉口地区为例,对该地区2015—2016年的故障工单数据进行统计分析,并绘制故障热力图;其次,运用Pearson相关分析法分析温度、雾霾、风力、雨量等因素对配电网故障的影响;最后,建立基于最小二乘和支持向量机(SVM)的配电网故障量预测模型,该模型成功预测了2017上半年汉口地区的故障量并得到验证。结果表明,该模型融合内外部数据,可对影响故障量的各因素进行相关性分析,将预测精度细化至街道办事处,为优化抢修驻点设置和提升抢修服务水平提供新方案。
Aiming at the current situation of complex network structure and high failure frequency of 10 kV distribution network,in order to improve the efficiency of distribution network emergency repair and to improve the quality of service in every aspect,first of all,a statistical analysis of the fault work data in Hankou area of Wuhan City,2015—2016 has been conducted and the fault heat diagram has been drawn.Then,it analyzes the influence of factors like temperature,haze,wind force,rainfall on distribution network fault by Pearson correlation analysis.At last,the fault prediction model of distribution network based on least square method and support vector machines(SVM)has been established,and the fault volume in Hankou area was predicted and verified in the first half of 2017.The results show that this model can combine the internal and external data,analyze the correlation of the factors that affect the amount of failure,refine the prediction accuracy to the street office,and provide a new scheme for setting up emergency repair site and improving the service level of emergency repair.
作者
高小芹
王梓玮
卢晨
伍茜
肖诗弋
GAO Xiaoqin;WANG Ziwei;LU Chen;WU Xi;XIAO Shiyi(Wuhan Poweo Supply Company,State Grid Hubei Electric Poweo Co.Ltd.,Wuhan 430014,Chona;不详)
出处
《武汉理工大学学报（信息与管理工程版）》
CAS
2020年第1期23-29,共7页
Journal of Wuhan University of Technology：Information & Management Engineering
基金
国家自然科学基金项目(71372201)。
关键词
故障量预测
最小二乘
支持向量机
Pearson相关分析
配电网
failure quantity prediction
least square method
Support Vector Machine(SVM)
Pearson correlation analysis
distribution network
