基于大数据分析的客户电费风险预测及防控
|
摘要
客户既是企业最大的财富,也是风险的最大来源,拖欠电费、违约用电等行为时有发生。在交易过程中对客户信誉信息收集调查和风险评估,制定针对性的防范和服务策略,对企业具有非常重要的作用。为了降低电费回收风险,提高企业经营效益和管理手段,需要对客户进行评价分级,电力企业可以利用现有资源和数据,建立客户电费信誉信息档案,为客户提供差异化的服务,并规避风险。以营销业务应用系统中客户的海量历史用电数据为依据,再借鉴客户经理、催费员等电力人员工作经验,梳理风险用户的业务规则,并构建多维度的电费风险分析指标体系。通过机器学习的大数据分析方式对用电行为与电费风险之前的潜在关系进行研究,实现对风险用户的精准定位。
Customers are not only the greatest wealth of enterprises,but also the biggest source of risk. Default of electricity charges,breach of contract and other behaviors occur from time to time. In the course of transaction,it is very important for enterprises to collect and investigate customer reputation information,to make risk assessment,and to formulate targeted prevention and service strategies. In order to reduce the risk of electricity tariff recovery and improve the business efficiency and management means of enterprises,customers need to be evaluated and graded. Electric power enterprises can use existing resources and data to establish customer credit information files,provide customers with differentiated services and avoid risks. Based on the huge amount of historical electricity data of customers in the marketing business application system,then draw lessons from the work experience of electric personnel such as customer managers and fee collectors,comb out the business rules of risk users,and build a multi-dimensional electricity risk analysis index system. The potential relationship between power consumption behavior and electricity risk was studied through the data analysis method of machine learning,and the precise location of risk users was realized.
Customers are not only the greatest wealth of enterprises,but also the biggest source of risk. Default of electricity charges,breach of contract and other behaviors occur from time to time. In the course of transaction,it is very important for enterprises to collect and investigate customer reputation information,to make risk assessment,and to formulate targeted prevention and service strategies. In order to reduce the risk of electricity tariff recovery and improve the business efficiency and management means of enterprises,customers need to be evaluated and graded. Electric power enterprises can use existing resources and data to establish customer credit information files,provide customers with differentiated services and avoid risks. Based on the huge amount of historical electricity data of customers in the marketing business application system,then draw lessons from the work experience of electric personnel such as customer managers and fee collectors,comb out the business rules of risk users,and build a multi-dimensional electricity risk analysis index system. The potential relationship between power consumption behavior and electricity risk was studied through the data analysis method of machine learning,and the precise location of risk users was realized.
引文
[1]王宇哲,雷霞,陈晓盛,等.基于BP神经网络电力大客户信用等级评价研究[J].电力需求侧管理,2015,17(05):49-53.WANG Yuzhe,LEI Xia,CHEN Xiaosheng,et al. Large powercustomer credit rating based on BP neural network[J]. PowerDemand Side Management,2015,17(05):49-53.(in Chinese)
[2]陈延超.供电公司大客户信用评价及电费风险管理研究[D].北京:华北电力大学,2015.
[3]苏宏帮,吕静梅,王贞,等.电费关键指标监控系统的开发应用[J].青海电力,2014,33(01):11-13.SU Hongbang,LV Jingmei,WANG Zhen,et al. Development andapplication of electricity rate key indicator monitoring system[J].Qinghai Electric Power,2014,33(01):11-13.
[4]李力,林懿伦,曹东璞,等.平行学习—机器学习的一个新型理论框架[J].自动化学报,2017,43(01):1-8.LILi,LINYilun,CAODongpu,et al. Parallel learning-a newframework for machine learning[J]. Acta AutomaticaSinica,2017,43(01):1-8.
[5]张宜浩,金澎,孙锐,等.基于改进K-means算法的中文词义归纳[J].计算机应用,2012,32(05):1332-1334.ZHANG Yihao,JIN Peng,SUN Yue,et al. Chinese wordsenseinduction based on improved K-means algorithm〔J〕. Journal ofComputer Applications,2012,32(05):1332-1334.
[6]曲朝阳,陈帅,杨帆,等.基于云计算技术的电力大数据预处理属性约简方法[J].电力系统自动化,2014,38(08):67-71.QU Chaoyang,CHEN Shuai,YANG Fan,et al. An attributereducing method for electric power big data preprocessing based oncloud computing technology[J]. Automation of Electric PowerSystem,2014,38(08):67-71.
[7]宋亚奇,周国亮,朱永利,等.智能电网大数据处理技术现状与挑战[J].电网技术,2013,37(04):927-935.SONG Yaqi,ZHOU Guoliang,ZHU Yongli,et al. Presentstatus andchallenges of big data processing in smart grid[J]. Power SystemTechnology,2013,37(04):927-935.
[8]武依明.分类型调查数据的插补效果研究[D].石家庄:河北经贸大学,2016.
[9]俞立平,李守伟.标准特征因子的特点与应用分析[J].中国科技期刊研究,2016,27(09):990-993.YU Liping,LI Shouwei. The characateristic of the normalizedeigenfactor and its usage[J]. Chinese Journal of scientific andtechnological periodicals,2016,27(09):990-993.
[10]陈云,石松,潘彦,等.基于SVM混合集成的信用风险评估模型[J].计算机工程与应用,2016,52(04):115-120.CHEN Yun,SHI Song,PAN Yan,et al. Hybrid ensembleapproach for credit risk assessment based on SVM[J]. ComputerEngineering and Applications,2016,52(04):115-120.(inChinese)
[11]匡立伟.基于张量的大数据统一表示及降维方法研究[D].武汉:华中科技大学,2016.
[12]曹军威,袁仲达,明阳阳,等.能源互联网大数据分析技术综述[J].南方电网技术,2015,9(11):1-12.CAO Junwei,YUAN Zhongda,MING Yangyang,et al. Survey ofbig data analysis technology for energy Internet[J]. SouthernPower System Technology,2015,9(11):1-12.
[13]邓松,岳东,朱力鹏,等.电力大数据智能化高效分析挖掘技术框架[J].电子测量与仪器学报,2016,30(11):1679-1686.DENG Song,YUE yang,ZHU Lipeng,et al. Framework ofintelligent and efficient analysis and mining technology for powerbig data[J]. Journal Of Electronic Measurement andInstrumentation,2016,30(11):1679-1686.
[14]张东霞,苗新,刘丽平,等.智能电网大数据技术发展研究[J].中国电机工程学报,2015,35(01):2-12.ZHANG Dongxia,MIAO Xin,LIU Liping,et al. Research ondevelopment strategy for smart grid big data[J]. Proceedings ofthe CSEE,2015,35(01):2-12.
[15]彭小圣,邓迪元,程时杰,等.面向智能电网应用的电力大数据关键技术[J].中国电机工程学报,2015,35(03):503-511.PENG Xiaosheng,DENG Diyuan,CHENG Shijie,et al. Keytechnologies of electric power big data and its applicationprospects in smart grid[J]. Proceedings of the CSEE,2015,35(03):503-511.
[16] LIN W Y,HU Y H,TSAI C F. Machine learning in financial crisisprediction:a survey[J]. Systems Man&Cybernetics Part CApplications&Reviews IEEE Transactions on,2012,42(04):421-436.
[17]刘建,赵加奎,杨维,等.电力95598客户服务质检抽样算法研究[J].电网技术,2015,39(11):3163-3168.LIU Jian,ZHAO Jiakui,YANG Wei,et al. Research on qualityinspection sampling algorithm of electric 95598 customer service[J]. Power System Technology,2015,39(11):3163-3168.
[18]卢建昌,樊围国.大数据时代下数据挖掘技术在电力企业中的应用[J].广东电力,2014,27(09):88-93.LU Jianchang,FAN Weiguo. Application of data miningtechnologyin electric power enterprisses in era of big data[J]. GuangdongElectric Power,2014,27(09):88-93.
[19]刘玉军.包头地区电费收缴方式及其发展趋势分析[J].内蒙古电力技术,2016,34(04):40-42+46.LIU Yujun. Analysis on development trend of electricity chargecollection pattern in baotou area[J]. Inner Mongolia ElectricPower,2016,34(04):40-42+46.
[20]刘彪,刘金长.基于用户画像分析预测电费敏感型客户的建模实践[J].电力大数据2017,20(08):20-24.LIU Biao, LIU Jinchang. Modeling practice of forecastingelectricity fees sensitive customers based on user profile analysis[J]. Power systems and big data. 2017,20(08):20-24.
[2]陈延超.供电公司大客户信用评价及电费风险管理研究[D].北京:华北电力大学,2015.
[3]苏宏帮,吕静梅,王贞,等.电费关键指标监控系统的开发应用[J].青海电力,2014,33(01):11-13.SU Hongbang,LV Jingmei,WANG Zhen,et al. Development andapplication of electricity rate key indicator monitoring system[J].Qinghai Electric Power,2014,33(01):11-13.
[4]李力,林懿伦,曹东璞,等.平行学习—机器学习的一个新型理论框架[J].自动化学报,2017,43(01):1-8.LILi,LINYilun,CAODongpu,et al. Parallel learning-a newframework for machine learning[J]. Acta AutomaticaSinica,2017,43(01):1-8.
[5]张宜浩,金澎,孙锐,等.基于改进K-means算法的中文词义归纳[J].计算机应用,2012,32(05):1332-1334.ZHANG Yihao,JIN Peng,SUN Yue,et al. Chinese wordsenseinduction based on improved K-means algorithm〔J〕. Journal ofComputer Applications,2012,32(05):1332-1334.
[6]曲朝阳,陈帅,杨帆,等.基于云计算技术的电力大数据预处理属性约简方法[J].电力系统自动化,2014,38(08):67-71.QU Chaoyang,CHEN Shuai,YANG Fan,et al. An attributereducing method for electric power big data preprocessing based oncloud computing technology[J]. Automation of Electric PowerSystem,2014,38(08):67-71.
[7]宋亚奇,周国亮,朱永利,等.智能电网大数据处理技术现状与挑战[J].电网技术,2013,37(04):927-935.SONG Yaqi,ZHOU Guoliang,ZHU Yongli,et al. Presentstatus andchallenges of big data processing in smart grid[J]. Power SystemTechnology,2013,37(04):927-935.
[8]武依明.分类型调查数据的插补效果研究[D].石家庄:河北经贸大学,2016.
[9]俞立平,李守伟.标准特征因子的特点与应用分析[J].中国科技期刊研究,2016,27(09):990-993.YU Liping,LI Shouwei. The characateristic of the normalizedeigenfactor and its usage[J]. Chinese Journal of scientific andtechnological periodicals,2016,27(09):990-993.
[10]陈云,石松,潘彦,等.基于SVM混合集成的信用风险评估模型[J].计算机工程与应用,2016,52(04):115-120.CHEN Yun,SHI Song,PAN Yan,et al. Hybrid ensembleapproach for credit risk assessment based on SVM[J]. ComputerEngineering and Applications,2016,52(04):115-120.(inChinese)
[11]匡立伟.基于张量的大数据统一表示及降维方法研究[D].武汉:华中科技大学,2016.
[12]曹军威,袁仲达,明阳阳,等.能源互联网大数据分析技术综述[J].南方电网技术,2015,9(11):1-12.CAO Junwei,YUAN Zhongda,MING Yangyang,et al. Survey ofbig data analysis technology for energy Internet[J]. SouthernPower System Technology,2015,9(11):1-12.
[13]邓松,岳东,朱力鹏,等.电力大数据智能化高效分析挖掘技术框架[J].电子测量与仪器学报,2016,30(11):1679-1686.DENG Song,YUE yang,ZHU Lipeng,et al. Framework ofintelligent and efficient analysis and mining technology for powerbig data[J]. Journal Of Electronic Measurement andInstrumentation,2016,30(11):1679-1686.
[14]张东霞,苗新,刘丽平,等.智能电网大数据技术发展研究[J].中国电机工程学报,2015,35(01):2-12.ZHANG Dongxia,MIAO Xin,LIU Liping,et al. Research ondevelopment strategy for smart grid big data[J]. Proceedings ofthe CSEE,2015,35(01):2-12.
[15]彭小圣,邓迪元,程时杰,等.面向智能电网应用的电力大数据关键技术[J].中国电机工程学报,2015,35(03):503-511.PENG Xiaosheng,DENG Diyuan,CHENG Shijie,et al. Keytechnologies of electric power big data and its applicationprospects in smart grid[J]. Proceedings of the CSEE,2015,35(03):503-511.
[16] LIN W Y,HU Y H,TSAI C F. Machine learning in financial crisisprediction:a survey[J]. Systems Man&Cybernetics Part CApplications&Reviews IEEE Transactions on,2012,42(04):421-436.
[17]刘建,赵加奎,杨维,等.电力95598客户服务质检抽样算法研究[J].电网技术,2015,39(11):3163-3168.LIU Jian,ZHAO Jiakui,YANG Wei,et al. Research on qualityinspection sampling algorithm of electric 95598 customer service[J]. Power System Technology,2015,39(11):3163-3168.
[18]卢建昌,樊围国.大数据时代下数据挖掘技术在电力企业中的应用[J].广东电力,2014,27(09):88-93.LU Jianchang,FAN Weiguo. Application of data miningtechnologyin electric power enterprisses in era of big data[J]. GuangdongElectric Power,2014,27(09):88-93.
[19]刘玉军.包头地区电费收缴方式及其发展趋势分析[J].内蒙古电力技术,2016,34(04):40-42+46.LIU Yujun. Analysis on development trend of electricity chargecollection pattern in baotou area[J]. Inner Mongolia ElectricPower,2016,34(04):40-42+46.
[20]刘彪,刘金长.基于用户画像分析预测电费敏感型客户的建模实践[J].电力大数据2017,20(08):20-24.LIU Biao, LIU Jinchang. Modeling practice of forecastingelectricity fees sensitive customers based on user profile analysis[J]. Power systems and big data. 2017,20(08):20-24.