电费回收风险预测模型研究与实践  

Research and Practice on Risk Prediction Model of Electricity Tariff Recovery

作　　者：夏洪涛[1] 施永益[1] 凌卫家[1] 李玮玮[1] 李懑君 赵悦 赵燕波 

机构地区：[1]国网浙江省电力公司,浙江杭州310012 [2]北京国电通网络技术有限公司,北京100070 [3]国网浙江海盐县供电公司,浙江嘉兴314300

出　　处：《电力信息与通信技术》2016年第8期58-63,共6页Electric Power Information and Communication Technology

摘　　要：电费是电网公司的主要收入来源,电费的及时有效回收对电网公司的良性运营至关重要。为了提高电费回收率,文章将量化分析方法引入到电费回收管理工作中,建立了分类预测模型预测电费的回收风险。基于模型预测结果,对影响电费回收的关键因素进行识别,并为改善电费回收状况提出针对性建议。Electricity tariff is a main revenue source for the Power Grid Corp. The timely and effective recovery of electricity is essential to the operation of the Power Grid Corp. To improve the recovery rate, this paper introduces the quantitative analysis method into the collection and management of electricity charges, and establishes the classification prediction model to forecast the risk of electricity collection. Based on the model prediction results, the key factors affecting the recovery of electricity tariff are identified, and the suggestions are put forward to improve the status of the recovery of electricity tariff.

关 键 词：电费回收 分类模型 风险预测 不平衡数据 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]