基于多维度电力大数据的高电压企业用户信用评估模型  

Customer Credit Evaluation Model of High-voltage Enterprises Based on Multi Dimensional Power Big Data

作　　者：王文烁 李天赐 WANG Wenshuo;LI Tianci(Electric Power Research Institute,State Grid Hubei Electric Power Co.,Ltd.,Wuhan 430000,China;Haikou Power Supply Bureau,Southern Power Grid Hainan Power Grid Co.,Ltd.,Haikou 570100,China)

机构地区：[1]国网湖北电力科学研究院,湖北武汉430000 [2]南网海南省海口供电局,海南海口570100

出　　处：《电工技术》2023年第12期190-194,197,共6页Electric Engineering

摘　　要：为实现不同信用等级高电压企业用户的差异化营销和电价风险识别,提高电力企业的风险管控水平,提出了基于多维度电力大数据的高电压企业用户信用评估模型。选取不同的原始凝聚点对企业用户的电网数据进行聚类分析;采用数据清洗、数据集成、数据规范和数据转换4个步骤对电网数据进行预处理;通过计算用户用电特征的信息增益,剔除错误数据,得到准确的用户用电特征;以守信能力为基础,建立高电压企业用户信用评价体系,设计基于支付积极性的高电压企业用户信用评价模型;计算企业加权定期支付周期值,建立信用价值权衡函数,完成对高电压企业用户支付信用的定量评价。实例测试结果表明,该方法能够准确获取高电压企业用户的用电特征,评价结果的准确性较高,为电网公司规避支付风险提供了数据依据。In order to realize the differentiated marketing and price risk identification of high-voltage enterprise users with different credit levels,and improve the risk management and control level of power enterprises,a credit evaluation model of high-voltage enterprise users based on multi-dimensional power big data is proposed.Different original condensation points were selected to cluster and analyze the power grid data of enterprise users.Four steps were used to preprocess the power grid data:data cleaning,data integration,data specification and data transformation.By calculating the information gain of the user's power consumption characteristics and eliminating the erroneous data,the accurate user power consumption characteristics were obtained.Based on the ability to keep promises,a credit evaluation system for high-voltage enterprise users was established,and a credit evaluation model for high-voltage enterprise users based on payment enthusiasm was designed.The weighted periodic payment cycle value of enterprises was calculated,the credit value trade-off function was established,and the quantitative evaluation of the payment credit of high-voltage enterprise users was completed.The case test results showed that the proposed method could accurately obtain the electricity consumption characteristics of high-voltage enterprise users,and the accuracy of the evaluation results was high,which provided a data basis for power grid companies to avoid payment risks.

关 键 词：多维度 电力大数据 高电压企业用户 信用评估 评估体系 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]