电费回收风险预测的大数据方法应用

doi:10.11959/j.issn.1000-0801.2019040

Abstract

Abstract:

Based on the historical data of electricity customers,the model index system was determined according to the customers’ basic attributes,the electricity consumption and the payment behavior,the customers’ credit,the industry prospects’ information and so on.Through the correlation coefficient matrix and the information value of the index,the index variables that enter the model were selected.At the same time,the best grouping method was used to group variables and WOE (weight of evidence) transformation was carried out.Based on the processed data,the logic regression algorithm were used to construct the electricity cost risk forecasting model of the electric customers,and output variable standard score card was quantified according to the model results.Thus the customers were divided into high,middle and low risk users that could provide the basis for taking differential marketing measures to the different customers.

Key words: tariff recovery, logical regression algorithm, index system

CLC Number:

O213.1

Yadi ZHAO,Zhao WU,Qingbing LI,Xiaofeng CHEN,Baoting WANG. Application of big data method in forecasting the risk of tariff recovery[J]. Telecommunications Science, 2019, 35(2): 125-133.

Figures/Tables 12

References 17

[1]	ROBU V , VINYALS M , ROGERS A ,et al. Efficient buyer groups for prediction-of-use electricity tariffs[C]// AAAI Conference on Artificial Intelligence,July 27-31,2014,Québec City,Québec,Canada. Palo Alto:AIII Press, 2014: 451-457.
[2]	PEPERMANS G , WILLEMS B . Cost recovery in congested electricity networks[J]. Zeitschrift Für Energiewirtschaft, 2010,34(3): 195-208.
[3]	张晓峰 . 电力大客户电费回收风险防范体系的构建[J]. 内蒙古科技与经济, 2013(24): 121-123.
	ZHANG X F . Construction of risk prevention system for electric power customer's electricity recovery[J]. Inner Mongolia Science Technology ＆ Economy, 2013(24): 121-123.
[4]	裘华东, 涂莹, 丁麒 . 基于标签库系统的电力企业客户画像构建与信用评估及电费风险防控应用[J]. 电信科学, 2017,33(1): 214-221.
	QIU H D , TU Y , DING Q . Construction of power customer portrait and its credit evaluation and electricity fee risk control based on tag library system[J]. Telecommunications Science, 2017,33(Z1): 214-221.
[5]	HUANG Y , YAN Y . Research of evaluating credit-risk in power enterprise based on SVM and VIKOR method[C]// IEEE International Conference on Industrial Engineering and Engineering Management,Dec 8-11,2009,Singapore. Piscataway:IEEE Press, 2009: 1596-1599.
[6]	刘金 . 基于改进层次分析法的电力客户欠费风险评价[J]. 电子世界, 2014(19): 182-183.
	LIU J . Risk evaluation of electric power customer arrears based on improved analytic hierarchy process[J]. Electronic World, 2014(19): 182-183.
[7]	WIGAN M R , CLARKE R . Big data’s big unintended consequences[J]. Computer, 2013,46(6): 46-53.
[8]	RAHMAN M N , ESMAILPOUR A , ZHAO J . Machine learning with big data an efficient electricity generation forecasting system[J]. Big Data Research, 2016(5): 9-15.
[9]	杨华飞, 李栋华, 程明 . 电力大数据关键技术及建设思路的分析和研究[J]. 电力信息与通信技术, 2015,13(1): 7-10.
	YANG H F , LI D H , CHENG M . Analysis and research of key technologies and construction ideas of power big data[J]. Electric Power Information and Communication Technology, 2015,13(1): 7-10.
[10]	赵永良, 秦萱, 吴尚远 ,等. 基于数据挖掘的高压用户电费回收风险预测[J]. 电力信息与通信技术, 2015,13(9): 57-61.
	ZHAO Y L , QIN X , WU S Y ,et al. Electricity recovery risk prediction of high-voltage customers based on data mining[J]. Electric Power Information and Communication Technology, 2015,13(9): 57-61.
[11]	黄文思, 郝悍勇, 李金湖 ,等. 基于决策树算法的电力客户欠费风险预测[J]. 电力信息与通信技术, 2016(1): 19-22.
	HUANG W S , HAO H Y , LI J H ,et al. Prediction of power customer arrear risk based on decision tree algorithm[J]. Electric Power Information and Communication Technology, 2016(1): 19-22.
[12]	吴漾, 朱州 . 基于特征选择改进LR-Bagging算法的电力欠费风险居民客户预测[J]. 电子产品世界, 2017(4): 70-75.
	WU Y , ZHU Z . The arrears risk prediction of power residential customers based on LR-Bagging algorithm improved by feature selection[J]. Electronic Engineering and Products World, 2017(4): 70-75.
[13]	张禄, 潘鸣宇, 田贺平 ,等. 基于数据挖掘技术的电力客户欠费风险预警研究[J]. 2017.
	ZHANG L , PAN M Y , TIAN H P ,et al. Risk early-warning of electricity customers’arrears based on data mining technology[J]. 2017.
[14]	GRANELL R , AXON C J , WALLOM D C H . Predicting winning and losing businesses when changing electricity tariffs[J]. Applied Energy, 2014,133(10): 298-307.
[15]	KAMILARIS A , KALLURI B , KONDEPUDI S ,et al. A literature survey on measuring energy usage for miscellaneous electric loads in offices and commercial buildings[J]. Renewable ＆Sustainable Energy Reviews, 2014(34): 536-550.
[16]	GRANELL R , AXON C J , WALLOM D C H ,et al. Power-use profile analysis of non-domestic consumers for electricity tariff switching[J]. Energy Efficiency, 2016,9(3): 825-841.
[17]	FELDMAN R , SANGER J . Text mining handbook:advanced approaches in analyzing unstructured data[M]. Oxford City: Cambridge University PressPress, 2006.

Metrics

Recommended 0

No Suggested Reading articles found!

	变量名	特征规则		变量名	特征规则
规则一	现金缴费占比	>0.4且<0.71	规则二	现金缴费占比	>0.96
	缴费渠道数	>1		缴费次数	>2且≤4
	6—10日缴费次数	>0		5日之前缴费次数	>0且≤2
	6个月内计费次数	>2		6—20日缴费次数	=0
				21—25日缴费次数	≤2
				历史6个月电费平均值	≤257.853
				供电单位编号	*******06
				缴费渠道	支付宝

变更前渠道	变更后渠道	变更用户数	欠费用户数	欠费用户占比
微信缴费	电力机构柜台收费	130	81	62.31%
电力机构柜台收费	95598网站缴费	77	39	50.65%
电力机构柜台收费	支付宝缴费	1 363	571	41.89%
电力机构柜台收费	微信缴费	833	311	37.33%
支付宝缴费	电力机构柜台收费	451	147	32.59%

入模指标	指标名称	入模指标	指标名称
近6个月缴费回款时长平均值	PAY_TIME_AVG_6M	近6个月计费次数	RELEASE_NUM_6M
近6个月连续逾期次数	YQ_LAST_NUM_6M	缴费时长波动情况	BD_PAY_RATIO
近6个月平均电量环比	PQ_HB_6M	缴费次数	PAY_NUM
近6个月平均电量同比	PQ_TB_6M	现金缴费占比	CASH_P
近6个月交费及时度	PAY_IN_TIME_6M	电费平均值	T_AMT_AVG
25日之后缴费次数	PAY_NUM_25AFTER	电量方差	T_PQ_S
21—25日缴费次数	PAY_NUM_25_21	本期电量同比	PQ_TB_LAG
11—15日缴费次数	PAY_NUM_15_11	交费渠道变更次数	CHANGE_PAY_NUM
6—10日缴费次数	PAY_NUM_10_6	缴费渠道	PAY_MODE
6日之前缴费次数	PAY_NUM_5_BFORE	供电单位	ORG_NO
缴费回款时长方差	PAY_TIME_S

入模指标	系数
常数项	-4.132 907 9
供电单位	0.926 893 9
近6个月缴费回款时长平均值	-0.551 051 3
电量方差	-0.527 719 6
缴费回款时长方差	-0.329 336
缴费次数	-0.362 532
25日之后缴费次数	-0.232 415 4
11—15日缴费次数	-0.362 621 2
6日之前缴费次数	-0.366 416 4
电费平均值	-0.290 359 2
本期电量同比	-0.279 614
近6个月平均电量同比	-0.288 589 2
缴费渠道	0.184 463 8

变量名称	变量分组	得分
近6个月缴费	(-1,0]	1.71
回款时长平均值	(0,3.33]	1.47
	(3.33,6.67]	2.02
	(6.67,10.33]	2.33
	(10.33,13.67]	2.59
	(13.67,17.33]	2.93
	(17.33,22]	3.20
	(22,57]	3.84
供电单位	S区A所	2.40
	S区B所	2.46
	S区C所	1.96
	S区D所	2.50
	S区E所	2.62
电量方差	(-1,0]	2.86
	(0,5.33]	2.34
	(5.33,32.33]	2.43
	(32.33,50]	2.20
	(50,544.5]	2.30
	(544.5,722]	2.44
	(722,2 094.33]	2.34
	(2 094.33,inf]	2.45
6日之前缴费次数	(-1,0]	2.44
	(0,1]	2.39
	(1,2]	2.36
	(2,3]	2.17
	(3,inf]	1.82

Application of big data method in forecasting the risk of tariff recovery

RichHTML

PDF下载

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 12

References 17

Related Articles 6

Metrics

Recommended 0

[1]	Shanshan LI, Jianhong XU, Tao LI, Yongjian ZHAO, Zhifei LIU. Research on digital twin network capability maturity model of telecom operators [J]. Telecommunications Science, 2023, 39(6): 149-158.
[2]	KANG Chongqing,WANG Yi,ZHANG Jing,WANG Yongzhen,GAO Feng,LIU Yi. Development of national energy internet:index system and future trend [J]. Telecommunications Science, 2019, 35(6): 2-14.
[3]	Li LI,Tiantian LUO,Yana ZHANG,Runze WU. Multi-objective layered diagnosis model of electric power ICT network based on extenics [J]. Telecommunications Science, 2017, 33(2): 113-119.
[4]	Huifang WANG. Fuzzy evaluation in the selecting for excellence of telecommunication engineering design [J]. Telecommunications Science, 2016, 32(11): 161-164.
[5]	Gaoji Zhang,Jiandong Lu,Wenxue Chen. Study on Energy Conservation and Emission Reduction’s Evaluation Index and Methods about Communication Base Station [J]. Telecommunications Science, 2011, 27(3): 101-105.
[6]	Lihua Deng,Hongjuan Liang. Research on Evaluation System of Business Intelligence System [J]. Telecommunications Science, 2009, 25(9): 38-42.