CN109697574A - Small customer electricity Risk Identification Method in a kind of electric power - Google Patents
Small customer electricity Risk Identification Method in a kind of electric power Download PDFInfo
- Publication number
- CN109697574A CN109697574A CN201811651701.6A CN201811651701A CN109697574A CN 109697574 A CN109697574 A CN 109697574A CN 201811651701 A CN201811651701 A CN 201811651701A CN 109697574 A CN109697574 A CN 109697574A
- Authority
- CN
- China
- Prior art keywords
- risk
- client
- electricity
- arrears
- customer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 143
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000011156 evaluation Methods 0.000 claims abstract description 28
- 238000012502 risk assessment Methods 0.000 claims abstract description 25
- 230000003542 behavioural effect Effects 0.000 claims abstract description 23
- 238000003064 k means clustering Methods 0.000 claims abstract description 11
- 230000011218 segmentation Effects 0.000 claims abstract description 11
- 239000000284 extract Substances 0.000 claims abstract description 7
- 238000009826 distribution Methods 0.000 claims description 19
- 238000009941 weaving Methods 0.000 claims description 16
- 239000004753 textile Substances 0.000 claims description 9
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000011161 development Methods 0.000 claims description 6
- 230000018109 developmental process Effects 0.000 claims description 6
- 235000013305 food Nutrition 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 3
- 230000002950 deficient Effects 0.000 claims description 3
- 239000004744 fabric Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000008520 organization Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000000926 separation method Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 235000013399 edible fruits Nutrition 0.000 claims description 2
- 238000003786 synthesis reaction Methods 0.000 claims description 2
- 230000004069 differentiation Effects 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 238000009472 formulation Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 2
- 230000035515 penetration Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses customer electricity Risk Identification Methods small in a kind of electric power, are related to power domain.Currently, medium and small client uses risk there are various.The technical program extracts the evaluation index and corresponding evaluation index data of medium and small client to be identified from three clean government's risk, arrears risk, security risk Risk Dimensions respectively, based on K-means clustering algorithm, it carries out the risk identification of three dimensions respectively to the medium and small client of pending risk assessment, and extracts the electricity consumption behavioural characteristic of small client in each risk；The risk identification result of three clean government's risk, arrears risk, security risk dimensions is integrated, medium and small customer risk identification model is constructed, realizes the medium and small customer segmentation based on customer risk, to provide basis for differentiation, personalized marketing service.
Description
Technical field
The present invention relates to small customer electricity Risk Identification Methods in power domain more particularly to a kind of electric power.
Background technique
With the deep development of " internet+", power customer differentiation, personalized service demand are highlighted, in order to more
Well, more targetedly meet client's multiple demands, it would be highly desirable to effectively be classified to client, be needed for deep grasp client
Offer analysis personnel is sought, the service strategy formulated accordingly can accurately meet the power supply service demand of different clients, realize
Differentiation and personalized service.
In terms of the research currently segmented to power customer, focus mostly in the subdivision research to large power customers, for medium and small
The research of client is related to less, but in practical customer group, medium and small customer quantity is more, industry is extensive, social influence is wide,
It is necessary to be finely divided to medium and small customer group.Moreover, medium and small client is during electricity consumption, there is Multifunctional electric classification contract,
The features such as average electricity price is high, mobility is big, various risks such as Yi Fasheng arrears risk, security risk, internal clean government.
Summary of the invention
The technical problem to be solved in the present invention and the technical assignment of proposition are to be improved and improved to prior art,
Small customer electricity Risk Identification Method in a kind of electric power is provided, with reduction risk purpose.For this purpose, the present invention takes following skill
Art scheme.
Small customer electricity Risk Identification Method in a kind of electric power, comprising the following steps:
1) according to the concept defining and feature of medium and small client, the target client of pending risk identification is determined；According to client from
Right attribute and electricity consumption behavior property carry out medium and small client's concept defining；Natural quality includes category of employment, mobility；Electricity consumption behavior
Attribute includes electricity consumption, arrearage number
The medium and small client includes the user of the classification containing commercial power in low pressure commercial user and low-voltage customer, such client's
Feature are as follows: Population is more, category of employment is more；The operator of electricity utilization site and user's separation, mobility are big；Electricity consumption is big,
Social influence is wider；Average electricity price is high；Multifunctional electric classification shares；Plurality of classes easily occurs uses risk；；
2) risk is carried out according to the behavioural characteristic of user to define, be divided into arrears risk, security risk, clean government's risk；
It include arrears risk, safety utilization of electric power risk, internal clean government risk with risk；Wherein, arrears risk refers in electricity consumption
The case where owing electricity charges generated in journey；Safety utilization of electric power risk refers to safety wind caused by safety accident occurred as customer side etc.
Danger；Internal clean government's risk refers to, in power industry, for the medium and small trade company of Multifunctional electric categories combination, appraises and decides carrying out electricity price
When, there is the case where quantitative, to compare surely；Personnel are appraised and decided when carrying out appraising and deciding electricity in the presence of the risk played one's own game；
3) the three classes consumer's risk classification of basis obtains risk assessment index；
Arrears risk evaluation index includes behavioral indicator and payment index, and the behavioral indicator of arrears risk includes for reflecting client
With the load character of electrical characteristics, the working capacity for reflecting customer electricity scale, for reflecting the monthly average electricity consumption of client
The monthly average electricity consumption of scale, for reflect client to the monthly electricity charge that are averaged of the contribution situation of Utilities Electric Co., for reflecting
The year electricity growth rate of customer electricity variation tendency and development potentiality；The payment index of arrears risk is for reflecting advance visitor
The monthly advance amount of money of family electricity charge situation, payment credit situation for reflecting in the customer evaluation phase accumulative arrearage number, use
In the monthly average arrearage amount of money of the arrearage total amount in the reflection customer evaluation phase, for reflecting arrears risk feelings in the customer evaluation phase
The paid penalty total value of condition；
Security risk assessment index includes behavioral indicator, safety index, and the behavioral indicator of security risk includes for reflecting client
Electricity consumption scale voltage class, for reflect client with the load character of electrical characteristics, for reflecting the set operation of client
The contract capacity of scale, the working capacity for reflecting the actual operation scale of client, for reflecting that client uses beyond rule
The over capacity of fixed capacity；The safety index of security risk includes the power supply points for reflecting the electricity consumption categorical measure of client
Amount, Electrical Safety behavior for reflecting client safety accident number；
Clean government's risk assessment index includes behavioral indicator, electricity price index, and the behavioral indicator of clean government's risk includes for reflecting client
Monthly average resident/business electricity accounting of electricity consumption distribution, electricity consumption category distribution for reflecting client monthly average residence
The people/business electricity charge accounting, the monthly average electricity for reflecting the monthly average electricity consumption scale of client, for reflecting client to electric power
The monthly average electricity charge of the contribution situation of company；The electricity price index of clean government's risk includes for reflecting the flat of advance electricity price situation
The electricity price categorical measure of equal electricity price, quantity for reflecting electricity consumption classification that client uses；
4) according to Risk Assessment Index System, three classes risk evaluation index data are extracted respectively, are known as pending risk
Other basis sample data；And the data of extraction are tentatively cleaned, are arranged, check that data whether there is abnormal conditions, it is right
It is substituted in exceptional value situation by the way of directly deleting or according to the quantile ratio of data distribution；
5) identification of three classes risk is carried out respectively to the medium and small client of pending risk identification based on K-means clustering algorithm, and
That extracts small client in each risk uses electrical feature；
6) according to the identification situation of three classes risk, the three classes risk identification result of medium and small client is integrated, by three classes risk
The different risk clients identified compare, if some client is the client of high risk in three classes risk,
Judge the client for high clean government's risk, high arrears risk, high safety risk client；It is integrated and is tied according to three classes risk identification result
Fruit constructs medium and small customer risk identification model, extracts the electricity consumption behavioural characteristic of user of all categories, realizes based in customer risk
Small customer segmentation.
The technical program is taken out from three clean government's risk, arrears risk, security risk Risk Dimensions respectively in measurement period
The evaluation index and corresponding evaluation index data of medium and small client to be identified are taken, K-means clustering algorithm is based on, to pending wind
The medium and small client nearly assessed carries out the risk identification of three dimensions respectively, and extracts the electricity consumption behavior of small client in each risk
Feature；The risk identification result of three clean government's risk, arrears risk, security risk dimensions is integrated, medium and small client is constructed
Risk identification model realizes the medium and small customer segmentation based on customer risk, to provide for differentiation, personalized marketing service
Basis.
Point of penetration is identified as with medium and small customer risk and carries out medium and small client segmentation, can either be reduced and be used from two level of inside and outside
Family risk, but the deep enough identification user's feature of energy, the formulation for targetedly marketing service strategy provide support.
As optimization technique means: step 5) the following steps are included:
S1) to the processing of evaluation index data normalization；
Initial data is standardized, data dimension is eliminated；
S2) by arrears risk evaluation index data, security risk assessment achievement data, clean government's risk assessment achievement data respectively into
Row K-means clustering, and carry out classification of risks；Multiple tentative calculation is carried out to cluster number in clustering, until cluster
As a result it is optimal；It in cluster process, if there are exceptional value, excluding outliers for cluster result, then is clustered, until cluster
As a result without abnormal.
As optimization technique means: in step 5), carrying out different electricity consumptions respectively to the medium and small client of pending risk assessment
When the risk class identification of risk, multiple assessment achievement data carries out clustering according to K-means clustering algorithm；Arrearage wind
Danger, security risk, clean government's risk carry out the cluster measuring and calculating of different cluster numbers respectively, determine optimum cluster number；It is determining most
After excellent cluster number, according to the cluster result of optimum clustering number, each classification sample number is determined；According to the cluster of optimum clustering number
As a result, determining the affiliated class categories of single sample.
As optimization technique means: the class categories of arrears risk evaluation index are as follows: high electricity high risk type, low battery are high
State of risk, devoid of risk type；The class categories of security risk assessment index are as follows: high risk type, average risk type, low-risk type；It is honest and clean
The class categories of atmosphere in government bodies danger evaluation index are as follows: high risk type, average risk type, low-risk type.
As optimization technique means: in the sample in the combination of small customer segmentation result, according to client's property and future
Policy development object, entire sample is subjected to the classification of sample customer risk, is classified as eight classes, it is respectively high arrears risk, cheap
Atmosphere in government bodies danger, lower security risk；High safety risk, low arrears risk, cheap atmosphere in government bodies danger；High clean government's risk, low arrears risk, low peace
Full blast danger；High clean government's risk, high arrears risk, lower security risk；High clean government's risk, high safety risk, low arrears risk；Gao An
Full blast danger, high arrears risk, cheap atmosphere in government bodies danger；High clean government's risk, high arrears risk, high safety risk；Lower security risk, it is low owe
Take risk, cheap atmosphere in government bodies danger.
As optimization technique means: to " high arrears risk, cheap atmosphere in government bodies danger, lower security risk；High safety risk, it is low owe
Take risk, cheap atmosphere in government bodies danger；High clean government's risk, low arrears risk, lower security risk；High clean government's risk, high arrears risk, low peace
Full blast danger；High clean government's risk, high safety risk, low arrears risk；High safety risk, high arrears risk, cheap atmosphere in government bodies danger；"
Clean government's risk, security risk, the index of arrears risk of client counts, analyze every a kind of client specific electricity consumption situation and
Behavioural characteristic；And stored, using the foundation of the identification as medium and small power customer risk.
As optimization technique means: specific electricity consumption situation and behavioural characteristic are compared with preset recording, when variation,
Modify database；The preset recording includes:
The first kind: high arrears risk, cheap atmosphere in government bodies danger, lower security risk
The client's arrearage and arrearage amount of money is higher frequent occurrence, adding up arrearage number is 1-2 time, and the non-350-20000 of the arrearage amount of money is first
Differ；Monthly average electricity price is lower, is 0.68-0.89 member；Electricity price classification is 1-2 kind；Occur safety accident number be almost
Zero；The industry of concentration includes that weaving, clothes and daily necessities are specially sold, dinner service, comprehensive retail business；Wherein weaving, clothes
And daily necessities are specially sold accounting rate higher than dinner service, comprehensive retail business；
Second class: high safety risk, low arrears risk, cheap atmosphere in government bodies danger
Safety accident frequent occurrence, safety accident number are 1-2 times；Arrearage situation is less, and adding up arrearage number is 0-1 times, the moon
The average arrearage amount of money is spent less than 160 yuan；Monthly average electricity price is higher, is 0.59-1.05 member；Electricity price classification is 1-2 kind；Main point
The cloth region more in rural area self-built housing；In industry distribution, including comprehensive retail business, dinner service, broadcast television transmissions clothes
Business, telecommunications, weaving, clothes and daily necessities are specially sold, and comprehensive retail business accounting rate is higher than dinner service, broadcast television transmissions
Service, telecommunications, weaving, clothes and daily necessities are specially sold；
Third class: Gao Lianzheng risk, low arrears risk, lower security risk
Such monthly average electricity price is lower, is 0.5-0.66 member；Arrearage situation is less, and arrearage number is 0-1 times, monthly average deficient
Take the amount of money lower than 150 yuan；Electricity price classification is 1-3 kind；The number for safety accident occur is seldom；In industry distribution, including synthesis
Retail business, weaving, clothes, daily necessities are specially sold, telecommunications, and radio and television propagate service, wherein comprehensive retail business accounting rate is high
It is specially sold in weaving, clothes, daily necessities, telecommunications, radio and television propagate service；
4th class: Gao Lianzheng risk, high arrears risk, lower security risk
Average electricity price is lower, is 0.56-0.67 member；Electricity price classification is 2-4 kind；There are arrearage situations, add up arrearage number 1-2
Secondary, the monthly average arrearage amount of money is up to more than 7000 yuan；Safety accident seldom occurs；Industry distribution includes textile garment and daily
Product retail, amusement, comprehensive retail, instant food manufacture；
5th class: Gao Lianzheng risk, high safety risk, low arrears risk
Average electricity price is lower, is 0.54-0.71 member；Electricity price classification is 2-4 kind；Arrearage situation did not occurred, adds up arrearage number
It is 0, the monthly average arrearage amount of money not 0；There are security risks, and safety accident occurred, at 1-2 times；Industry includes food and drink, lives
Place, textile garment and daily necessities retail, telecommunications, broadcast television transmissions；
6th class: high safety risk, high arrears risk, cheap atmosphere in government bodies danger
Average electricity price is higher, in 0.81-0.86 member；Only a kind of electricity price classification；There are arrears risks, add up arrearage number in 1-
2 times, the monthly average arrearage amount of money is at 1500 yuan or less；1-2 safety accident occurred；Industry includes textile garment and daily necessities
Retail, comprehensive retail, organization and administration service.
As optimization technique means: when being standardized evaluation index data, being marked using Z-score method
Standardization.
The utility model has the advantages that
The technical program is identified as point of penetration with medium and small customer risk and carries out medium and small client segmentation, can either be from two level of inside and outside
User power utilization risk is reduced, and the deep enough identification user's feature of energy, the formulation for targetedly marketing service strategy provide support.
The technical program carries out classification of risks to medium and small client using K-means clustering algorithm, by the risk of three dimensions
Recognition result is integrated, and is formed medium and small customer electricity risk identification model, is realized medium and small customer segmentation.It is accurate and comprehensive, have
Risk is used conducive to the medium and small client of reduction.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is distribution situation of the medium and small client in every profession and trade；
Fig. 3 is the small accumulative arrearage number of client year in every profession and trade；
Fig. 4 is small client secure fault occurrences in every profession and trade；
Fig. 5 is each risk client number distribution situation.
Specific embodiment
Technical solution of the present invention is described in further detail below in conjunction with Figure of description.
As shown in Figure 1, small customer electricity Risk Identification Method in electric power, comprising the following steps:
Step 1, concept defining and signature analysis are carried out to medium and small client, determines the target client of pending risk identification.
Medium and small client refers to the user of the classification containing commercial power in low pressure commercial user and low-voltage customer, the spy of such client
Sign are as follows: Population is more, category of employment is more；The operator of electricity utilization site and user's separation, mobility are big；Electricity consumption is big, society
It is wider to will affect face；Average electricity price is high；Multifunctional electric classification shares；Plurality of classes easily occurs uses risk.
Step 2, user power utilization risk is bound, determines consumer's risk classification to be identified, is divided into three classes；
The medium and small customer electricity risk analyzed in the present invention refers to raw arrears risk, safety utilization of electric power risk, internal clean government risk,
Wherein, arrears risk refers to during electricity consumption, client's flowing, supervise the ineffective, usage of trade etc. due to generate be in arrears with
The case where electricity charge；Safety utilization of electric power risk refers to security risk caused by safety accident occurred as customer side etc..Internal clean government's wind
Danger refers to, in power industry, for the medium and small trade company of Multifunctional electric categories combination, when progress electricity price is appraised and decided, there are quantitative, fixed
Compare the case where.The personnel of appraising and deciding have certain right, when carrying out appraising and deciding electricity, there is the risk played one's own game.
Step 3, according to determining three classes consumer's risk classification, risk assessment index is determined respectively.Arrears risk, safety wind
Below the evaluation index of dangerous, internal clean government risk shown in three tables.
One arrears risk evaluation index system of table
Two security risk assessment index system of table
Three clean government's Risk Assessment Index System of table
Step 4, according to Risk Assessment Index System, three classes risk evaluation index data are extracted respectively, and to the number of extraction
According to tentatively being cleaned；
According to fixed arrears risk, security risk, clean government's Risk Assessment Index System, corresponding achievement data is extracted, is made
For the basic sample data of pending risk identification；
Edit is carried out to the data of extraction, checks that data are abnormal with the presence or absence of missing values, unknown-value, invalid value of variable etc.
Situation substitutes exceptional value situation by the way of directly deleting or according to the quantile ratio of data distribution.
Step 5, it is based on K-means clustering algorithm, three classes risk is carried out respectively to the medium and small client of pending risk identification
Identification, and that extracts small client in each risk uses electrical feature；
Based on improved K-means clustering algorithm, the process classified to small customer risk in electric power to be identified are as follows:
S1) to the processing of evaluation index data normalization；
Since there are larger differences in terms of dimension for initial data, the accuracy of cluster result will affect, it is therefore desirable to original
Data are standardized, and eliminate data dimension, and the method that standardization mainly takes Z-Score carries out.
S2 evaluation index data) are directed respectively into K-means clustering software, carry out classification of risks.
Arrears risk evaluation index data are imported into K-means clustering software, pass through the tentative calculation to different cluster numbers, hair
When present poly- 5 class, sample classification is more suitble to.Electrical feature is used by 5 classes of analysis, this five class can be summarized as three classes by discovery,
That is: high electricity high risk type, low battery high risk type, devoid of risk type.
Security risk assessment achievement data is imported into K-means clustering software, passes through the tentative calculation to different cluster numbers, hair
When present poly- 3 class, sample classification is more suitble to.Electrical feature is used by 3 classes of analysis, three classes can be summarized are as follows: high risk type, one
As state of risk, low-risk type.
Clean government's risk assessment achievement data is imported into K-means clustering software, passes through the tentative calculation to different cluster numbers, hair
When present poly- 3 class, sample classification is more suitble to.Electrical feature is used by 3 classes of analysis, three classes can be summarized are as follows: high risk type, one
As state of risk, low-risk type.
Step 6, according to the identification situation of three classes risk, the three classes risk identification result of medium and small client is integrated, is used
It is identical in the sample data of three classes risk identification, therefore the different risk clients that three classes risk identification goes out is carried out pair
Than if some client is the client of high risk in three classes risk, which is high clean government's risk, high arrearage wind
Danger, high safety risk client.According to three classes risk identification result integrated results, medium and small customer risk identification model is constructed, is extracted
The electricity consumption behavioural characteristic of user of all categories realizes the medium and small customer segmentation based on customer risk.
Find that high risk classification situation is less according to the cluster result analysis to three classes risk.Therefore, by risk height and wind
Danger is higher to be classified as one kind, is referred to as high risk.Finally by the three classes of each dimension, it is classified as two classes: high risk, low-risk.
It, will according to client's property and the policy development object in future in the sample in the combination of small customer segmentation result
Entire sample is classified as eight classes.Wherein, wherein 1-6 class is paid close attention to, as shown in the table.
Four sample customer risk of table is sorted out
The clean government's risk, security risk, the index of arrears risk of 1-6 class client are counted respectively, analyze every a kind of client
Specific electricity consumption situation and behavioural characteristic.
Every class electrical feature is as follows:
The first kind: high arrears risk, cheap atmosphere in government bodies danger, lower security risk
Such client arrearage and arrearage amount of money is higher frequent occurrence, adding up arrearage number is 1-2 times, the non-350- of the arrearage amount of money
20000 yuan are differed；Monthly average electricity price is lower, is 0.68-0.89 member；Electricity price classification is 1-2 kind；There is the number of safety accident
It is almost nil；It is concentrated mainly on weaving, clothes and daily necessities to be specially sold, dinner service, comprehensive retail business, wherein weaving, clothes
It is more that dress and daily necessities are specially sold accounting.
Second class: high safety risk, low arrears risk, cheap atmosphere in government bodies danger
Safety accident frequent occurrence, safety accident number are 1-2 times；Arrearage situation is less, and adding up arrearage number is 0-1 times, the moon
The average arrearage amount of money is spent less than 160 yuan；Monthly average electricity price is higher, is 0.59-1.05 member；Electricity price classification is 1-2 kind；Main point
The cloth region more in rural area self-built housing, it will usually there is the big electricity user of resident to cause safety problem；In industry distribution, mainly
Comprehensive retail business is concentrated on, is specially sold in dinner service, broadcast television transmissions service, telecommunications, weaving, clothes and daily necessities
Industry is also distributed larger.
Third class: Gao Lianzheng risk, low arrears risk, lower security risk
Such monthly average electricity price is lower, is 0.5-0.66 member；Arrearage situation is less, and arrearage number is 0-1 times, monthly average deficient
Take the amount of money lower than 150 yuan；Electricity price classification is 1-3 kind；The number for safety accident occur is seldom；It is concentrated mainly on comprehensive retail business,
Weaving, clothes and daily necessities are specially sold, telecommunications, and radio and television propagate service etc., wherein comprehensive retail business accounting is more.
4th class: Gao Lianzheng risk, high arrears risk, lower security risk
Average electricity price is lower, is 0.56-0.67 member；Electricity price classification is 2-4 kind；There are arrearage situations, add up arrearage number 1-2
Secondary, the monthly average arrearage amount of money is up to more than 7000 yuan；Safety accident seldom occurs.Industry distribution includes textile garment and daily
Product retail, amusement, comprehensive retail, instant food manufacture etc..
5th class: Gao Lianzheng risk, high safety risk, low arrears risk
Average electricity price is lower, is 0.54-0.71 member；Electricity price classification is 2-4 kind；Arrearage situation did not occurred, adds up arrearage number
It is 0, the monthly average arrearage amount of money not 0；There are security risks, and safety accident occurred, at 1-2 times.Industry includes food and drink, lives
Place, textile garment and daily necessities retail, telecommunications, broadcast television transmissions, other special retails etc..
6th class: high safety risk, high arrears risk, cheap atmosphere in government bodies danger
Average electricity price is higher, in 0.81-0.86 member；Only a kind of electricity price classification；There are arrears risks, add up arrearage number in 1-
2 times, the monthly average arrearage amount of money is at 1500 yuan or less；1-2 safety accident occurred.Industry includes textile garment and daily necessities
Retail, comprehensive retail, organization and administration service etc..
The present embodiment takes the electricity consumption situation in certain period and area, is counted, is obtained: medium and small client as shown in Figure 2 exists
The distribution situation of every profession and trade；The small accumulative arrearage number of client year in every profession and trade as shown in Figure 3；In every profession and trade as shown in Figure 4
Small client secure fault occurrences；Fig. 5 is each risk client number distribution situation；According to above-mentioned data as this area
The foundation of medium and small power customer electricity consumption classification of risks, reduces user power utilization risk, but can deep enough identification user's feature, for for
Property marketing service strategy formulation provide support.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, several improvement and deformations can also be made, these improvement and deformations
Also it should be regarded as protection scope of the present invention.
Claims (8)
1. small customer electricity Risk Identification Method in a kind of electric power, it is characterised in that the following steps are included:
1) according to the concept defining and feature of medium and small client, the target client of pending risk identification is determined；According to client from
Right attribute and electricity consumption behavior property carry out medium and small client's concept defining；Natural quality includes category of employment, mobility；Electricity consumption behavior
Attribute includes electricity consumption, arrearage number；
The medium and small client includes the user of the classification containing commercial power in low pressure commercial user and low-voltage customer, such client's
Feature are as follows: Population is more, category of employment is more；The operator of electricity utilization site and user's separation, mobility are big；Electricity consumption is big,
Social influence is wider；Average electricity price is high；Multifunctional electric classification shares；Plurality of classes easily occurs uses risk；
2) risk is carried out according to the behavioural characteristic of user to define, be divided into arrears risk, security risk, clean government's risk；
It include arrears risk, safety utilization of electric power risk, internal clean government risk with risk；Wherein, arrears risk refers in electricity consumption
The case where owing electricity charges generated in journey；Safety utilization of electric power risk refers to safety wind caused by safety accident occurred as customer side etc.
Danger；Internal clean government's risk refers to, in power industry, for the medium and small trade company of Multifunctional electric categories combination, appraises and decides carrying out electricity price
When, there is the case where quantitative, to compare surely；Personnel are appraised and decided when carrying out appraising and deciding electricity in the presence of the risk played one's own game；
3) the three classes consumer's risk classification of basis obtains risk assessment index；
Arrears risk evaluation index includes behavioral indicator and payment index, and the behavioral indicator of arrears risk includes for reflecting client
With the load character of electrical characteristics, the working capacity for reflecting customer electricity scale, for reflecting the monthly average electricity consumption of client
The monthly average electricity consumption of scale, for reflect client to the monthly electricity charge that are averaged of the contribution situation of Utilities Electric Co., for reflecting
The year electricity growth rate of customer electricity variation tendency and development potentiality；The payment index of arrears risk is for reflecting advance visitor
The monthly advance amount of money of family electricity charge situation, payment credit situation for reflecting in the customer evaluation phase accumulative arrearage number, use
In the monthly average arrearage amount of money of the arrearage total amount in the reflection customer evaluation phase, for reflecting arrears risk feelings in the customer evaluation phase
The paid penalty total value of condition；
Security risk assessment index includes behavioral indicator, safety index, and the behavioral indicator of security risk includes for reflecting client
Electricity consumption scale voltage class, for reflect client with the load character of electrical characteristics, for reflecting the set operation of client
The contract capacity of scale, the working capacity for reflecting the actual operation scale of client, for reflecting that client uses beyond rule
The over capacity of fixed capacity；The safety index of security risk includes the power supply points for reflecting the electricity consumption categorical measure of client
Amount, Electrical Safety behavior for reflecting client safety accident number；
Clean government's risk assessment index includes behavioral indicator, electricity price index, and the behavioral indicator of clean government's risk includes for reflecting client
Monthly average resident/business electricity accounting of electricity consumption distribution, electricity consumption category distribution for reflecting client monthly average residence
The people/business electricity charge accounting, the monthly average electricity for reflecting the monthly average electricity consumption scale of client, for reflecting client to electric power
The monthly average electricity charge of the contribution situation of company；The electricity price index of clean government's risk includes for reflecting the flat of advance electricity price situation
The electricity price categorical measure of equal electricity price, quantity for reflecting electricity consumption classification that client uses；
4) according to Risk Assessment Index System, three classes risk evaluation index data are extracted respectively, are known as pending risk
Other basis sample data；And the data of extraction are tentatively cleaned, are arranged, check that data whether there is abnormal conditions, it is right
It is substituted in exceptional value situation by the way of directly deleting or according to the quantile ratio of data distribution；
5) identification of three classes risk is carried out respectively to the medium and small client of pending risk identification based on K-means clustering algorithm, and
That extracts small client in each risk uses electrical feature；
6) according to the identification situation of three classes risk, the three classes risk identification result of medium and small client is integrated, by three classes risk
The different risk clients identified compare, if some client is the client of high risk in three classes risk,
Judge the client for high clean government's risk, high arrears risk, high safety risk client；It is integrated and is tied according to three classes risk identification result
Fruit constructs medium and small customer risk identification model, extracts the electricity consumption behavioural characteristic of user of all categories, realizes based in customer risk
Small customer segmentation.
2. small customer electricity Risk Identification Method in a kind of electric power according to claim 1, it is characterised in that: step 5) packet
Include following steps:
S1) to the processing of evaluation index data normalization；
Initial data is standardized, data dimension is eliminated；
S2) by arrears risk evaluation index data, security risk assessment achievement data, clean government's risk assessment achievement data respectively into
Row K-means clustering, and carry out classification of risks；Multiple tentative calculation is carried out to cluster number in clustering, until cluster
As a result it is optimal；It in cluster process, if there are exceptional value, excluding outliers for cluster result, then is clustered, until cluster
As a result without abnormal.
3. small customer electricity Risk Identification Method in a kind of electric power according to claim 2, it is characterised in that: in step
5) when, carrying out the different risk class with risk respectively to the medium and small client of pending risk assessment and identifying, multiple assessment refers to
It marks data and clustering is carried out according to K-means clustering algorithm；Arrears risk, security risk, clean government's risk carry out difference respectively
The cluster measuring and calculating for clustering number, determines optimum cluster number；After determining optimum cluster number, according to the cluster of optimum clustering number
As a result, determining each classification sample number；According to the cluster result of optimum clustering number, the affiliated class categories of single sample are determined.
4. small customer electricity Risk Identification Method in a kind of electric power according to claim 3, it is characterised in that: arrears risk
The class categories of evaluation index are as follows: high electricity high risk type, low battery high risk type, devoid of risk type；Security risk assessment index
Class categories are as follows: high risk type, average risk type, low-risk type；The class categories of clean government's risk assessment index are as follows: high risk
Type, average risk type, low-risk type.
5. small customer electricity Risk Identification Method in a kind of electric power according to claim 4, it is characterised in that: in the sample
In the combination of small customer segmentation result, according to client's property and the policy development object in future, entire sample is subjected to sample
Customer risk is sorted out, and eight classes are classified as, respectively high arrears risk, cheap atmosphere in government bodies danger, lower security risk；High safety risk, it is low owe
Take risk, cheap atmosphere in government bodies danger；High clean government's risk, low arrears risk, lower security risk；High clean government's risk, high arrears risk, low peace
Full blast danger；High clean government's risk, high safety risk, low arrears risk；High safety risk, high arrears risk, cheap atmosphere in government bodies danger；Gao Lian
Atmosphere in government bodies danger, high arrears risk, high safety risk；Lower security risk, low arrears risk, cheap atmosphere in government bodies danger.
6. small customer electricity Risk Identification Method in a kind of electric power according to claim 5, it is characterised in that: to " height is owed
Take risk, cheap atmosphere in government bodies danger, lower security risk；High safety risk, low arrears risk, cheap atmosphere in government bodies danger；High clean government's risk, it is low owe
Take risk, lower security risk；High clean government's risk, high arrears risk, lower security risk；High clean government's risk, high safety risk, it is low owe
Take risk；High safety risk, high arrears risk, cheap atmosphere in government bodies danger；" clean government's risk of client, security risk, arrears risk
Index is counted, and the specific electricity consumption situation and behavioural characteristic of every a kind of client are analyzed；And stored, using as medium and small electric power
The foundation of the identification of customer electricity risk.
7. small customer electricity Risk Identification Method in a kind of electric power according to claim 6, it is characterised in that: will specifically use
Electric situation and behavioural characteristic are compared with preset recording, when variation, modify database；The preset recording includes:
The first kind: high arrears risk, cheap atmosphere in government bodies danger, lower security risk
The client's arrearage and arrearage amount of money is higher frequent occurrence, adding up arrearage number is 1-2 time, and the non-350-20000 of the arrearage amount of money is first
Differ；Monthly average electricity price is lower, is 0.68-0.89 member；Electricity price classification is 1-2 kind；Occur safety accident number be almost
Zero；The industry of concentration includes that weaving, clothes and daily necessities are specially sold, dinner service, comprehensive retail business；Wherein weaving, clothes
And daily necessities are specially sold accounting rate higher than dinner service, comprehensive retail business；
Second class: high safety risk, low arrears risk, cheap atmosphere in government bodies danger
Safety accident frequent occurrence, safety accident number are 1-2 times；Arrearage situation is less, and adding up arrearage number is 0-1 times, the moon
The average arrearage amount of money is spent less than 160 yuan；Monthly average electricity price is higher, is 0.59-1.05 member；Electricity price classification is 1-2 kind；Main point
The cloth region more in rural area self-built housing；In industry distribution, including comprehensive retail business, dinner service, broadcast television transmissions clothes
Business, telecommunications, weaving, clothes and daily necessities are specially sold, and comprehensive retail business accounting rate is higher than dinner service, broadcast television transmissions
Service, telecommunications, weaving, clothes and daily necessities are specially sold；
Third class: Gao Lianzheng risk, low arrears risk, lower security risk
Such monthly average electricity price is lower, is 0.5-0.66 member；Arrearage situation is less, and arrearage number is 0-1 times, monthly average deficient
Take the amount of money lower than 150 yuan；Electricity price classification is 1-3 kind；The number for safety accident occur is seldom；In industry distribution, including synthesis
Retail business, weaving, clothes, daily necessities are specially sold, telecommunications, and radio and television propagate service, wherein comprehensive retail business accounting rate is high
It is specially sold in weaving, clothes, daily necessities, telecommunications, radio and television propagate service；
4th class: Gao Lianzheng risk, high arrears risk, lower security risk
Average electricity price is lower, is 0.56-0.67 member；Electricity price classification is 2-4 kind；There are arrearage situations, add up arrearage number 1-2
Secondary, the monthly average arrearage amount of money is up to more than 7000 yuan；Safety accident seldom occurs；Industry distribution includes textile garment and daily
Product retail, amusement, comprehensive retail, instant food manufacture；
5th class: Gao Lianzheng risk, high safety risk, low arrears risk
Average electricity price is lower, is 0.54-0.71 member；Electricity price classification is 2-4 kind；Arrearage situation did not occurred, adds up arrearage number
It is 0, the monthly average arrearage amount of money not 0；There are security risks, and safety accident occurred, at 1-2 times；Industry includes food and drink, lives
Place, textile garment and daily necessities retail, telecommunications, broadcast television transmissions；
6th class: high safety risk, high arrears risk, cheap atmosphere in government bodies danger
Average electricity price is higher, in 0.81-0.86 member；Only a kind of electricity price classification；There are arrears risks, add up arrearage number in 1-
2 times, the monthly average arrearage amount of money is at 1500 yuan or less；1-2 safety accident occurred；Industry includes textile garment and daily necessities
Retail, comprehensive retail, organization and administration service.
8. small customer electricity Risk Identification Method in a kind of electric power according to claim 2, it is characterised in that: will evaluate
When achievement data is standardized, it is standardized using Z-score method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811651701.6A CN109697574A (en) | 2018-12-31 | 2018-12-31 | Small customer electricity Risk Identification Method in a kind of electric power |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811651701.6A CN109697574A (en) | 2018-12-31 | 2018-12-31 | Small customer electricity Risk Identification Method in a kind of electric power |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109697574A true CN109697574A (en) | 2019-04-30 |
Family
ID=66233125
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811651701.6A Pending CN109697574A (en) | 2018-12-31 | 2018-12-31 | Small customer electricity Risk Identification Method in a kind of electric power |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109697574A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288216A (en) * | 2019-06-17 | 2019-09-27 | 国网甘肃省电力公司兰州供电公司 | A kind of power customer service Risk-warning and trend analysis method |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111784042A (en) * | 2020-06-28 | 2020-10-16 | 佛山市南海区公共安全技术研究院 | Power utilization node safety risk prediction method and device and storage medium |
| CN112163781A (en) * | 2020-10-15 | 2021-01-01 | 国网冀北电力有限公司智能配电网中心 | Park electricity utilization group life cycle evaluation method based on multi-dimensional index clustering |
| CN113706336A (en) * | 2021-09-01 | 2021-11-26 | 知能汇融(北京)咨询有限公司 | Risk assessment method and device, computer equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120087300A (en) * | 2010-12-31 | 2012-08-07 | 한양대학교 산학협력단 | Method for evaluating risk level of power system |
| CN103824148A (en) * | 2013-08-29 | 2014-05-28 | 国家电网公司 | Power marketing risk pre-warning method and system |
| CN106504125A (en) * | 2016-12-27 | 2017-03-15 | 北京中电普华信息技术有限公司 | A kind of method and device for building power customer overall evaluation system |
| CN106557882A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | Power consumer screening technique and system based on various dimensions Risk Evaluation Factors |
-
2018
- 2018-12-31 CN CN201811651701.6A patent/CN109697574A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120087300A (en) * | 2010-12-31 | 2012-08-07 | 한양대학교 산학협력단 | Method for evaluating risk level of power system |
| CN103824148A (en) * | 2013-08-29 | 2014-05-28 | 国家电网公司 | Power marketing risk pre-warning method and system |
| CN106557882A (en) * | 2016-11-29 | 2017-04-05 | 国网山东省电力公司电力科学研究院 | Power consumer screening technique and system based on various dimensions Risk Evaluation Factors |
| CN106504125A (en) * | 2016-12-27 | 2017-03-15 | 北京中电普华信息技术有限公司 | A kind of method and device for building power customer overall evaluation system |
Non-Patent Citations (1)
| Title |
|---|
| 卢海明等: "电力客户细分及增值服务系统研究与应用", 《华北电力技术》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110288216A (en) * | 2019-06-17 | 2019-09-27 | 国网甘肃省电力公司兰州供电公司 | A kind of power customer service Risk-warning and trend analysis method |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111738462B (en) * | 2020-06-08 | 2022-08-30 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111784042A (en) * | 2020-06-28 | 2020-10-16 | 佛山市南海区公共安全技术研究院 | Power utilization node safety risk prediction method and device and storage medium |
| CN112163781A (en) * | 2020-10-15 | 2021-01-01 | 国网冀北电力有限公司智能配电网中心 | Park electricity utilization group life cycle evaluation method based on multi-dimensional index clustering |
| CN113706336A (en) * | 2021-09-01 | 2021-11-26 | 知能汇融(北京)咨询有限公司 | Risk assessment method and device, computer equipment and storage medium |
| CN113706336B (en) * | 2021-09-01 | 2024-02-02 | 知能汇融(北京)咨询有限公司 | Risk assessment method, risk assessment device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109697574A (en) | Small customer electricity Risk Identification Method in a kind of electric power | |
| CN110097297B (en) | Multi-dimensional electricity stealing situation intelligent sensing method, system, equipment and medium | |
| CN106780140B (en) | Power credit evaluation method based on big data | |
| CN107220906A (en) | Multiple Time Scales multiplexing electric abnormality analysis method based on electricity consumption acquisition system | |
| CN110739686B (en) | Method and system for managing line loss of transformer area based on total table anomaly analysis | |
| CN107862467A (en) | A kind of electric network synthetic data target monitoring method and system based on big data platform | |
| CN109063945A (en) | A kind of 360 degree of customer portrait construction methods of sale of electricity company based on Value accounting system | |
| CN107340492A (en) | Electric power meter failure analysis methods with scene anticipation are excavated based on big data | |
| CN111177389A (en) | NLP technology-based classification method, system and storage medium for power charge notification and customer appeal collection | |
| CN103455855B (en) | A kind of intermittent electricity stealing prevention detection method based on power information data analysis | |
| CN110119948B (en) | Power consumer credit evaluation method and system based on time-varying weight dynamic combination | |
| CN108256723B (en) | Economic benefit evaluation method for accessing coal-to-electricity into power grid and terminal equipment | |
| CN110427418A (en) | Customer analysis grouping method based on customer energy value index system | |
| CN113189418B (en) | Topological relation identification method based on voltage data | |
| CN106372747A (en) | Random forest-based zone area reasonable line loss rate estimation method | |
| CN112561339A (en) | High-quality customer identification method | |
| CN108389069A (en) | Top-tier customer recognition methods based on random forest and logistic regression and device | |
| CN106530139A (en) | Method for calculating the index parameter of grid investment analysis model | |
| CN116611862A (en) | Prediction method for monthly marketized purchase electric quantity of power supply company | |
| CN114022205A (en) | Power consumer payment channel preference matching method and system based on improved clustering method | |
| CN116187808A (en) | Electric power package recommendation method based on virtual power plant user-package label portrait | |
| CN114462983A (en) | Audit data processing method suitable for distribution network engineering | |
| CN112801542A (en) | Credit assessment method for electricity utilization client | |
| CN111932078A (en) | Risk user identification method based on measurement data multi-situation evaluation | |
| CN110070256B (en) | Zero-power user investigation priority weight calculation method based on CRITIC method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190430 |
|
| RJ01 | Rejection of invention patent application after publication |