A risk assessment method and system based on user electricity consumption information

Abstract

A risk assessment method and system based on user electricity consumption information, characterized in that the method includes the following steps: collecting electricity consumption data of electricity users and filtering out a key data set containing multiple key indicators from the electricity consumption data; constructing a multi-dimensional classification system, and dividing the key indicator data set according to different electricity users based on the multi-dimensional classification system to obtain multiple classified user sets; for each electricity user, calculating the evaluation value of each key indicator of the current electricity user according to the deviation of the classified user set corresponding to the current electricity user in the multi-dimensional classification system; obtaining the risk assessment result of the current electricity user according to the evaluation value of each key indicator of the current electricity user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and more specifically, to a risk assessment method and system based on user power consumption information. Background Art

[0002] In recent years, the use of power data to analyze corporate or individual electricity consumption behavior, conduct user profiling, corporate credit, and business risk assessment has been applied to a certain extent. Some technologies collect corporate power data and business data to build a corporate credit risk database and analyze the characteristics of dishonest companies and high-risk companies. Some use the Delphi survey method and hierarchical analysis method based on specific power data to determine the corporate risk assessment model and build a risk assessment system based on the model.

[0003] All kinds of applications basically collect some power data, build an indicator library for specific application scenarios, and build an analysis model based on the indicator library, or directly build an analysis model based on the data, and finally use the model for analysis and provide scenario services. In addition, the existing technology mainly focuses on evaluating the customer's operating risk, credit risk, etc. through corresponding analysis models or indicators, obtaining evaluation scores or grades or evaluation reports, and then applying them to specific scenarios.

[0004] However, the prior art still has the following problems:

[0005] First, different electricity users have great differences in electricity usage characteristics due to their different regions, industries, and sizes. Existing technologies lack pre-classification of users when using electricity data, or only a few technical solutions simply classify them according to regions and industries, without considering the correlation between regions, the correlation between different industries, and the correlation between users of different sizes.

[0006] Secondly, most solutions in the existing technology only analyze the power data of a single user (enterprise). Even if an analysis model is constructed through big data, when conducting user (enterprise) evaluation, only the power data of the user is used as the input of the model, and there is no real-time dynamic collection of power data from other users to conduct comprehensive analysis.

[0007] In addition, most existing technologies output single analysis results for specific application scenarios, such as enterprise operational risks, other enterprise risks, or comprehensive scores and other assessment results.

[0008] Finally, there are deficiencies in the protection of data assets in current risk assessment methods. Many solutions will output or provide the original value of the user's electricity consumption data to external users for use, or external users can reversely calculate the user's original electricity consumption data through corresponding analysis results. Even if the external user obtains the user's authorization, there is still a risk of leakage of the original data assets.

[0009] In view of the above problems, there is an urgent need for a risk assessment method and system based on user electricity usage information. Summary of the invention

[0010] In order to solve the deficiencies in the prior art, the present invention provides a risk assessment method and system based on user electricity usage information, which calculates each key indicator of power users by pre-dividing and classifying user sets and constructing key indicators, thereby implementing risk assessment of power users.

[0011] The present invention adopts the following technical solution.

[0012] A first aspect of the present invention relates to a risk assessment method based on user electricity usage information, the method comprising the following steps: collecting electricity usage data of electricity users and screening out a key data set containing multiple key indicators from the electricity usage data; constructing a multi-dimensional classification system, and dividing the key indicator data set according to different electricity users based on the multi-dimensional classification system to obtain multiple classified user sets; for each electricity user, calculating the evaluation value of each key indicator in the current electricity user according to the deviation of the classified user set corresponding to the current electricity user in the multi-dimensional classification system; obtaining the risk assessment result of the current electricity user according to the evaluation value of each key indicator in the current electricity user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator.

[0013] Preferably, electricity consumption data of power users are collected and a key data set containing multiple key indicators is screened out from the electricity consumption data, including: the electricity consumption data of power users includes basic information of power users, electricity consumption, and electricity charges; the key indicators include basic electricity consumption information, electricity consumption level, electricity consumption fluctuation, electricity consumption trend, electricity charge analysis, and breach of contract behavior; the key indicators are calculated based on the value results of one or more electricity consumption data within a first preset time period and a second preset time period.

[0014] Preferably, a multidimensional classification system is constructed, and based on the multidimensional classification system, the key indicator data set is divided according to different electricity users to obtain multiple classified user sets, including: the multidimensional classification system includes regional dimension, industry dimension and scale dimension, and the scale dimension is divided based on the user's electricity consumption capacity; based on the multidimensional classification system, the key indicator data set is divided according to different electricity users, including: multiple division levels of the regional dimension, multiple division levels of the industry dimension, and division levels of the scale dimension are arranged and combined to obtain multidimensional classifications of multiple electricity users; all electricity users that meet any multidimensional classification are extracted to construct a classified user set for the current multidimensional classification.

[0015] Preferably, for each power user, according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system, it includes: if the classified user set corresponding to the current power user is , then the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system is:

[0016] ,

[0017] In the formula, Characterization Region Industry, scale The classified user set corresponding to all users of ;

[0018] For all classified user sets A collection of;

[0019] To distinguish from the current user category set User category set In the current key indicator, all users The average value of the indicators below;

[0020] Classify sets for users User Classification Set The correlation coefficient between .

[0021] Preferably, the evaluation amount of each key indicator of the current power user is calculated, including: the current key indicator The evaluation amount for:

[0022] ,

[0023] In the formula, It is a key indicator for current electricity users The value of .

[0024] Preferably, the risk assessment result of the current power user is obtained based on the evaluation amount of each key indicator among the current power users, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator, including: extracting the power consumption data of the power users within a first preset time period, and calculating the basic evaluation weight; extracting the power consumption data of the power users within a second preset time period, and calculating the dynamic evaluation weight; the first preset time period is equal to the sum of multiple second preset time periods.

[0025] Preferably, the risk assessment result of the current power user is obtained based on the evaluation amount of each key indicator of the current power user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator, including: calculating the weighted sum of the evaluation amount of each key indicator of the current power user and the basic evaluation weight to obtain the basic evaluation of the current power user; calculating the weighted sum of the evaluation amount of each key indicator of the current power user and the dynamic evaluation weight to obtain the dynamic evaluation of the current power user.

[0026] Preferably, user profiling and user risk assessment are implemented based on the evaluation quantity, basic evaluation, and dynamic evaluation of each key indicator among current electricity users.

[0027] The second aspect of the present invention relates to a risk assessment system based on user electricity usage information using the method in the first aspect of the present invention, the system comprising a screening module, a division module, a calculation module, and an evaluation module; wherein the screening module is used to collect electricity usage data of power users and to filter out a key data set containing multiple key indicators from the electricity usage data; the division module is used to construct a multi-dimensional classification system, and based on the multi-dimensional classification system, the key indicator data set is divided according to different power users to obtain multiple classified user sets; the calculation module is used to calculate, for each power user, the evaluation value of each key indicator in the current power user according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system; the evaluation module is used to obtain the risk assessment result of the current power user based on the evaluation value of each key indicator in the current power user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator.

[0028] A third aspect of the present invention relates to a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method in the first aspect of the present invention.

[0029] The beneficial effect of the present invention is that, compared with the prior art, the risk assessment method and system based on user electricity usage information in the present invention calculates each key indicator of the power user by pre-dividing the classified user set and constructing key indicators, thereby implementing risk assessment of the power user. The present invention fully considers the inherent characteristics of users in different regions, different industries, and different scales, and implements a comprehensive assessment of user risks in an extended scenario through a comprehensive analysis of other user information under the same classification and different classifications. The application scenarios are wide and the assessment results are accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of a process of a risk assessment method based on user electricity usage information of the present invention;

[0031] Figure 2The present invention is a schematic diagram of the architecture of a risk assessment system based on user electricity usage information. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in the present invention are only embodiments of a part of the present invention, not all embodiments. Based on the spirit of the present invention, all other embodiments not recorded in the present invention obtained by ordinary technicians in this field according to the embodiments recorded in the present invention without making creative work should belong to the protection scope of the present invention.

[0033] Figure 1 FIG. 1 is a flow chart of a risk assessment method based on user electricity usage information of the present invention. Figure 1 As shown, the first aspect of the present invention relates to a risk assessment method based on user electricity usage information, and the method includes steps 1 to 4.

[0034] Step 1: Collect electricity consumption data of power users and filter out key data sets containing multiple key indicators from the electricity consumption data.

[0035] In order to analyze the electricity consumption data of power users, it is necessary to extract relevant data of power users. In the present invention, the power user to be analyzed is associated with other power users for analysis to overcome the influence of the power grid on the power consumption behavior of a single user and more accurately analyze the power consumption risks of individual users and corporate users due to user-side factors.

[0036] Preferably, the electricity consumption data of the power users are collected and a key data set containing multiple key indicators is screened out from the electricity consumption data, including: the electricity consumption data of the power users includes basic information of the power users, electricity consumption, and electricity charges.

[0037] The basic information of power users includes the enterprise's unified social credit code, user number, user address, user industry, power business type, power capacity, and historical changes in the above information. Electricity consumption includes information such as daily electricity consumption and hourly electricity consumption extracted from the user's electricity consumption curve. Electricity charges include electricity charge balance, payment type, historical payment information, historical arrears information, etc. The above data is obtained from the various information systems of the power supply enterprise, or obtained on demand through the data center of the power supply enterprise.

[0038] For the original electricity consumption data of users, normalization and standardization should be performed. For example, the electricity consumption capacity of users can be calculated by dividing the level of their capacity by the upper limit of the level. The indicator reflecting the electricity consumption of users in a certain month can be divided by a specific electricity consumption value. The specific electricity consumption value can be determined according to the indicator system, such as the average electricity consumption of the industry in the same month, a fixed value agreed in advance, etc. The indicator reflecting the electricity bill of users in a certain month can be divided by a specific electricity bill value, such as the average electricity bill of the industry in the same month, a fixed value agreed in advance, etc. The indicator reflecting the length of time that users open an electricity account can be divided by the length of time (month or year) that the account is opened by a specific value. For non-original electricity consumption data in the indicator system, dimensionless processing should be performed. The indicator reflecting "the electricity consumption level of users in the past year" can be divided by the electricity consumption of users in the past year by the average electricity consumption of users in the same region, industry, and scale in the past year. The indicator reflecting "the abnormal electricity consumption of users on a certain day" can be divided by the electricity consumption of that day by the daily average electricity consumption of the previous month or last week.

[0039] Based on the above data content, the method also calculates multiple key indicators. In one embodiment of the present invention, the key indicators include basic electricity consumption information, electricity consumption level, electricity fluctuation, electricity trend, electricity fee analysis, and breach of contract. These indicators are obtained by calculation. For example, electricity fluctuation is obtained based on the maximum difference between the user's electricity consumption at multiple times in the same day, and electricity fee analysis is calculated based on multiple information in the user's electricity bill, and so on. By setting multiple key indicators, a key data set is constructed based on the key indicators. The data set includes the above six dimensions, and the quantifiable values ​​of the six dimensional indicators within a time period can be calculated for each electricity user.

[0040] In other embodiments of the present invention, similar methods are used to select key indicators of different quantities and meanings according to analysis requirements, and obtain key data sets.

[0041] Preferably, the key indicator is calculated based on the value results of one or more electricity consumption data in the first preset time period and the second preset time period.

[0042] In the present invention, the first and second preset time periods can be used to calculate the evaluation weights in subsequent steps. At the same time, in order to implement reasonable prediction, the method extracts data within the first preset time period and the second preset time period in the stage of extracting the data set to obtain the key data set. Generally speaking, the first preset time period is greater than the second preset time period and can be an integer multiple of the second preset time period.

[0043] Step 2: construct a multi-dimensional classification system, and divide the key indicator data set according to different power users based on the multi-dimensional classification system to obtain multiple classified user sets.

[0044] After obtaining the key data set, the method classifies the electricity users according to the multi-dimensional classification system, and decomposes the key data set into multiple classified user sets according to the classification. In the present invention, the construction process of the multi-dimensional classification system is obtained after analyzing a large amount of user electricity usage information. In the user electricity usage information, there are great differences between the electricity usage information due to the differences in multiple dimensional features. After decomposing these dimensions, the information of a large number of users tends to be similar in the same dimensional interval. Therefore, users of multiple dimensions are divided, and the initial electricity usage information is adjusted according to the degree of deviation of the data information of a classified user set among all users, that is, the group bias of the current set is removed, which can overcome the evaluation errors of different types of users in the same risk assessment model.

[0045] Preferably, the multi-dimensional classification system includes regional dimension, industry dimension and scale dimension, and the scale dimension is divided based on the user's electricity consumption capacity.

[0046] In order to divide user groups, in one embodiment of the present invention, the characteristics of power users are taken into consideration, and the regional dimension, industry dimension and scale dimension are selected. With the development of the future power grid and the development of power purchase business, the number and content of the dimensions can be adjusted or increased according to actual needs.

[0047] In the present invention, the regional dimension is based on the administrative division where the electricity user number (e.g., each terminal meter that collects electricity consumption data corresponds to a household number) is located. In the present invention, the regional dimension should be accurate to at least the township (street) level. The industry dimension is obtained based on the industry classification specified in the national standard (e.g., GB / T4754). The scale dimension is based on the user's electricity consumption capacity and is divided into several levels.

[0048] In one embodiment, a set of empirical grading is provided including a plurality of capacity intervals, namely [0, 100), [100, 200), [200, 300), [300, 400), [400, 600), [600, 800), [800, 1200), [1200, 2000), [2000, 3000), [3000, 4000), [4000, 5000), [5000, +∞), with the unit being “kilovolt-ampere”).

[0049] It is easy to imagine that the process of dividing the electricity consumption capacity can refer to the sum of the rated powers of the electricity equipment owned by different types of electricity users, and divide the capacity intervals by analyzing the electricity equipment of different individual users and corporate users, as well as the utilization rate of electricity equipment. The division process can consider dividing multiple users more evenly into multiple unequal capacity intervals, or it can consider implementing a grouping algorithm based on the similarity of data such as actual electricity consumption levels, electricity fluctuations, electricity trends, and electricity charges, and reasonably divide all users into multiple user groups, and then implement the capacity interval division based on the user capacity situation within each group.

[0050] Preferably, the key indicator data set is divided according to different electricity users based on a multi-dimensional classification system, including: arranging and combining multiple division levels of the regional dimension, multiple division levels of the industry dimension, and division levels of the scale dimension to obtain a multi-dimensional classification of multiple electricity users; extracting all electricity users that meet any multi-dimensional classification to construct a classified user set for the current multi-dimensional classification.

[0051] Since each dimension is divided into multiple intervals, the three dimensions can be divided into multiple three-dimensional spaces according to the arrangement and combination of different intervals in the three-dimensional space. After dividing the user types, the key indicators of all users in a space are calculated to solve the average value of each key indicator. Based on the six key indicators in step 2, the method calculates the solution results of each key indicator in turn. For the key indicators , calculation space The key indicator values ​​of all users in the system are averaged to obtain the calculation result. Characterization Region Industry, scale The three-dimensional space, the classified user set corresponding to this three-dimensional space is .

[0052] Step 3: For each power user, the evaluation value of each key indicator of the current power user is calculated according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system.

[0053] Preferably, for each power user, according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system, it includes: if the classified user set corresponding to the current power user is , then the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system is:

[0054] ,

[0055] In the formula, To distinguish from the current user category set User category set In the current key indicator, all users The average value of the indicators below; Classify sets for users User Classification Set The correlation coefficient between .

[0056] In the above formula, the correlation coefficient The calculation method is:

[0057] ,

[0058] In the formula, For key indicators In two spaces and The covariance between

[0059] For space Key indicators The variance of

[0060] For space Key indicators The variance of .

[0061] In the above formula, the parameters and Long-term historical electricity consumption can be used as a reference. Therefore, when calculating the above data, the original electricity consumption information used can be the electricity consumption of users over a long period of time, and a more accurate indicator value can be calculated from this. In one embodiment, the time length is 3 years.

[0062] In another embodiment, the parameter and The value of is updated in a shorter period of time. In one embodiment, the above parameters are updated every 1 month.

[0063] Step 3: Obtain the risk assessment result of the current power user according to the assessment value of each key indicator of the current power user, as well as the basic assessment weight and dynamic assessment weight of each key indicator.

[0064] In order to obtain the risk status of an electricity user in the current situation, the method can use the calculation results obtained in step 1 and step 2.

[0065] Preferably, the evaluation amount of each key indicator of the current power user is calculated, including: the current key indicator The evaluation amount for:

[0066] ,

[0067] In the formula, It is a key indicator for current electricity users The data may be the value of the power user at the current forecast moment, or may be the average value in a recent shorter time period, such as the value in the second preset time period.

[0068] In this formula, As the current value of a key indicator, the indicator value of the key indicator is obtained through the classification of the user, so as to implement the analysis of the key indicator. The process of obtaining the key indicator evaluation quantity ensures that the power user overcomes the influence of its region, industry, scale and other dimensions on the electricity consumption behavior data. In the process of analyzing electricity consumption behavior, it ensures that the evaluation data excludes errors in these dimensions, and then explores the electricity consumption risks on the user side in the key indicators.

[0069] The basic evaluation weight and the dynamic evaluation weight are obtained by using a variety of algorithms. The present invention is based on one of the following calculations: entropy method, analytic hierarchy process, Delphi method, priority diagram method, and principal component method.

[0070] Multi-dimensional data are selectively organized according to the common methods of various algorithms, and indicator weights are obtained through comprehensive evaluation.

[0071] The Delphi method is an expert-oriented decision-making method that reaches consensus through multiple rounds of anonymous expert surveys and feedback. The Delphi method process includes expert selection, question raising, multiple rounds of feedback, analysis of survey results, and final agreement. Experts from various fields are invited to solicit opinions. After forming an initial set of weights for each indicator based on existing theoretical results, multiple rounds of opinions are solicited from experts to finally obtain the basic evaluation weights.

[0072] Preferably, the risk assessment result of the current power user is obtained based on the evaluation amount of each key indicator among the current power users, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator, including: extracting the power consumption data of the power users within a first preset time period, and calculating the basic evaluation weight; extracting the power consumption data of the power users within a second preset time period, and calculating the dynamic evaluation weight; the first preset time period is equal to the sum of multiple second preset time periods.

[0073] After obtaining the evaluation value of each key indicator of the current power user, a key indicator vector constructed by multiple key indicators of the current power user is obtained. The key indicator vector can be used to conduct a comprehensive analysis of various different power consumption situations of the current power user.

[0074] In one embodiment, the above key indicator vector is depicted in the user portrait, so as to obtain a reasonable representation of the basic electricity usage status of the user. Through the user portrait, the user's electricity usage in various dimensions can be obtained more clearly and quickly.

[0075] For corporate electricity users, by calculating the evaluation amount of key indicators and generating user portraits, the credit of corporate electricity users' electricity consumption behavior can be further analyzed, and the business risks and other risks of the enterprise can be evaluated. In addition, based on the differences between the basic evaluation and dynamic evaluation of current electricity users, a reliable corporate credit risk database can be built, and on this basis, the characteristics of dishonest enterprises and high-risk enterprises can be analyzed, and the electricity consumption behavior of enterprises can be given an evaluation grade and evaluation report.

[0076] Through the improvement of the method of the present invention, the process of pre-classifying users supports the investigation of the correlation between users according to dimensions such as user region, industry, and user scale, and on this basis, overcomes the deviation caused by multi-dimensional similarity, and filters out the user's own electricity consumption behavior characteristics for evaluation. In addition, when evaluating the electricity consumption behavior of a single user, the behavioral characteristics of other electricity users in the same time period are also collected and considered. By conducting a comprehensive analysis, the abnormal electricity consumption behavior of a single user is effectively extracted. Through a multi-dimensional, multi-user, and comprehensive analysis method of various indicators, it is possible to output comprehensive evaluation results of various aspects such as enterprise operating risks, other enterprise risks, and comprehensive scores through a single analysis.

[0077] Furthermore, in the process of implementing the index extraction and filtering the user's own electricity consumption behavior characteristics, the method adds special calculation processes on the basis of the original data. These calculation processes can be relatively simple, and are added and implemented in various types of power data acquisition systems. Each time the data is retrieved for calculation, what is actually collected is non-original data that has been simply processed. This ensures that only the risk assessment system where the present invention is located can agree on the transmission message format with the original power acquisition system, and extract relevant indicators from the accurate location of the corresponding transmission data. In the process of data transmission, there is no risk of data asset leakage. Even if other applications are connected to the same power data acquisition system at the same time, and other applications obtain the same level of information retrieval authority, they cannot implement the analysis of these intermediate indicator data, thereby greatly improving the security of the original power data assets.

[0078] Similarly, the evaluation results are also used for secondary development in different scenarios. For example, financial institutions use the above-processed series of indicators to conduct a single pre-loan evaluation of target customers, and then decide on the loan credit limit for customers based on the historical and current operating conditions of the target customers in their regions, industries, and scale classifications. By dynamically tracking the above-processed series of indicators, selecting some indicators with strong timeliness for further combined analysis, the dynamic changes in the relative operations of the target customers in their regions, industries, and scale classifications can be obtained, and negative warning information or positive business investment opportunity reminder information can be obtained, thereby assisting decision-making financial institutions in real-time adjustments to their policies for target customers. Some companies evaluate and dynamically manage suppliers to screen high-quality suppliers. For analysis of specific regions and industries, through the above-processed series of indicators, in-depth and detailed analysis of the historical development and dynamic changes of the operations of enterprises in specific regions and industries can be conducted based on the region, industry, and scale.

[0079] Figure 2 FIG. 1 is a schematic diagram of the architecture of a risk assessment system based on user electricity usage information of the present invention. Figure 2 As shown, the second aspect of the present invention relates to a risk assessment system based on user electricity usage information using the method in the first aspect of the present invention, the system comprising a screening module, a division module, a calculation module, and an evaluation module; wherein the screening module is used to collect electricity usage data of power users and to filter out a key data set containing multiple key indicators from the electricity usage data; the division module is used to construct a multi-dimensional classification system, and based on the multi-dimensional classification system, the key indicator data set is divided according to different power users to obtain multiple classified user sets; the calculation module is used to calculate, for each power user, the evaluation value of each key indicator in the current power user according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system; the evaluation module is used to obtain the risk assessment result of the current power user according to the evaluation value of each key indicator in the current power user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator.

[0080] A third aspect of the present invention relates to a terminal, comprising a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method in the first aspect of the present invention.

[0081] A fourth aspect of the present invention relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method in the first aspect of the present invention.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A risk assessment method based on user electricity usage information, characterized in that: The method comprises the following steps: Collecting electricity consumption data of power users and filtering out a key data set containing multiple key indicators from the electricity consumption data; The electricity consumption data of the power user includes basic information of the power user, electricity consumption, and electricity charges; The key indicators include basic electricity consumption information, electricity consumption level, electricity consumption fluctuation, electricity consumption trend, electricity fee analysis, and breach of contract behavior; The key indicator is calculated based on the value results of one or more of the electricity consumption data in the first preset time period and the second preset time period; Constructing a multi-dimensional classification system, and dividing the key indicator data set according to different power users based on the multi-dimensional classification system to obtain multiple classified user sets; For each power user, according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system, the evaluation value of each key indicator of the current power user is calculated; If the classified user set corresponding to the current power user is , then the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system is: In the formula, Characterization Region Industry, scale The classified user set corresponding to all users; For all the classified user sets A collection of; To distinguish from the current one the classified user set Classification user set In the current key indicator, all users The average value of the indicators below; For the classified user set With the classified user set The correlation coefficient between The risk assessment result of the current power user is obtained according to the assessment value of each key indicator of the current power user, as well as the basic assessment weight and dynamic assessment weight of each key indicator.

2. A risk assessment method based on user electricity usage information according to claim 1, characterized in that: The multi-dimensional classification system is constructed, and based on the multi-dimensional classification system, the key indicator data set is divided according to different power users to obtain multiple classified user sets, including: The multi-dimensional classification system includes a regional dimension, an industry dimension and a scale dimension, and the scale dimension is divided based on the power consumption capacity of the user; The dividing the key indicator data set according to different power users based on the multi-dimensional classification system includes: Arrange and combine the multiple classification levels of the region dimension, the multiple classification levels of the industry dimension, and the multiple classification levels of the scale dimension to obtain a multi-dimensional classification of the multiple power users; All power users that meet any of the multi-dimensional classifications are extracted to construct the classified user set for the current multi-dimensional classification.

3. A risk assessment method based on user electricity usage information according to claim 2, characterized in that: The calculating of the evaluation amount of each key indicator of the current power user includes: A key indicator The evaluation amount for: In the formula, A key indicator for current electricity users The value of .

4. A risk assessment method based on user electricity usage information according to claim 3, characterized in that: The risk assessment result of the current power user is obtained according to the assessment amount of each key indicator of the current power user, as well as the basic assessment weight and dynamic assessment weight of each key indicator, including: Extracting the electricity consumption data of the power user within the first preset time period, and calculating the basic evaluation weight; extracting the electricity consumption data of the power user within the second preset time period, and calculating the dynamic evaluation weight; The first preset time period is equal to the sum of multiple second preset time periods.

5. A risk assessment method based on user electricity usage information according to claim 4, characterized in that: The risk assessment result of the current power user is obtained according to the assessment amount of each key indicator of the current power user, as well as the basic assessment weight and dynamic assessment weight of each key indicator, including: Calculating a weighted sum of an evaluation amount of each key indicator of the current power user and the basic evaluation weight to obtain a basic evaluation of the current power user; The weighted sum of the evaluation amount of each key indicator of the current power user and the dynamic evaluation weight is calculated to obtain the dynamic evaluation of the current power user.

6. A risk assessment method based on user electricity usage information according to claim 5, characterized in that: User profiling and user risk assessment are implemented based on the evaluation quantity, basic evaluation and dynamic evaluation of each key indicator of the current electricity users.

7. A risk assessment system based on user electricity usage information using the method according to any one of claims 1 to 6, characterized in that: The system includes a screening module, a division module, a calculation module, and an evaluation module; wherein, The screening module is used to collect the power consumption data of the power users and screen out a key data set containing multiple key indicators from the power consumption data; The division module is used to construct a multi-dimensional classification system, and divide the key indicator data set according to different power users based on the multi-dimensional classification system to obtain multiple classified user sets; The calculation module is used to calculate, for each power user, an evaluation value of each key indicator of the current power user according to the deviation of the classified user set corresponding to the current power user in the multi-dimensional classification system; The evaluation module is used to obtain the risk evaluation result of the current power user according to the evaluation value of each key indicator of the current power user, as well as the basic evaluation weight and dynamic evaluation weight of each key indicator.

8. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-6.

Images