Power customer value grading method and system based on customer grouping

Abstract

The invention discloses an electric power customer value grade division method and system based on customer grouping, and relates to the technical field of electric power customer value grade division, and the method comprises the steps: obtaining index data corresponding to a preset index from the power consumption data of each electric power customer; performing standardization processing on the preset indexes, and based on the standardized preset indexes, performing calculation by using an entropy weight method to obtain an index weight of each index in the preset indexes; based on the index weight of each index in the preset indexes, calculating to obtain a value score of each power customer; based on each preset value grade category, obtaining a target value grade of each power customer according to the value score and by using an improved k-means clustering algorithm; differentiated services are provided for users with different value levels in a targeted mode, and the user satisfaction degree is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power customer value grade classification, and more specifically, to a method and system for power customer value grade classification based on customer grouping. Background Art

[0002] With the gradual implementation of my country's power reform plan, power supply companies are facing increasingly fierce market competition. At present, there is no real intelligent two-way interaction between power customers and power supply companies, and information belongs to a one-way transmission mode; with the continuous construction and development of smart grids, a large amount of customer data assets have been accumulated, including basic data, business data, metering data and service data. In the era of customer first, the reasonable configuration of power services will directly determine the cost and customer service satisfaction of the enterprise. Therefore, power supply companies urgently need to conduct scientific analysis of customers, scientifically and reasonably group customers, identify the different demand points of different customer groups, formulate different marketing strategies for different customer groups, and provide customers with lean services to improve customer service satisfaction. This will enable power grid companies to further expand the depth and breadth of services and provide data support for the formulation of future power demand-side response policies. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for classifying power customer value levels based on customer grouping, so as to provide differentiated services to users of different value levels in a targeted manner and improve user satisfaction.

[0004] The above technical objectives of the present invention are achieved through the following technical solutions:

[0005] In a first aspect, the present application provides a method for classifying power customer value based on customer grouping, comprising the following specific steps:

[0006] Obtaining index data corresponding to preset indexes from the electricity consumption data of each electricity customer, the preset indexes including the electricity fee receivable index, the electricity consumption growth rate index and the arrears ratio index;

[0007] The preset indicators are standardized respectively, and based on the standardized preset indicators, the indicator weights of each indicator in the preset indicators are calculated using the entropy weight method;

[0008] Based on the indicator weights of each indicator in the preset indicators, the value score of each power customer is calculated;

[0009] Based on the preset value level categories, the target value level of each electricity customer is obtained according to the value score and the improved k-means clustering algorithm.

[0010] The beneficial effects of the present invention are as follows: in this solution, customer value is evaluated through three aspects: customer profitability, growth and risk, and profitability, growth and risk correspond to the receivable electricity fee index, electricity consumption growth rate index and arrears ratio index respectively; in terms of indicator weight calculation, by comparing the commonly used weight calculation methods of hierarchical analysis method, principal component analysis method and entropy weight method, the objective, scientific and clear indicator meaning entropy weight method is selected to calculate the indicator weight; customer value level division, the improved k-means clustering algorithm with high computational efficiency, stability and good effect is applied to finally obtain the target value level of each power customer, so that the power grid can provide differentiated services to users of different target value levels in a targeted manner, thereby improving user satisfaction.

[0011] Based on the above technical solution, the present invention can also be improved as follows.

[0012] Furthermore, the above standardization process is specifically as follows:

[0013] The indicators of electricity charges receivable and electricity consumption growth rate are as follows:

[0014] Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data;

[0015] The specific indicators for the proportion of arrears are as follows:

[0016] Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data.

[0017] The beneficial effects of adopting the above further scheme are: if the electricity bill receivable index and the electricity consumption growth rate index are high-quality indicators, the larger the value of the index, the better; if the arrears ratio index is a low-quality indicator, the smaller the value of the index, the better; before analyzing the data, the data set is normalized and the sample data is limited to a certain range. This not only facilitates the subsequent processing of the data, but also improves the operating efficiency of the algorithm.

[0018] Furthermore, the indicator weights of each indicator in the above preset indicators are specifically as follows:

[0019] in,

[0020] If p ij =0, then

[0021] Where W j represents the indicator weight of the jth indicator in the preset indicators, k represents the total amount of indicators in the preset indicators, k = 3; E j represents the information entropy of the jth indicator in the preset indicators, Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer, and n represents the total number of power customers.

[0022] The beneficial effect of adopting the above further scheme is that the index weight adopts the entropy weight method, which is scientific, objective, and has clear index meanings. It is a more suitable method for calculating index weights.

[0023] Furthermore, the value scores of the above-mentioned electricity customers are as follows:

[0024]

[0025] In the formula, F i represents the value score of electricity customer i, W j represents the index weight of the jth index in the preset index of power customer i, k represents the total amount of the index in the preset index, k = 3; Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer.

[0026] Furthermore, the target value level of the electricity customer is obtained by using the improved k-means clustering algorithm based on the value score, specifically:

[0027] A1, establish a corresponding number of clusters according to the number of value level categories, and apply the initial cluster center algorithm to determine the initial cluster center of each cluster;

[0028] A2, calculate the Euclidean distance between the value score of each electricity customer and each initial cluster center, and assign each value score to the cluster with the closest Euclidean distance;

[0029] A3, calculate the average value of all data objects in each cluster and use the average value as the new cluster center of the cluster;

[0030] A4, loop A2 to A3 until the new cluster centers of each cluster obtained after two iterations are the same.

[0031] Furthermore, the above-mentioned initial cluster center algorithm is used to determine the initial cluster center of each cluster, specifically:

[0032] A11, determining the minimum value score of each power customer as the first target value;

[0033] A12, calculating the Euclidean distance between the value scores of other power customers and the first target value, and arranging the value scores in ascending order according to the Euclidean distance;

[0034] A13, in the sequence of value scores arranged in ascending order, the value score numbered 1+(i-1)(n-2) / (k-1) is assigned to the kth cluster, where i=1, 2...k, and n represents the total number of value scores;

[0035] A14, for the value scores that are not classified into the cluster, calculate the Euclidean distance between each value score that is not classified into the cluster and each value score that is classified into the cluster, and classify each value score that is not classified into the cluster into the cluster with the smallest Euclidean distance;

[0036] A15, calculate the average value of each value score in each cluster, and determine the average value as the initial cluster center.

[0037] The beneficial effect of adopting the above further scheme is that the initial cluster center algorithm avoids random steps, making the clustering results more reliable. At the same time, the calculated initial cluster center points are evenly distributed and can better represent the overall distribution of the original data, which can appropriately reduce the number of iterations.

[0038] Furthermore, the above method also includes:

[0039] The number of times of each electricity payment method is obtained from the electricity consumption data of each electricity customer, and the electricity payment method with the most times of each electricity customer is determined as the preferred channel;

[0040] The deviation of each electricity customer's preferred channel relative to the preferred channels of other electricity customers is calculated, and based on the deviation of each electricity customer, the activity index corresponding to the preferred channel of each electricity customer is calculated.

[0041] The beneficial effect of adopting the above further scheme is: analyzing the customer's channel preference based on the customer's payment channel. The channel preference analysis is conducive to the precise marketing of the power grid, saving operation and marketing costs, reducing losses, and is conducive to the deployment of power grid service outlets to achieve the effect of precise marketing.

[0042] Furthermore, the deviation degree of one of the above-mentioned electricity customers is specifically:

[0043] in, In the formula, P represents the deviation of the preferred channel B of power customer A, X represents the number of times the preferred channel B of power customer A is used, represents the average number of times preferred channel B is used in the group of power customer A, σ represents the standard deviation of the number of times used in the group of power customer A, N represents the number of power customers in the group, X i represents the number of times the i-th electricity customer in the group uses the preferred channel B;

[0044] The activity index corresponding to one of the preferred channels of an electricity customer is as follows:

[0045] V=60+20*P, where V represents the activity index corresponding to the preferred channel B of electricity customer A, and P represents the deviation of the preferred channel B of electricity customer A.

[0046] Furthermore, the above method also includes:

[0047] The electricity consumption in each quarter is obtained from the electricity consumption data of each electricity customer, and the season with the largest electricity consumption is determined as the active season of each electricity customer.

[0048] The beneficial effect of adopting the above further solution is: analyzing the season with the highest electricity consumption based on the user's electricity consumption behavior is conducive to the adjustment of the power grid load.

[0049] In a second aspect, the present application provides a system for classifying power customer value based on customer grouping, which is applied to a method for classifying power customer value based on customer grouping in any one of the first aspects, including:

[0050] The first module is used to obtain index data corresponding to preset indicators from the electricity consumption data of each power customer, and the preset indicators include the receivable electricity fee index, the electricity consumption growth rate index and the arrears ratio index;

[0051] The second module is used to perform standardization processing on the preset indicators respectively, and based on the preset indicators after standardization, use the entropy weight method to calculate the indicator weight of each indicator in the preset indicators;

[0052] The third module is used to calculate the value score of each power customer based on the indicator weights of each indicator in the preset indicators;

[0053] The fourth module is used to obtain the target value level of each electricity customer based on the preset value level categories according to the value score and using the improved k-means clustering algorithm.

[0054] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the methods in the first aspect when executing the computer program.

[0055] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any one of the methods in the first aspect.

[0056] Compared with the prior art, the present invention has at least the following beneficial effects:

[0057] In this application, customer value is evaluated through three aspects: profitability, growth, and risk. Profitability, growth, and risk correspond to the receivable electricity bill index, electricity consumption growth rate index, and arrears ratio index, respectively. In terms of indicator weight calculation, the commonly used weight calculation methods, hierarchical analysis method, principal component analysis method, and entropy weight method are compared, and the entropy weight method, which is objective, scientific, and has clear indicator meanings, is selected to calculate the indicator weights. For customer value level division, an improved k-means clustering algorithm with high computational efficiency, stability, and good results is applied to finally obtain the target value level of each electricity customer. In this way, the power grid can provide differentiated services to users of different target value levels in a targeted manner, thereby improving user satisfaction.

[0058] In this application, the receivable electricity bill index and the electricity consumption growth rate index are high-quality indicators, and the larger the value of the index, the better; the arrears ratio index is a low-quality indicator, and the smaller the value of the index, the better; the data set is normalized before analyzing the data, and the sample data is limited to a certain range, which not only facilitates the subsequent processing of the data, but also improves the operating efficiency of the algorithm; the entropy weight method is used for indicator weights. This method is scientific, objective, and has clear indicator meanings. It is a more suitable method for calculating indicator weights; at the same time, the initial cluster center algorithm avoids random steps, making the clustering results more reliable. At the same time, the calculated initial cluster center points are evenly distributed, which can better represent the overall distribution of the original data, and the number of iterations can be appropriately reduced.

[0059] In this application, the customer's channel preference is analyzed based on the customer's payment channel. Channel preference analysis is beneficial to the precise marketing of the power grid, saving operation and marketing costs, reducing losses, and is beneficial to the deployment of power grid service outlets to achieve the effect of precise marketing; the season with the highest electricity consumption is analyzed based on the user's electricity consumption behavior, which is beneficial to the load adjustment of the power grid; comprehensive implementation of group management for customers, correctly grasping, satisfying and guiding customer needs, formulating differentiated service strategies in close accordance with customer needs, and providing customers with safe, reliable, efficient, convenient, green and environmentally friendly power supply services comparable to international advanced levels are the inevitable requirements for power supply companies to establish the three good service image of "good service, good management and good image". BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings:

[0061] Figure 1 A method flow chart of a division method in an embodiment of the present invention;

[0062] Figure 2 A connection diagram of a partitioning system in an embodiment of the present invention;

[0063] Figure 3 Schematic diagram of the connection of electronic equipment in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0065] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0067] In the description of the embodiments of the present invention, "plurality" means at least 2.

[0068] Embodiment 1:

[0069] The present application embodiment provides a method for classifying power customer value based on customer grouping, such as Figure 1 As shown, the specific steps include:

[0070] S1, obtaining index data corresponding to preset indexes from the electricity consumption data of each electricity customer, wherein the preset indexes include an index of electricity charges receivable, an index of electricity consumption growth rate, and an index of proportion of arrears.

[0071] Among them, the customer value level assessment is mainly considered from three aspects: first, profitability, which is measured by income. Income refers to the value created by customers for power supply companies, mainly electricity charges receivable. Therefore, profitability is measured by electricity charges receivable; second, growth. The increase in customer electricity consumption will increase the grid revenue and thus create greater value. Therefore, growth is measured by the electricity consumption growth rate; third, risk. Arrears will directly affect the power supply company's electricity charge recovery, making electricity charge recovery risky. Therefore, risk is measured by the proportion of arrears. Specifically, the electricity consumption data of power customers can be data for the past year.

[0072] S2, respectively standardize the preset indicators, and based on the standardized preset indicators, use the entropy weight method to calculate the indicator weight of each indicator in the preset indicators.

[0073] Among them, due to the different load sizes of electricity users, there are large differences between the original load data values, sometimes even multiple orders of magnitude, which will affect the entire model analysis effect if not processed. Therefore, before modeling and analyzing the data, the data set must be normalized to limit the sample data to a certain range. This not only facilitates the subsequent processing of the data, but also improves the running time of the algorithm in the model.

[0074] Optionally, the above standardization process is specifically as follows:

[0075] The indicators of electricity charges receivable and electricity consumption growth rate are as follows:

[0076] Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data.

[0077] Furthermore, the specific indicators for the proportion of arrears are as follows:

[0078] Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data.

[0079] S3, based on the indicator weights of each indicator in the preset indicators, calculate the value score of each electricity customer.

[0080] Among them, when evaluating customer value, the indicator weight reflects the position or role of each indicator in the decision-making process, which directly affects the scoring results. At present, the weight is usually given based on experience. It is undeniable that this can reflect the actual situation to a certain extent, and the evaluation results are more in line with reality; but the weight given based on experience is often subjective, and sometimes cannot objectively reflect the actual situation, and the evaluation results may be "distorted", so a scientific weight determination method should be used.

[0081] Specifically, the hierarchical analysis method has the characteristics of multiple indicators and the combination of qualitative and quantitative methods. However, in the process of judging the importance of each pair of indicators, it still needs experts or experienced people to judge, which is subjective to a certain extent. The principal component analysis method is a more objective method. It mainly calculates the weights based on the variance of the indicator data. It cannot truly reflect the contribution of the indicators to the comprehensive evaluation. The meaning of the principal component is generally a bit vague, not as clear and accurate as the meaning of the original variable. This is the price that must be paid in the process of variable dimensionality reduction. Therefore, the entropy weight method is used for indicator weights. This method is scientific, objective, and has clear indicator meanings. It is a more suitable method for calculating indicator weights.

[0082] Optionally, the indicator weight of each indicator in the above preset indicators is specifically:

[0083] in,

[0084] If p ij =0, then

[0085] Where W j represents the indicator weight of the jth indicator in the preset indicators, k represents the total amount of indicators in the preset indicators, k = 3; E j represents the information entropy of the jth indicator in the preset indicators, Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer, and n represents the total number of power customers.

[0086] S4, based on the preset value level categories, the target value level of each electricity customer is obtained according to the value score and using the improved k-means clustering algorithm.

[0087] Optionally, the value score of the above electricity customer is as follows:

[0088]

[0089] In the formula, F i represents the value score of electricity customer i, W jrepresents the index weight of the jth index in the preset index of power customer i, k represents the total amount of the index in the preset index, k = 3; Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer.

[0090] Among them, the user value level can be divided into 5 levels. Because it is not clear in advance what value level category the user belongs to, clustering technology is used to classify data without category information; clustering technology is usually called unsupervised learning, and there is no classification information that represents the data category in clustering. Clustering simply means grouping similar things into one group and dissimilar things into different groups; clustering algorithms mainly include hierarchical clustering algorithms and partition clustering algorithms.

[0091] Specifically, hierarchical clustering algorithms are divided into agglomerative hierarchical clustering and divisive hierarchical clustering. There is no difference between the two. Usually, agglomerative hierarchical clustering is used. The strategy of agglomerative hierarchical clustering is to first treat each object as a cluster, and then merge these atomic clusters into larger and larger clusters until all objects are in one cluster or a certain termination condition is met; hierarchical clustering algorithms need to repeatedly calculate the distance between two objects, and their time complexity is high ο(n^3), where n is the number of data objects. In addition, once two clusters are merged, they will not be undone. If one step is wrong, all steps will be wrong.

[0092] Furthermore, the k-means clustering algorithm is simple and efficient in calculation, but is sensitive to the initial cluster centers; in addition, randomly selecting the initial cluster centers will lead to unstable clustering results; therefore, an improved k-means clustering algorithm is used to divide customer value levels; the improved k-means clustering algorithm, based on the original k-means algorithm, improves its method of randomly selecting initial center points.

[0093] Optionally, the target value level of the electricity customer is obtained according to the value score and using an improved k-means clustering algorithm, specifically:

[0094] A1, establish a corresponding number of clusters according to the number of value level categories, and apply the initial cluster center algorithm to determine the initial cluster center of each cluster.

[0095] Optionally, the initial cluster center algorithm is applied to determine the initial cluster center of each cluster, specifically:

[0096] A11, determining the smallest value score of each electricity customer as the first target value.

[0097] A12, calculating the Euclidean distance between the value scores of other electricity customers and the first target value, and arranging the value scores in ascending order according to the Euclidean distance.

[0098] A13, in the sequence of value scores arranged in ascending order, the value score numbered 1+(i-1)(n-2) / (k-1) is assigned to the kth cluster, where i=1, 2…k, and n represents the total number of value scores.

[0099] A14, for the value scores that are not classified into the cluster, calculate the Euclidean distance between each value score that is not classified into the cluster and each value score that is classified into the cluster, and classify each value score that is not classified into the cluster into the cluster with the smallest Euclidean distance.

[0100] A15, calculate the average value of each value score in each cluster, and determine the average value as the initial cluster center.

[0101] Among them, the initial cluster center algorithm avoids random steps, making the clustering results more reliable. At the same time, the calculated initial cluster center points are evenly distributed and can better represent the overall distribution of the original data, which can appropriately reduce the number of iterations.

[0102] A2, calculate the Euclidean distance between the value score of each electricity customer and each initial cluster center, and assign each value score to the cluster with the closest Euclidean distance.

[0103] A3, calculate the average value of all data objects in each cluster and use the average value as the new cluster center of the cluster.

[0104] A4, loop A2 to A3 until the new cluster centers of each cluster obtained after two iterations are the same.

[0105] Optionally, the above method further includes:

[0106] S5, obtaining the number of times each electricity payment method is used from the electricity consumption data of each electricity customer, and determining the electricity payment method with the largest number of times for each electricity customer as the preferred channel; calculating the deviation of each electricity customer's preferred channel relative to the preferred channels of other electricity customers, and based on the deviation of each electricity customer, calculating the activity index corresponding to the preferred channel of each electricity customer.

[0107] Optionally, the deviation degree of one of the above-mentioned electricity customers is specifically:

[0108] in, In the formula, P represents the deviation of the preferred channel B of power customer A, X represents the number of times the preferred channel B of power customer A is used, represents the average number of times preferred channel B is used in the group of power customer A, σ represents the standard deviation of the number of times used in the group of power customer A, N represents the number of power customers in the group, X irepresents the number of times the i-th electricity customer in the group uses the preferred channel B; the activity index corresponding to the preferred channel of one electricity customer is specifically:

[0109] V=60+20*P, where V represents the activity index corresponding to the preferred channel B of electricity customer A, and P represents the deviation of the preferred channel B of electricity customer A.

[0110] Among them, the customer's channel preference is analyzed based on the customer's payment channel. Channel preference analysis is conducive to the precise marketing of the power grid, saving operation and marketing costs, reducing losses, and is conducive to the deployment of power grid service outlets to achieve the effect of precise marketing.

[0111] Optionally, the above method further includes:

[0112] S6, obtaining the electricity consumption of each season from the electricity consumption data of each electricity customer, and determining the season with the largest electricity consumption as the active season of each electricity customer.

[0113] Among them, analyzing the season with the highest electricity consumption based on users' electricity consumption behavior is conducive to grid load adjustment.

[0114] Embodiment 2:

[0115] The embodiment of the present application provides a system for classifying power customer value based on customer grouping, which is applied to a method for classifying power customer value based on customer grouping in any one of the embodiments 1, such as Figure 2 As shown, including:

[0116] The first module is used to obtain indicator data corresponding to preset indicators from the electricity consumption data of each electricity customer. The preset indicators include the electricity bill receivable indicator, the electricity consumption growth rate indicator and the arrears ratio indicator.

[0117] The second module is used to standardize the preset indicators respectively, and based on the preset indicators after standardization, use the entropy weight method to calculate the indicator weight of each indicator in the preset indicators.

[0118] The third module is used to calculate the value score of each electricity customer based on the indicator weights of each indicator in the preset indicators.

[0119] The fourth module is used to obtain the target value level of each electricity customer based on the preset value level categories according to the value score and using the improved k-means clustering algorithm.

[0120] Embodiment 3:

[0121] The present application embodiment provides an electronic device, such as Figure 3As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, any method in Embodiment 1 is implemented.

[0122] Embodiment 4:

[0123] An embodiment of the present application provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions enable a computer to execute any method in Example 1.

[0124] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for classifying power customer value based on customer grouping, characterized in that: The specific steps include: Obtaining index data corresponding to preset indexes from the electricity consumption data of each electricity customer, wherein the preset indexes include an index of electricity charges receivable, an index of electricity consumption growth rate, and an index of proportion of arrears; The preset indicators are standardized respectively, and based on the standardized preset indicators, the indicator weights of each indicator in the preset indicators are calculated by using the entropy weight method; Based on the indicator weights of each indicator in the preset indicators, the value score of each power customer is calculated; Based on the preset value level categories, the target value level of each electricity customer is obtained according to the value score and using an improved k-means clustering algorithm.

2. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The standardization process is specifically as follows: The indicators of electricity charges receivable and electricity consumption growth rate are as follows: Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data; The specific indicators for the proportion of arrears are as follows: Where, X x represents the index data after standardization, X represents the original index data without standardization, and X min Indicates the minimum value in the original indicator data, X max Indicates the maximum value in the original indicator data.

3. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The indicator weight of each indicator in the preset indicators is specifically: in, If p ij =0, then Where W j represents the indicator weight of the jth indicator in the preset indicators, k represents the total amount of indicators in the preset indicators, k = 3; E j represents the information entropy of the jth indicator in the preset indicators, Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer, and n represents the total number of power customers.

4. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The value score of the electricity customer is specifically: In the formula, F i represents the value score of electricity customer i, W j represents the index weight of the jth index in the preset index of power customer i, k represents the total amount of the index in the preset index, k=3; Y ij It represents the value of the jth indicator in the preset indicators of the ith power customer in the power consumption data of each power customer.

5. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The target value level of the electricity customer is obtained according to the value score and using the improved k-means clustering algorithm, specifically: A1, establish a corresponding number of clusters according to the number of value level categories, and apply the initial cluster center algorithm to determine the initial cluster center of each cluster; A2, calculate the Euclidean distance between the value score of each electricity customer and each initial cluster center, and assign each value score to the cluster with the closest Euclidean distance; A3, calculate the average value of all data objects in each cluster and use the average value as the new cluster center of the cluster; A4, loop A2 to A3 until the new cluster centers of each cluster obtained after two iterations are the same.

6. A method for classifying power customer value based on customer grouping according to claim 5, characterized in that: The initial cluster center algorithm is applied to determine the initial cluster center of each cluster, specifically: A11, determining the minimum value score of each power customer as the first target value; A12, calculating the Euclidean distance between the value scores of other electricity customers and the first target value, and arranging the value scores in ascending order according to the Euclidean distance; A13, in the sequence of value scores arranged in ascending order, the value score numbered 1+(i-1)(n-2) / (k-1) is assigned to the kth cluster, where i=1, 2...k, and n represents the total number of value scores; A14, for the value scores that are not classified into the cluster, calculate the Euclidean distance between each value score that is not classified into the cluster and each value score that is classified into the cluster, and classify each value score that is not classified into the cluster into the cluster with the smallest Euclidean distance; A15, calculate the average value of each value score in each cluster, and determine the average value as the initial cluster center.

7. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The method further comprises: The number of times of each electricity payment method is obtained from the electricity consumption data of each electricity customer, and the electricity payment method with the most times of each electricity customer is determined as the preferred channel; The deviation of each electricity customer's preferred channel relative to the preferred channels of other electricity customers is calculated, and based on the deviation of each electricity customer, the activity index corresponding to the preferred channel of each electricity customer is calculated.

8. A method for classifying power customer value based on customer grouping according to claim 1, characterized in that: The deviation of one of the electricity customers is specifically: in, In the formula, P represents the deviation of the preferred channel B of power customer A, X represents the number of times the preferred channel B of power customer A is used, represents the average number of times preferred channel B is used in the group of power customer A, σ represents the standard deviation of the number of times used in the group of power customer A, N represents the number of power customers in the group, X i represents the number of times the i-th electricity customer in the group uses the preferred channel B; The activity index corresponding to the preferred channel of one of the electricity customers is specifically: V=60+20*P, where V represents the activity index corresponding to the preferred channel B of electricity customer A, and P represents the deviation of the preferred channel B of electricity customer A.

9. The method for classifying power customer value based on customer grouping according to claim 1 is characterized in that: The method further comprises: The electricity consumption in each quarter is obtained from the electricity consumption data of each electricity customer, and the season with the largest electricity consumption is determined as the active season of each electricity customer.

10. A system for classifying power customer value based on customer grouping, applied to a method for classifying power customer value based on customer grouping as claimed in any one of claims 1 to 9, characterized in that: include: The first module is used to obtain index data corresponding to preset indicators from the electricity consumption data of each power customer, wherein the preset indicators include the electricity fee receivable index, the electricity consumption growth rate index and the arrears ratio index; The second module is used to perform standardization processing on the preset indicators respectively, and based on the preset indicators after standardization, calculate the indicator weight of each indicator in the preset indicators by using the entropy weight method; The third module is used to calculate the value score of each power customer based on the indicator weights of each indicator in the preset indicators; The fourth module is used to obtain the target value level of each power customer based on the preset value level categories according to the value score and using the improved k-means clustering algorithm.

Images