Charging station service fee pricing method, device, equipment and storage medium

Abstract

The present application provides a charging station service fee pricing method, device, equipment and storage medium, which can be used in the field of artificial intelligence technology. The method includes: obtaining historical charging feature data corresponding to at least one preset charging station; performing multi-level clustering processing on each preset charging station according to the various historical charging feature data corresponding to each preset charging station, so as to obtain the category corresponding to each preset charging station; performing counterfactual prediction on the correlation between the charging demand and pricing of each preset charging station in the current period according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous period in the historical time period and the preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current period; determining the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and pricing of each preset charging station in the current period and the pre-constructed target pricing model.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a charging station service fee pricing method, device, equipment and storage medium. Background Art

[0002] With the large-scale use of new energy vehicles, the construction of charging and swapping infrastructure is crucial. The core issue of the construction of charging and swapping infrastructure is the pricing of charging station service fees.

[0003] At present, when charging station service fees are priced, they are generally priced based on cost or based on the prices of surrounding competing charging stations. Therefore, the current pricing method is generally artificially priced based on certain factors, and the pricing is rarely changed after pricing.

[0004] Therefore, the current pricing method has low pricing flexibility. Moreover, the pricing is not based on the charging demand of the charging station. For the rare cases of pricing changes, it is also impossible to determine the correlation between charging demand and pricing in the charging station based on historical samples. As a result, it is impossible to provide charging stations with effective pricing references for charging demand, and it is also impossible for charging stations to balance charging efficiency and revenue to make more reasonable pricing. Summary of the invention

[0005] The present application provides a charging station service fee pricing method, device, equipment and storage medium to solve the technical problem in the prior art that when charging stations rarely change their pricing, it is impossible to determine the correlation between charging demand and pricing in the charging station based on historical samples, resulting in the inability to provide charging stations with effective pricing references regarding charging demand.

[0006] In a first aspect, the present application provides a charging station service fee pricing method, including:

[0007] Acquire historical charging characteristic data corresponding to at least one preset charging station, wherein the historical charging characteristic data includes: a plurality of historical charging non-price characteristic data and historical charging price characteristic data;

[0008] Perform multi-level clustering processing on each preset charging station according to a variety of historical charging feature data corresponding to each preset charging station to obtain a category corresponding to each preset charging station;

[0009] The following operations are performed for the two charging user types respectively to complete the current time period charging pricing corresponding to each charging user type respectively. The charging user types include: car owner type and non-car owner type. The following operations include:

[0010] According to the categories corresponding to the preset charging stations, the observation samples of the charging user types in the previous period in the historical time period corresponding to the preset charging stations, and the preset counterfactual prediction model, a counterfactual prediction is performed on the correlation between the charging demand and the pricing of the current period of each preset charging station, so as to obtain the correlation between the charging demand and the pricing of the current period of each preset charging station;

[0011] The optimal pricing of each of the preset charging stations in the current period is determined based on the correlation between the charging demand and pricing of each of the preset charging stations in the current period and a pre-constructed target pricing model, wherein the target pricing model includes a target charging pile effect sub-model and a target revenue sub-model.

[0012] In a possible design, the multiple historical charging non-price characteristic data include: historical charging pile efficiency characteristic data, historical charging user types and historical power types;

[0013] The multi-level clustering process is performed on each preset charging station according to the various historical charging feature data corresponding to each preset charging station to obtain the category corresponding to each preset charging station, including:

[0014] Performing a first-level clustering process on each preset charging station using the historical power type corresponding to each preset charging station to obtain a power category corresponding to each preset charging station;

[0015] The historical pile efficiency characteristic data corresponding to each preset charging station is used to perform a second-level clustering process on the preset charging stations in each power category to obtain the pile efficiency category of each preset charging station under each power category;

[0016] Calculate the historical vehicle owner ratio corresponding to each preset charging station based on the historical charging user type corresponding to each preset charging station;

[0017] The third-level clustering process is performed on the preset charging stations in each pile effect category using the historical car owner ratio corresponding to each preset charging station, so as to obtain the car owner ratio category of each preset charging station under the pile effect category.

[0018] In a possible design, the counterfactual prediction of the correlation between the charging demand and pricing of each preset charging station in the current period is performed according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current period, including:

[0019] Constructing a first observation sample of the charging user type in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the first observation sample includes: the charging amount of the preset charging station in the previous period, the average price of the preset charging station in the previous period in the historical time period, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous period;

[0020] Constructing a second observation sample of the charging user type in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the second observation sample includes: the average charging amount of the preset charging station in the historical time period, the category corresponding to the preset charging station, and the price change rate between the preset charging station and the surrounding competing charging stations in the previous period;

[0021] The correlation between the charging demand and pricing of each preset charging station during the current period is determined based on the first observation sample and the second observation sample corresponding to each preset charging station and a preset counterfactual prediction model.

[0022] In a possible design, the preset counterfactual prediction model includes a preset benchmark charging demand sub-model and a preset quantity-price sub-model; the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity;

[0023] The determining of the correlation between the charging demand and pricing of each preset charging station in the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model includes:

[0024] Determine the benchmark parameters of the current time period in the benchmark charging demand submodel according to the first observation samples corresponding to each preset charging station, so as to determine the benchmark charging demand submodel of the current time period;

[0025] Determine the quantity and price parameters of the current period in the quantity and price sub-model according to the second observation samples corresponding to each preset charging station, so as to determine the quantity and price sub-model of the current period, wherein the quantity and price parameters of the current period include: a price elasticity parameter of the current period;

[0026] The correlation between the charging demand and price elasticity of each preset charging station in the current period is determined according to the benchmark charging demand sub-model of the current period corresponding to each preset charging station and the quantity-price sub-model of the current period.

[0027] In a possible design, determining the optimal pricing of each of the preset charging stations in the current period according to the correlation between the charging demand and pricing of each of the preset charging stations in the current period and a pre-built target pricing model includes:

[0028] The target charging pile efficiency sub-model and the target revenue sub-model are converted according to the correlation between the charging demand and pricing of each of the preset charging stations in the current period, so as to obtain the target pricing model of each of the preset charging stations in the current period;

[0029] The target pricing model for the current time period pricing of each of the preset charging stations is solved to obtain the optimal pricing of each of the preset charging stations in the current time period.

[0030] In a possible design, the target pricing model is represented by a linear weighting method of the target pile effect sub-model and the target income sub-model;

[0031] Solving the target pricing model for the current time period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current time period includes:

[0032] A linear weighted method is used to convert a multi-objective optimization problem corresponding to a target pricing model for solving the current period pricing of each of the preset charging stations into a single-objective optimization problem;

[0033] The optimal solution of the single-objective optimization problem of each of the preset charging stations is obtained, and each of the optimal solutions is determined as the optimal pricing of each of the preset charging stations in the current time period.

[0034] In a possible design, after determining the optimal pricing of each of the preset charging stations in the current period according to the correlation between the charging demand and pricing of each of the preset charging stations in the current period and the pre-built target pricing model, the method further includes:

[0035] The optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type are sent to the user terminal, so that the user terminal displays the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type.

[0036] In a second aspect, the present application provides a charging station service fee pricing device, comprising:

[0037] An acquisition module, used to acquire historical charging characteristic data corresponding to at least one preset charging station, wherein the historical charging characteristic data includes: a plurality of historical charging non-price characteristic data and historical charging price characteristic data;

[0038] A clustering processing module, used to perform multi-level clustering processing on each preset charging station according to a variety of historical charging characteristic data corresponding to each preset charging station, so as to obtain a category corresponding to each preset charging station;

[0039] An execution module is used to perform the following operations for two charging user types respectively, so as to complete the current time period charging pricing corresponding to each charging user type respectively. The charging user types include: car owner type and non-car owner type. The execution module includes:

[0040] A prediction submodule, configured to perform a counterfactual prediction on the correlation between the charging demand and pricing of each preset charging station in the current period according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous period in the historical time period corresponding to each preset charging station, and a preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current period;

[0041] A determination submodule determines the optimal pricing of each of the preset charging stations in the current period according to the correlation between the charging demand and pricing of each of the preset charging stations in the current period and a pre-constructed target pricing model, wherein the target pricing model includes a target charging pile effect submodel and a target revenue submodel.

[0042] In a third aspect, the present application provides a method comprising: a processor, and a memory communicatively connected to the processor;

[0043] The memory stores computer-executable instructions;

[0044] The processor executes the computer-executable instructions stored in the memory to implement the method as described in any one of the first aspects.

[0045] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method as described in any one of the first aspects.

[0046] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0047] The charging station service fee pricing method, device, equipment and storage medium provided in the present application obtain historical charging feature data corresponding to at least one preset charging station, wherein the historical charging feature data includes: a variety of historical charging non-price feature data and historical charging price feature data; perform multi-level clustering processing on each preset charging station according to the multiple historical charging feature data corresponding to each preset charging station to obtain the category corresponding to each preset charging station; perform the following operations respectively for two charging user types to complete the current time period charging pricing corresponding to each charging user type respectively, wherein the charging user types include: car owner type and non-car owner type, and the following operations The operation includes: counterfactual prediction of the correlation between the charging demand and pricing of each preset charging station in the current period according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current period; determine the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and pricing of each preset charging station in the current period and the pre-built target pricing model, and the target pricing model includes a target pile effect sub-model and a target revenue sub-model. Since the server performs counterfactual prediction of the correlation between the charging demand and pricing of each preset charging station in the current period according to the category corresponding to the preset charging station, the observation sample of the charging user type in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, the correlation between the charging demand and pricing of the preset charging station in the current period is obtained, thereby determining the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and pricing of each preset charging station in the current period and the pre-built target pile effect sub-model and target revenue sub-model. Therefore, through counterfactual prediction, this scheme can determine the correlation between charging demand and pricing in charging stations based on historical samples when facing the situation that charging stations rarely make pricing changes. Then, effective pricing can be made with reference to charging demand, so that charging stations can take into account both charging efficiency and revenue for more reasonable pricing. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0049] Figure 1 An application scenario diagram of a charging station service fee pricing method provided in one embodiment of the present application;

[0050] Figure 2 A flowchart of a charging station service fee pricing method provided in an embodiment of the present application;

[0051] Figure 3 A flowchart of a charging station service fee pricing method provided in another embodiment of the present application;

[0052] Figure 4 A schematic diagram of multi-level clustering processing provided in an embodiment of the present application;

[0053] Figure 5 A flowchart of a charging station service fee pricing method provided in another embodiment of the present application;

[0054] Figure 6 A schematic diagram of the structural framework for pricing charging station service fees provided for this application;

[0055] Figure 7 A schematic diagram of the structure of a charging station service fee pricing device provided in one embodiment of the present application;

[0056] Figure 8 A schematic diagram of the structure of an electronic device provided in one embodiment of the present application.

[0057] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0058] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0059] In order to clearly understand the technical solution of the present application, the solution of the prior art is first introduced in detail.

[0060] At present, when pricing charging station service fees, it is generally priced by cost or based on the prices of surrounding competing charging stations. Therefore, the current pricing method is generally artificially priced based on certain factors, and there are very few changes in pricing after pricing. Moreover, there are few studies on the pricing of charging service fees at home and abroad, and most of them are just analytical strategies, lacking specific mathematical models. The pricing of charging station service fees is mainly concerned with price optimization. The main challenge of this problem is that the prices of most charging stations do not change frequently, or even never change. Therefore, the current pricing method has low pricing flexibility. And it is not based on the charging demand of the charging station. For the rare case of pricing changes, it is also impossible to determine the correlation between charging demand and pricing in the charging station based on historical samples, so it is impossible to provide charging stations with effective pricing references for charging demand, and it is also impossible for charging stations to take into account charging efficiency and revenue for more reasonable pricing.

[0061] Therefore, in the face of technical problems in the prior art, in order to enable pricing to effectively refer to charging demand and truly perform pricing based on charging demand, the correlation between charging demand and pricing can be obtained based on counterfactual prediction. And because the types of charging users are different, the historical pricing and charging conditions are also different, so the correlation between charging demand and pricing can be obtained based on counterfactual prediction for the two types of charging users respectively. Specifically, this scheme performs counterfactual prediction on the correlation between charging demand and pricing in the current period of each preset charging station through the categories corresponding to the preset charging stations, the observation samples of charging user types in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, and obtains the correlation between charging demand and pricing in the current period of the preset charging station, so as to determine the optimal pricing of each preset charging station in the current period according to the correlation between charging demand and pricing in the current period of each preset charging station and the pre-constructed target pile effect sub-model and target income sub-model, so as to balance charging efficiency and income for more reasonable pricing. Moreover, after determining the optimal pricing, the price can be adjusted more flexibly based on the optimal pricing, making the pricing more flexible.

[0062] The application scenarios of the embodiments of the present invention are introduced below.

[0063] Figure 1 This is an application scenario diagram of a charging station service fee pricing method provided in one embodiment of the present application, such as Figure 1As shown, the specific application scenario of the present application can be a scenario of intelligent pricing of service fees when providing charging and battery replacement services to new energy vehicles. The system corresponding to the charging station service fee pricing method provided in the present application includes: user terminal 1, electronic equipment corresponding to charging station service fee pricing, such as server 2. The user sends a pricing request for the current period to server 2 through user terminal 1, and the identification of the preset charging station and the current period can be included in the pricing request. Based on the current pricing request, server 2 obtains the category of the preset charging station, and based on the category corresponding to the preset charging station, the observation sample of the charging user type in the previous period in the historical time period, and the preset counterfactual prediction model, the correlation between the charging demand and pricing of the preset charging station in the current period is predicted by counterfactual prediction, and the correlation between the charging demand and pricing of the preset charging station in the current period is obtained. According to the correlation between the charging demand and pricing of the preset charging station in the current period and the pre-built target pricing model, the optimal pricing of the preset charging station in the current period is determined, and finally server 2 sends the optimal pricing of the preset charging station in the current period to user terminal 1.

[0064] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0065] Figure 2 A flowchart of a charging station service fee pricing method provided in an embodiment of the present application, such as Figure 2 As shown, the charging station service fee pricing method provided in this embodiment includes the following steps:

[0066] Step 201 : acquiring historical charging characteristic data corresponding to at least one preset charging station, wherein the historical charging characteristic data includes: a plurality of historical charging non-price characteristic data and historical charging price characteristic data.

[0067] Among them, the preset charging stations refer to the charging stations that need to be priced.

[0068] The historical charging characteristic data refers to the characteristic data of the charging behavior that has occurred, which can be obtained from the charging flow data and the background charging station information management system.

[0069] Among them, the various historical charging non-price characteristic data may include historical charging pile efficiency characteristic data, historical charging user types and historical power types, charging station attribute characteristic data, time characteristic data, etc.

[0070] Among them, the historical charging price characteristic data includes charging station price characteristic data.

[0071] Among them, the historical charging pile efficiency characteristic data can be expressed by the daily charging volume of a single charging gun at the charging station, the historical charging user types include car owner type and non-car owner type, the historical power types include high power and low power, the charging station attribute characteristic data is a high power station or a low power station, and the time characteristic data may include the previous period in the historical time period.

[0072] Specifically, in this embodiment, the server establishes a communication relationship with the charging flow data system and the background charging station information management system, and obtains historical charging characteristic data from the charging flow data and the background charging station information management system.

[0073] Among them, by associating the charging flow data with the information in the background charging station information management system, a variety of historical charging non-price characteristic data and historical charging price characteristic data can be obtained.

[0074] Step 202 : performing multi-level clustering processing on each preset charging station according to a variety of historical charging characteristic data corresponding to each preset charging station, so as to obtain a category corresponding to each preset charging station.

[0075] Among them, multi-level clustering processing refers to the process of dividing a collection of physical or abstract objects into multiple classes consisting of similar objects.

[0076] Among them, the category corresponding to each preset charging station refers to the multi-level category composed of clusters of different levels into which each preset charging station i is divided after multi-level clustering processing, denoted as category i , each cluster corresponds to a category.

[0077] Specifically, in this embodiment, the server assigns each preset charging station i to a single cluster. After the server device obtains a variety of historical charging non-price feature data corresponding to each preset charging station, the server device selects the historical charging pile efficiency feature data, the historical power type and the historical charging user type from the multiple historical charging non-price feature data as clustering features, and divides the clusters into different clusters in the order of "historical power type-historical charging pile efficiency feature data-owner ratio", forming a multi-level category, recorded as category i .

[0078] Among them, the historical charging pile efficiency characteristic data refers to the daily charging volume of a single gun at a charging station.

[0079] The historical power type refers to the type of charging power at the charging station, such as high power and low power.

[0080] Among them, historical charging user types are divided into car owners and non-car owners. Car owners are car owners who use cars from this manufacturer, and non-car owners are car owners who do not use cars from this manufacturer.

[0081] Among them, the proportion of car owners refers to the proportion of the number of car owners to the total number of users.

[0082] Step 203, performing the following operations for the two charging user types respectively, so as to complete the current time period charging pricing corresponding to each charging user type respectively. The charging user types include: car owner type and non-car owner type. The following operations include:

[0083] Step 2031, based on the categories corresponding to each preset charging station, the observation samples of charging user types in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, a counterfactual prediction is performed on the correlation between the charging demand and pricing in the current period of each preset charging station to obtain the correlation between the charging demand and pricing in the current period of each preset charging station.

[0084] There are two types of charging users, namely, the car owner type and the non-car owner type. Since the historical charging characteristic data corresponding to the car owner type and the non-car owner type are quite different, and pricing for the car owner type and the non-car owner type separately is also a pricing requirement of the preset charging station, step 2031 and step 2032 are respectively performed for these two different types of charging users.

[0085] The current period is represented by t+1, and the previous period in the historical time period is represented by t.

[0086] The observation samples include a first observation sample and a second observation sample. The observation samples are obtained based on historical charging characteristic data of a preset charging station.

[0087] The preset counterfactual prediction model is used to make a counterfactual prediction of the correlation between charging demand and pricing at each preset charging station in the current period t+1. Counterfactual prediction is a prediction method based on machine learning algorithms, which aims to infer the opposite results that may occur under different decision conditions by analyzing and modeling existing data.

[0088] Specifically, in this embodiment, the charging user types are first divided into car owner types and non-car owner types, and the charging pricing for the current time period t+1 corresponding to the charging user type is completed for the car owner type and the non-car owner type respectively.

[0089] For charging users of the car owner type, the server obtains the category cate corresponding to each preset charging station after multi-level clustering processing iand the observation samples of the charging user types in the previous time period t in the historical time period corresponding to each preset charging station, obtain a pre-built preset counterfactual prediction model, solve the preset counterfactual prediction model through the categories corresponding to each preset charging station and the observation samples of the charging user types in the previous time period in the historical time period corresponding to each preset charging station, so as to make a counterfactual prediction on the correlation between the charging demand and pricing in the current time period t+1 of each preset charging station, thereby obtaining the correlation between the charging demand and pricing in the current time period t+1 of each preset charging station.

[0090] For charging users who are not car owners, a similar solution to that for charging users who are not car owners is implemented, which will not be described in detail here.

[0091] Step 2032, determining the optimal pricing of each preset charging station in the current period based on the correlation between charging demand and pricing of each preset charging station in the current period and a pre-built target pricing model, wherein the target pricing model includes a target charging pile efficiency sub-model and a target revenue sub-model.

[0092] Among them, the target pricing model is used to determine the optimal pricing model for each preset charging station in the current period t+1.

[0093] Among them, the target pricing model includes the target pile effect sub-model and the target profit sub-model.

[0094] Among them, the target pile effect sub-model is represented by f i E (P it ) represents the charging efficiency and the pricing P of charging station i in the previous period t. it The correlation relationship between them.

[0095] Among them, the target return sub-model is represented by f i P (P it ) represents the revenue and the price P of charging station i in the previous period t. it The correlation relationship between them.

[0096] Specifically, in this embodiment, the target pile effect sub-model f i E (P it ) can be expressed as the ratio of charging demand to the number of charging piles. The target revenue sub-model f i P (P it ) can be expressed as the difference between the product of pricing and charging demand and cost, so the target charging efficiency sub-model f in the target pricing model i E (P it ) and target return sub-model f iP (P it ) are all correlated with charging demand, and the target pricing model can be represented by charging demand. The correlation between charging demand and pricing can be obtained through the counterfactual prediction model, so that the target pricing model can be represented by pricing. Solving the target pricing model can obtain the optimal pricing of each preset charging station in the current period.

[0097] The charging station service fee pricing method provided in the present embodiment obtains historical charging feature data corresponding to at least one preset charging station, the historical charging feature data including: a plurality of historical charging non-price feature data and historical charging price feature data; performs multi-level clustering processing on each preset charging station according to the plurality of historical charging feature data corresponding to each preset charging station, so as to obtain the category corresponding to each preset charging station; performs the following operations respectively for two charging user types, so as to complete the charging pricing for the current time period corresponding to each charging user type respectively, the charging user types including: the owner type and the non-owner type, the following operations including: performs counterfactual prediction on the correlation between the charging demand and pricing of each preset charging station in the current time period according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous time period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current time period; determines the optimal pricing of each preset charging station in the current time period according to the correlation between the charging demand and pricing of each preset charging station in the current time period and the pre-constructed target pricing model, the target pricing model including the target pile effect sub-model and the target revenue sub-model. Since the server performs counterfactual predictions on the correlation between charging demand and pricing in the current period of each preset charging station based on the categories corresponding to the preset charging stations, the observation samples of charging user types in the previous period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, the correlation between charging demand and pricing in the current period of the preset charging station is obtained, and the optimal pricing of each preset charging station in the current period is determined based on the correlation between charging demand and pricing in the current period of each preset charging station and the pre-constructed target pile effect sub-model and target revenue sub-model. Therefore, through counterfactual predictions, this solution can also determine the correlation between charging demand and pricing in charging stations based on historical samples when facing the situation where charging stations rarely make pricing changes, and then can refer to charging demand for effective pricing, so that charging stations can take into account both charging efficiency and revenue for more reasonable pricing.

[0098] Figure 3 A flowchart of a charging station service fee pricing method provided in another embodiment of the present application, such as Figure 3 As shown, the various historical charging non-price characteristic data include: historical charging pile efficiency characteristic data, historical charging user types and historical power types.

[0099] Accordingly, step 202 includes the following steps:

[0100] Step 2021 : performing a first-level clustering process on each preset charging station using the historical power type corresponding to each preset charging station to obtain a power category corresponding to each preset charging station.

[0101] Among them, the first level of clustering processing refers to first performing clustering processing according to the first historical charging non-price feature data in the clustering order.

[0102] Specifically, in this embodiment, the clustering order is "historical power type-historical charging pile efficiency feature data-owner ratio". The first historical charging non-price feature data in the clustering order is the historical power type. Exemplarily, the power categories of the charging stations can be clustered at the first level according to 100W, 200W, and 300W. Each charging station corresponds to a power category, so that the power category corresponding to each preset charging station can be obtained.

[0103] Step 2022: Perform a second-level clustering process on the preset charging stations in each power category using the historical charging pile efficiency feature data corresponding to each preset charging station, so as to obtain the charging pile efficiency category of each preset charging station in each power category.

[0104] Among them, the second-level clustering processing refers to clustering processing based on the first-level clustering processing and the second historical charging non-price feature data in the clustering order.

[0105] Specifically, in this embodiment, the clustering order is "historical power type-historical pile efficiency feature data-owner ratio", and the second historical charging non-price feature data in the clustering order is the historical pile efficiency feature data. For example, under the 100W power category, the charging station pile efficiency category can be clustered at the second level according to 100kW·h, 200kW·h, 300kW·h... Each charging station corresponds to a pile efficiency category, so that the pile efficiency category of each preset charging station under each power category can be obtained. It is worth noting that the second-level clustering method under other power categories is similar and will not be repeated here.

[0106] Step 2023: Calculate the historical vehicle owner ratio corresponding to each preset charging station according to the historical charging user type corresponding to each preset charging station.

[0107] Among them, the historical car owner ratio refers to the proportion of car owners in each preset charging station over a period of history to the total number of users.

[0108] Specifically, in this embodiment, the server device obtains the historical charging user types corresponding to each preset charging station, calculates the number of historical car owners, that is, the total number of users, and calculates the ratio of the number of historical car owners to the total number of users as the historical car owner ratio.

[0109] Step 2024 , using the historical vehicle owner ratio corresponding to each preset charging station, perform third-level clustering processing on the preset charging stations in each charging pile effect category to obtain the vehicle owner ratio category of each preset charging station under the charging pile effect category.

[0110] Among them, the third-level clustering processing refers to clustering processing based on the second-level clustering processing and the third historical charging non-price feature data in the clustering order.

[0111] Specifically, in this embodiment, the clustering order is "historical power type-historical pile efficiency feature data-owner ratio", and the third historical charging non-price feature data in the clustering order is the owner ratio. For example, under the 100kW·h pile efficiency category, the charging station owner ratio category can be clustered at the third level according to 1:2, 1:3, 1:4... Each charging station corresponds to a car owner ratio category, so that the owner ratio category of each preset charging station under the pile effect category can be obtained. It is worth noting that the third-level clustering method under other pile effect categories is similar and will not be repeated here.

[0112] For example, Figure 4 A schematic diagram of multi-level clustering processing provided by an embodiment of the present application is shown in FIG. Figure 4As shown, assuming that there are 750 charging stations in total, the server assigns the 750 preset charging stations to a single cluster, and divides the cluster into different clusters in the order of "historical power type-historical charging pile efficiency feature data-owner ratio". The charging station power categories are clustered at the first level according to 100W, 200W, and 300W, respectively. Among them, 300 charging stations belong to the 100W power category, 300 charging stations belong to the 200W power category, and 150 charging stations belong to the 300W power category. Each charging station corresponds to a power category. Subsequently, under the 100W power category, the charging station charging pile efficiency categories can be clustered at the second level according to 100kW·h, 200kW·h, 300kW·h..., among which 60 charging stations belong to the 100W power category. 0kW·h pile efficiency category, 60 charging stations belong to 200kW·h pile efficiency category, 60 charging stations belong to 300kW·h pile efficiency category… Each charging station corresponds to a pile efficiency category, and the second-level clustering processing under other power categories is the same. Finally, under the 100kW·h pile efficiency category, the charging station owner ratio categories can be clustered according to 1:2, 1:3, 1:4… for the third-level clustering processing. Among them, 20 charging stations belong to the 1:2 owner ratio category, 20 charging stations belong to the 1:3 owner ratio category, and 20 charging stations belong to the 1:4 owner ratio category… Each charging station corresponds to a car owner ratio category, and the third-level clustering processing under other pile efficiency categories is the same. Finally, each charging station i corresponds to a category, denoted as category i For example, the cate corresponding to charging station 1 i The ratio is: "charging station power 100W - charging station pile efficiency 100kW·h - charging station owner ratio 1:2", where cate iIt can be represented by a three-dimensional vector. In the charging station service fee pricing method provided in this embodiment, the various historical charging non-price feature data include: historical pile effect feature data, historical charging user type and historical power type. Accordingly, when performing multi-level clustering processing on each preset charging station according to the various historical charging feature data corresponding to each preset charging station to obtain the category corresponding to each preset charging station, the historical power type corresponding to each preset charging station is used to perform the first-level clustering processing on each preset charging station to obtain the power category corresponding to each preset charging station; the historical pile effect feature data corresponding to each preset charging station is used to perform the second-level clustering processing on the preset charging stations in each power category to obtain the pile effect category of each preset charging station under each power category; the historical car owner ratio corresponding to each preset charging station is calculated according to the historical charging user type corresponding to each preset charging station; the historical car owner ratio corresponding to each preset charging station is used to perform the third-level clustering processing on the preset charging stations in each pile effect category to obtain the car owner ratio category of each preset charging station under the pile effect category. Due to the multi-level clustering processing of each preset charging station, the charging stations can be classified in more detail, so that more detailed categories corresponding to each preset charging station can be obtained, and then the charging demand of each preset charging station under the corresponding category can be determined, so that the final calculated optimal pricing is more accurate.

[0113] As an optional implementation, based on any of the above embodiments, in this embodiment, step 2031 includes the following steps:

[0114] Step 301, construct a first observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, the first observation sample includes: the charging amount of the preset charging station in the previous period, the average price of the preset charging station in the previous period of the historical time period, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous period.

[0115] The first observed sample is used to determine a sample of a preset benchmark charging demand sub-model.

[0116] Among them, the charging amount of the preset charging station in the previous period is recorded as Specifically represents the charging amount of charging station i in the previous period t. Distinguish between car owners and non-car owners, among which, represents the charging amount of the owner in the charging station i in the previous period t, represents the charging amount of non-owners in charging station i during period t.

[0117] Among them, the average price of the preset charging station in the previous period of the historical time period is recorded as Specifically, it represents the average price of charging station i in the previous period t during the historical period.

[0118] Among them, the average charging capacity of the preset charging station category in the historical time period is recorded as Specifically, it indicates the average charging amount of the charging station category corresponding to charging station i in the previous time period t during the historical time period, wherein the charging station category is obtained when multi-level clustering processing is performed on the preset charging stations.

[0119] Among them, the non-price characteristics of the preset charging station in the previous period are recorded as X it The non-price characteristics of the preset charging station in the previous period include charging station attribute characteristic data, time characteristic data, historical charging pile efficiency characteristic data, historical charging user type and historical power type, etc.

[0120] Specifically, in this embodiment, the server calculates the charging capacity of the preset charging station in the previous period t The average price of the preset charging station in the previous period t in the historical time period And the average charging capacity of the preset charging station category in the historical time period At the same time, obtain the non-price characteristics X of the preset charging station in the previous period t it , according to the type of each preset charging station, the charging amount of the preset charging station in the previous period t The average price of the preset charging station in the previous period t in the historical time period The average charging capacity of the preset charging station category during the historical period and the non-price characteristics X of the pre-set charging station in the previous period t it The first observation sample of the charging user type in the previous period t in the historical time period corresponding to the preset charging station is constructed, and the first observation sample includes: X it .

[0121] Step 302, construct a second observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, the second observation sample includes: the average charging amount of the station category to which the preset charging station belongs in the historical time period, the category corresponding to the preset charging station, and the price change rate between the preset charging station and the surrounding competing charging stations in the previous period.

[0122] Among them, the second observation sample is used to determine the sample of the preset quantity and price sub-model.

[0123] The categories corresponding to the preset charging stations are obtained when multi-level clustering is performed on each preset charging station.

[0124] Among them, the price change rate of the preset charging station and the surrounding competing charging stations in the previous period is recorded as in, represents the average selling price of the preset charging station i in the previous period t, Represents the average selling price of surrounding competing charging stations in the previous period t.

[0125] Specifically, in this embodiment, the server calculates the average charging capacity of the preset charging station in the historical time period. And the price change rate ΔP of the preset charging station and the surrounding competing charging stations in the previous period it At the same time, when performing multi-level clustering processing on each preset charging station, the category corresponding to the preset charging station is obtained. i , and based on the average charging capacity of the preset charging station category within the historical time period Category of preset charging station i , the price change rate ΔP of the preset charging station and the surrounding competing charging stations in the previous period it Construct a second observation sample of the charging user type in the previous period of the historical time period corresponding to each preset charging station, and the second observation sample includes: cate i , ΔP it .

[0126] Step 303 : determining the correlation between the charging demand and pricing of each preset charging station during the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model.

[0127] Among them, the preset counterfactual prediction model can be expressed as shown in formula (1):

[0128]

[0129] Among them, g(P it ,cate i ,θ) is the preset quantity-price sub-model, It is the preset quantity-price sub-model under the historical average selling price of station i in period t within one month. It is a preset benchmark charging demand sub-model.

[0130] Specifically, in this embodiment, the preset counterfactual prediction model can be obtained by substituting the first observation sample into the preset benchmark charging demand sub-model and substituting the second observation sample into the preset quantity-price sub-model. The preset counterfactual prediction model represents the correlation between the charging demand and pricing of each preset charging station. The server obtains the correlation between the charging demand and pricing of each preset charging station in the current period t+1 based on the preset counterfactual prediction model.

[0131] The charging station service fee pricing method provided in this embodiment performs counterfactual prediction on the correlation between charging demand and pricing of each preset charging station in the current time period according to the category corresponding to each preset charging station, the observation sample of charging user type in the previous time period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model to obtain the correlation between charging demand and pricing of each preset charging station in the current time period. A first observation sample of charging user type in the previous time period in the historical time period corresponding to each preset charging station is constructed according to the category corresponding to each preset charging station. The first observation sample includes: the charging amount of the preset charging station in the previous time period, the charging amount of the preset charging station in the previous time period in the historical time period, and the charging amount of the preset charging station in the previous time period in the historical time period. The first observation sample and the second observation sample are used to determine the correlation between the charging demand and pricing of each preset charging station in the current period. The first observation sample and the second observation sample are used to determine the correlation between the charging demand and pricing of each preset charging station in the current period. By constructing the preset counterfactual prediction model, even if the pricing rarely changes, the correlation between the charging demand and pricing of each preset charging station in the current period can still be predicted from the observation sample, enriching the observation sample data, and obtaining the correlation between the charging demand and pricing of each preset charging station in the current period.

[0132] As an optional implementation, in this embodiment, the preset counterfactual prediction model includes a preset baseline charging demand sub-model and a preset quantity-price sub-model; the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity.

[0133] Among them, the preset benchmark charging demand sub-model is denoted as The following describes the process of determining the preset benchmark charging demand sub-model. First, the prediction target of the preset benchmark charging demand sub-model is determined. The prediction target includes the charging demand of car owners and non-car owners at different time periods of the charging station, which are respectively denoted as Secondly, a regression model is constructed based on the extreme gradient boosting algorithm (Extreme Gradient Boosting, abbreviated as XGB), which can be expressed as shown in formula (2):

[0134]

[0135] The mathematical symbols have the following meanings:

[0136] Represents the historical average selling price of charging station i in the previous period t during the historical time period.

[0137] It represents the charging amount of charging station i in the previous period t. Distinguish between car owners and non-car owners, among which, represents the charging amount of the owner in the charging station i in the previous period t, represents the charging amount of non-owners in charging station i during period t.

[0138] It represents the average charging amount of the charging station category corresponding to charging station i in the previous period t during the historical time period.

[0139] represents the charging capacity of charging station i in the current period t+1, and Distinguish between car owners and non-car owners, among which, represents the charging amount of the owner in the charging station i in the current period t+1, It represents the charging amount of non-owners in charging station i in the current period t+1.

[0140] It represents the average charging amount of the charging station category corresponding to charging station i in the current period t+1 during the historical time period.

[0141] It represents the historical average selling price of charging station i in the current period t+1 during the historical time period.

[0142] X i,t+1 , represents non-price characteristics, including charging station attribute characteristic data, time characteristic data, historical charging pile efficiency characteristic data, historical charging user types and historical power types, etc.

[0143] Among them, X i,t+1 It is obtained by performing unique-hot processing on multiple non-price features such as observation samples, charging station attribute feature data, time feature data, historical charging pile efficiency feature data, historical charging user types and historical power types. It is a multi-dimensional vector.

[0144] Specifically, the server obtains X by performing unique heat treatment on multiple non-price features such as the historical average selling price of charging station i in the previous period t, the charging amount of charging station i in the previous period t, charging station attribute feature data, time feature data, historical pile efficiency feature data, historical charging user type and historical power type. it , calculated based on the first observation sample and Ratio Multiple observation sample data and X it Multiple known points can be determined, and the known points can be fitted through the fitting algorithm to obtain the values ​​of the benchmark parameters, and then the preset benchmark charging demand sub-model can be obtained.

[0145] Among them, the preset quantity-price sub-model is denoted as g(P it ,cate i ,θ), using charging demand and price elasticity to quantitatively measure the sensitivity of charging demand to price, and defining charging demand as a function based on price elasticity, which can be expressed as shown in formula (3):

[0146]

[0147] Taking logarithmic transformation on both sides of formula (3) yields formula (4):

[0148]

[0149] The mathematical symbols have the following meanings:

[0150] represents the price elasticity parameter, where θ2∈R m .

[0151] Among them, θ1, θ2, cate i It is obtained by performing unique heat treatment on the observed samples, historical power types, historical pile efficiency feature data, and historical charging user types. i is a three-dimensional vector.

[0152] Among them, P it To express ΔP it , we can get the functional relationship between charging demand and price, as shown in formula (3).

[0153] Specifically, the server aggregates the data to classify all charging stations. iThe server obtains the price elasticity of multiple charging stations by performing unique heat treatment on the historical average selling price of charging station i in the previous period t, the charging volume of charging station i in the previous period t, the historical pile efficiency feature data, the historical power type and the historical charging user type. i , calculate the average charging capacity of the charging station category corresponding to charging station i in the previous period t P in multiple observation sample data it and Multiple known points can be determined, and the known points can be fitted by the fitting algorithm to obtain the value of the quantity and price parameters, and then the functional relationship between charging demand and price can be obtained, as shown in formula (3). Then, the logarithmic transformation of both sides of formula (3) can be obtained to obtain the quantity and price sub-model g(P it ,cate i ,θ), quantity-price sub-model g(P it ,cate i ,θ) is shown in formula (4).

[0154] Specifically, in this embodiment, by presetting the benchmark charging demand sub-model And the preset quantity-price sub-model g(P it ,cate i ,θ) constructs a preset counterfactual prediction model.

[0155] Accordingly, step 303 includes the following steps:

[0156] Step 3031 : determining the benchmark parameters of the current time period in the benchmark charging demand submodel according to the first observation samples corresponding to each preset charging station, so as to determine the benchmark charging demand submodel of the current time period.

[0157] The reference parameter is the coefficient of the polynomial obtained by the fitting algorithm, and the reference parameter is a known constant.

[0158] Among them, the baseline charging demand sub-model is only affected by non-price characteristics, so it is used to predict the baseline charging demand without price intervention.

[0159] Specifically, in this embodiment, the server obtains X by performing unique heat processing on multiple non-price features such as the historical average selling price of charging station i in the previous period t, the charging amount of charging station i in the previous period t, charging station attribute feature data, time feature data, historical charging pile efficiency feature data, historical charging user type and historical power type. it ,calculate and Ratio Multiple observation sample data and Xit Multiple known points can be determined, and the known points can be fitted through the fitting algorithm to obtain the values ​​of the benchmark parameters, and then the preset benchmark charging demand sub-model can be obtained. When the previous period t changes to the current period t+1, the preset benchmark charging demand sub-model for the current period t+1 is obtained

[0160] Step 3032, determining the quantity and price parameters of the current period in the quantity and price sub-model according to the second observation samples corresponding to each preset charging station, so as to determine the quantity and price sub-model of the current period, wherein the quantity and price parameters of the current period include: the price elasticity parameter of the current period.

[0161] Among them, the quantity and price parameters include θ1, θ2, cate i and c, the quantity and price parameters are obtained by analyzing P in multiple observation sample data. it and Obtained by fitting processing.

[0162] Among them, the price elasticity parameter of the current period refers to Among them, θ2∈R m .

[0163] Specifically, in this embodiment, the server aggregates the data into categories of all charging stations. i The server obtains the price elasticity of multiple charging stations by performing unique heat treatment on the historical average selling price of charging station i in the previous period t, the charging volume of charging station i in the previous period t, the historical pile efficiency feature data, the historical power type and the historical charging user type. i , calculate the average charging capacity of the charging station category corresponding to charging station i in the previous period t P in multiple observation sample data it and Multiple known points can be determined, and the known points can be fitted by the fitting algorithm to obtain the value of the quantity and price parameters, and then the functional relationship between charging demand and price can be obtained, as shown in formula (3). Then, the logarithmic transformation of both sides of formula (3) can be obtained to obtain the quantity and price sub-model g(P it ,cate i ,θ), quantity-price sub-model g(P it ,cate i ,θ) As shown in formula (4), when the previous period t changes to the current period t+1, the quantity-price sub-model g(P i,t+1 ,cate i ,θ).

[0164] Among them, multi-task learning (MTL) is a promising field of machine learning that improves generalization performance by using domain knowledge contained in supervisory signals of related tasks.

[0165] Step 3033: Determine the correlation between the charging demand and price elasticity of each preset charging station in the current period according to the benchmark charging demand sub-model of the current period and the quantity-price sub-model of the current period corresponding to each preset charging station.

[0166] Specifically, in this embodiment, the benchmark charging demand sub-model of the current period t+1 is and the quantity-price sub-model g(P i,t+1 ,cate i ,θ) minus the preset quantity-price sub-model of the historical average selling price of station i in the current period t+1 within the past month The correlation between the charging demand and price elasticity of each preset charging station in the current period is obtained, which can be expressed as shown in formula (5):

[0167]

[0168] in, It is the preset quantity-price sub-model under the historical average selling price of station i in the current period t+1, which is a known quantity.

[0169] The charging station service fee pricing method provided in this embodiment includes a preset counterfactual prediction model including a preset benchmark charging demand sub-model and a preset quantity-price sub-model; the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity. Accordingly, when determining the correlation between charging demand and pricing of each preset charging station in the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model, the benchmark parameter of the current period in the benchmark charging demand sub-model is determined according to the first observation sample corresponding to each preset charging station to determine the benchmark charging demand sub-model of the current period; the quantity-price parameter of the current period in the quantity-price sub-model is determined according to the second observation sample corresponding to each preset charging station to determine the quantity-price sub-model of the current period, and the quantity-price parameter of the current period includes: the price elasticity parameter of the current period; the correlation between charging demand and price elasticity of each preset charging station in the current period is determined according to the benchmark charging demand sub-model of the current period corresponding to each preset charging station and the quantity-price sub-model of the current period. By using the preset benchmark charging demand sub-model without price intervention and the price-related quantity-price submodel g(P it ,cate i,θ) constructs a counterfactual prediction model, so that the charging demand under price intervention and the charging demand without price intervention can be predicted respectively, making the counterfactual prediction model more comprehensive in predicting the charging demand, thus making the counterfactual prediction model more accurate in predicting the charging demand.

[0170] As an optional implementation, in this embodiment, step 2032 includes the following steps:

[0171] Step 401 : transform the target charging pile efficiency sub-model and the target revenue sub-model according to the correlation between the charging demand and the pricing of each preset charging station in the current period, so as to obtain the target pricing model of each preset charging station in the current period.

[0172] Among them, the target pile effect sub-model f i E (P it ) can be expressed as the ratio of charging demand to the number of charging piles.

[0173] Among them, the target return sub-model f i P (P it ) can be expressed as the difference between the product of pricing and charging demand and cost.

[0174] Among them, the target pricing model is a comprehensive utility function, which is obtained by linearly weighting the target pile effect sub-model and the target income sub-model, and can be expressed as shown in formula (6):

[0175]

[0176] Among them, α represents the weight of the target pile effect sub-model, β represents the weight of the target return sub-model, and α+β=1.

[0177] Specifically, in this embodiment, since the target pile effect sub-model and the target revenue sub-model are both related to the charging demand, it can be seen from formula (5) that the charging demand is related to pricing. The target pricing model is the linear weighting of the target pile effect sub-model and the target revenue sub-model. Therefore, the target pricing model is related to pricing. The target pricing model of each preset charging station in the current period can be expressed by pricing, as shown in formula (6).

[0178] Among them, the weight of the target pile effect sub-model and the weight of the target income sub-model are customized by the user through the user terminal.

[0179] Step 402 , solving the target pricing model for the current time period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current time period.

[0180] Among them, optimal pricing refers to the solution corresponding to the optimal target pricing model.

[0181] Specifically, in this embodiment, since the target pricing model of each preset charging station in the current period is expressed by pricing, the optimal solution for pricing of each preset charging station in the current period can be obtained by solving the target pricing model through mathematical methods.

[0182] The charging station service fee pricing method provided in this embodiment determines the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and pricing of each preset charging station in the current period and the pre-built target pricing model. The target pile effect sub-model and the target income sub-model are converted according to the correlation between the charging demand and pricing of each preset charging station in the current period to obtain the target pricing model of each preset charging station in the current period; the target pricing model of the pricing of each preset charging station in the current period is solved to obtain the optimal pricing of each preset charging station in the current period. Since the target pile effect sub-model and the target income sub-model are converted according to the correlation between the charging demand and pricing of each preset charging station in the current period, the target pile effect sub-model and the target income sub-model are both related to the charging demand. It can be seen from formula (5) that the charging demand is related to the pricing. The target pricing model is the linear weighting of the target pile effect sub-model and the target income sub-model. Therefore, the target pricing model is related to the pricing. The target pricing model of each preset charging station in the current period can be expressed by the pricing, so that the direct relationship between the target pricing model and the pricing can be obtained, making it easier to solve the optimal pricing.

[0183] Figure 5 A flowchart of a charging station service fee pricing method provided in another embodiment of the present application is as follows: Figure 5 As shown, the target pricing model is represented by linearly weighting the target pile effect sub-model and the target revenue sub-model.

[0184] Accordingly, step 402 includes the following steps:

[0185] Step 501 , using a linear weighted method, converts a multi-objective optimization problem corresponding to a target pricing model for solving the current time period pricing of each preset charging station into a single-objective optimization problem.

[0186] Among them, multiple objectives mean that the target pricing model is related to the target pile effect sub-model and the target revenue sub-model, so pile effect and revenue are two objectives that affect pricing.

[0187] Among them, the single target refers to converting pile efficiency and revenue into the same related target, so that pricing is only affected by the single target.

[0188] Specifically, in this embodiment, the target pile effect sub-model and the target revenue sub-model are both related to charging demand, which is in turn related to pricing. The target pricing model is the linear weighting of the target pile effect sub-model and the target revenue sub-model. Therefore, the target pricing model is related to pricing, and pricing can be used to represent the target pricing model of each preset charging station in the current period, thereby converting the multi-objective optimization problem corresponding to the target pricing model of each preset charging station in the current period into a single-objective optimization problem. The single-objective optimization problem is solved by using the linear weighted method to obtain the functional relationship between the target pricing model and pricing.

[0189] Step 502 , find the optimal solution of the single-objective optimization problem of each preset charging station, and determine each optimal solution as the optimal pricing of each preset charging station in the current time period.

[0190] Specifically, in this embodiment, the target pile effect sub-model and the target income sub-model are first normalized, and different target models may have different units and ranges, and the normalization formula may be expressed as shown in formula (7):

[0191]

[0192] The corresponding weights are set for the target pile effect sub-model and the target revenue sub-model after normalization. If a higher pile effect is desired, the weight α corresponding to the target pile effect sub-model can be increased accordingly. If a higher revenue is desired, the weight β of the target pile effect sub-model can be increased accordingly. Then, the target pile effect sub-model and the target revenue sub-model are linearly weighted to obtain formula (6). Formula (6) represents the functional relationship between pricing and the target pricing model. By solving the optimal solution of the single objective problem corresponding to the target pricing model through mathematical methods, the optimal pricing of each preset charging station in the current period can be obtained.

[0193] The charging station service fee pricing method provided in this embodiment is represented by a linear weighted target pricing model for the target pile effect sub-model and the target revenue sub-model. Accordingly, when solving the target pricing model for the current period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current period, the multi-objective optimization problem corresponding to the target pricing model for solving the current period pricing of each preset charging station is converted into a single-objective optimization problem by using a linear weighted method; the optimal solution of the single-objective optimization problem of each preset charging station is obtained, and each optimal solution is determined as the optimal pricing of each preset charging station in the current period. Since the multi-objective problem is converted into a single-objective problem, the solution of the optimal pricing is simpler. At the same time, the linear weighted method can be used to formulate corresponding weights for the target pile effect sub-model and the target revenue sub-model respectively. If you want to obtain a higher pile effect, you can adjust the weight α corresponding to the target pile effect sub-model accordingly. If you want to obtain a higher revenue, you can adjust the weight β of the target pile effect sub-model accordingly, so that the final pricing can be more flexible according to user needs.

[0194] As an optional implementation manner, in this embodiment, after step 2032, the following steps are further included:

[0195] Step 601, sending the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type to the user terminal, so that the user terminal displays the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type.

[0196] Among them, the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type are calculated separately, using the same model and the same calculation method.

[0197] Specifically, in this embodiment, the server sends the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type that are finally calculated to the user terminal, and the user terminal can display the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type respectively through a display window.

[0198] The charging station service fee pricing method provided in this embodiment determines the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and pricing of each preset charging station in the current period and the pre-built target pricing model, and then sends the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type to the user terminal, so that the user terminal displays the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type. Since the user terminal can display the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type respectively through the display window, the user can obtain the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type more intuitively.

[0199] Figure 6 A schematic diagram of a structural framework for pricing charging station service fees provided in an embodiment of the present application, such as Figure 6 As shown, the structural framework for pricing the charging station service fee provided by the present application is located in the server, and the structural framework for pricing the charging station service fee includes a data processing layer, a multi-level clustering processing layer, a counterfactual prediction layer, and a single-objective price optimization layer.

[0200] In this embodiment, the data processing layer is used to process the charging flow data, the background charging station information management system information, etc., and construct the historical charging pile efficiency characteristic data, historical charging user type, historical power type, historical charging price characteristic data, etc. by using statistical analysis methods, continuous feature discretization processing, data normalization methods and other methods.

[0201] Among them, the statistical analysis method is used to count the charging amount over a period of time in history.

[0202] Among them, the continuous feature discretization processing is used to discretize the pile effect and the proportion of car owners.

[0203] Among them, the data normalization method is used to normalize the target pile effect sub-model and the target income sub-model.

[0204] In this embodiment, the multi-level clustering processing layer is used to perform multi-level clustering processing on the historical charging pile efficiency feature data, the historical charging user types, and the historical power types.

[0205] In this embodiment, the counterfactual prediction layer is used to perform counterfactual prediction on the correlation between charging demand and pricing at each preset charging station during the current period. The counterfactual prediction layer includes a preset benchmark charging demand sub-model and a preset quantity price sub-model.

[0206] Among them, the preset benchmark charging demand sub-model is a model that predicts charging demand under non-price characteristics without price intervention.

[0207] Among them, the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity.

[0208] In this embodiment, the single-objective price optimization layer is used to convert the multi-objective problem into a single-objective problem and solve the single-objective problem, so as to determine the optimal pricing of each preset charging station in the current period.

[0209] Among them, Figure 6 The corresponding charging station service fee pricing method includes the following steps:

[0210] Step 701, obtaining historical charging characteristic data corresponding to at least one preset charging station, where the historical charging characteristic data includes: a plurality of historical charging non-price characteristic data and historical charging price characteristic data.

[0211] Step 702 : performing a first-level clustering process on each preset charging station using the historical power type corresponding to each preset charging station to obtain a power category corresponding to each preset charging station.

[0212] Step 703 : Performing a second-level clustering process on the preset charging stations in each power category using the historical charging pole efficiency feature data corresponding to each preset charging station, so as to obtain the charging pole efficiency category of each preset charging station in each power category.

[0213] Step 704 , calculating the historical vehicle owner ratio corresponding to each preset charging station according to the historical charging user type corresponding to each preset charging station.

[0214] Step 705 , using the historical vehicle owner ratio corresponding to each preset charging station, respectively perform third-level clustering processing on the preset charging stations in each pile effect category to obtain the vehicle owner ratio category of each preset charging station under the pile effect category.

[0215] Step 706, construct a first observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, the first observation sample including: the charging amount of the preset charging station in the previous period, the average price of the preset charging station in the previous period of the historical time period, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous period.

[0216] Step 707, construct a second observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, the second observation sample includes: the charging amount of the preset charging station in the previous period, the category corresponding to the preset charging station, and the price change rate of the preset charging station and surrounding competing charging stations in the previous period.

[0217] Step 708 : determining the benchmark parameters of the current time period in the benchmark charging demand sub-model according to the first observation samples corresponding to each preset charging station, so as to determine the benchmark charging demand sub-model of the current time period.

[0218] Step 709, determining the quantity and price parameters of the current period in the quantity and price sub-model according to the second observation samples corresponding to each preset charging station, so as to determine the quantity and price sub-model of the current period, wherein the quantity and price parameters of the current period include: the price elasticity parameter of the current period.

[0219] Step 710 , determining the correlation between the charging demand and price elasticity of each preset charging station in the current period according to the benchmark charging demand sub-model of the current period and the quantity-price sub-model of the current period corresponding to each preset charging station.

[0220] Step 711, transform the target charging pile efficiency sub-model and the target revenue sub-model according to the correlation between the charging demand and pricing of each preset charging station in the current period, so as to obtain the target pricing model of each preset charging station in the current period.

[0221] Step 712 , using a linear weighted method, converts the multi-objective optimization problem corresponding to the target pricing model for solving the current time period pricing of each preset charging station into a single-objective optimization problem.

[0222] Step 713, find the optimal solution of the single-objective optimization problem of each preset charging station, and determine each optimal solution as the optimal pricing of each preset charging station in the current time period.

[0223] Step 714, the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type are sent to the user terminal, so that the user terminal displays the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type.

[0224] The implementation method of the charging station service fee pricing method provided in this embodiment is similar to the implementation method of the corresponding steps in the above embodiments, and will not be repeated here.

[0225] Figure 7 This is a schematic diagram of the structure of a charging station service fee pricing device provided in an embodiment of the present application, such as Figure 7 As shown, the charging station service fee pricing device provided in this embodiment is located in an electronic device, and the charging station service fee pricing device 80 provided in this embodiment includes: an acquisition module 81, a clustering processing module 82, an execution module 83, a prediction submodule 831, and a determination submodule 832.

[0226] Among them, the acquisition module 81 is used to obtain historical charging feature data corresponding to at least one preset charging station, and the historical charging feature data includes: a variety of historical charging non-price feature data and historical charging price feature data; the clustering processing module 82 is used to perform multi-level clustering processing on each preset charging station according to the various historical charging feature data corresponding to each preset charging station, so as to obtain the category corresponding to each preset charging station; the execution module 83 is used to perform the following operations for two charging user types respectively, so as to complete the current time period charging pricing corresponding to each charging user type respectively, and the charging user types include: car owner type and non-car owner type, and the execution module includes: prediction Submodule 831 is used to perform counterfactual prediction on the correlation between charging demand and pricing of each preset charging station in the current period according to the category corresponding to each preset charging station, the observation sample of charging user type in the previous period in the historical time period corresponding to each preset charging station and the preset counterfactual prediction model, so as to obtain the correlation between charging demand and pricing of each preset charging station in the current period; determine submodule 832, according to the correlation between charging demand and pricing of each preset charging station in the current period and the pre-built target pricing model, the target pricing model includes a target pile effect sub-model and a target revenue sub-model.

[0227] The charging station service fee pricing device provided in this embodiment can be executed Figure 2 The specific implementation principles and technical effects of the method embodiments shown are similar and will not be repeated here.

[0228] Optionally, the plurality of historical charging non-price characteristic data include: historical charging pile efficiency characteristic data, historical charging user types and historical power types.

[0229] Correspondingly, the clustering processing module 82 is specifically used to perform a first-level clustering processing on each preset charging station using the historical power type corresponding to each preset charging station to obtain the power category corresponding to each preset charging station; perform a second-level clustering processing on the preset charging stations in each power category using the historical pile efficiency feature data corresponding to each preset charging station to obtain the pile efficiency category of each preset charging station under each power category; calculate the historical car owner ratio corresponding to each preset charging station according to the historical charging user type corresponding to each preset charging station; perform a third-level clustering processing on the preset charging stations in each pile efficiency category using the historical car owner ratio corresponding to each preset charging station to obtain the car owner ratio category of each preset charging station under the pile efficiency category.

[0230] Optionally, the prediction submodule 831 performs counterfactual prediction on the correlation between the charging demand and pricing of each preset charging station in the current time period according to the category corresponding to each preset charging station, the observation sample of the charging user type in the previous time period in the historical time period corresponding to each preset charging station, and the preset counterfactual prediction model, so as to obtain the correlation between the charging demand and pricing of each preset charging station in the current time period, and is specifically used to: construct a first observation sample of the charging user type in the previous time period in the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, and the first observation sample includes: the charging amount of the preset charging station in the previous time period, the charging amount of the preset charging station in the previous time period in the historical time period, and the charging amount of the preset charging station in the previous time period in the historical time period. The average price of the preset charging station, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous time period; according to the category corresponding to each preset charging station, a second observation sample of the charging user type in the previous time period corresponding to each preset charging station is constructed, and the second observation sample includes: the charging amount of the preset charging station in the previous time period, the category corresponding to the preset charging station, and the price change rate of the preset charging station and the surrounding competing charging stations in the previous time period; according to the first observation sample and the second observation sample corresponding to each preset charging station, and the preset counterfactual prediction model, the correlation between the charging demand and pricing of each preset charging station in the current time period is determined.

[0231] Optionally, the preset counterfactual prediction model includes a preset baseline charging demand sub-model and a preset quantity-price sub-model; the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity.

[0232] Accordingly, when determining the correlation between the charging demand and pricing of each preset charging station in the current time period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model, the prediction submodule 831 is specifically used to: determine the benchmark parameters of the current time period in the benchmark charging demand submodel according to the first observation sample corresponding to each preset charging station, so as to determine the benchmark charging demand submodel for the current time period; determine the quantity and price parameters of the current time period in the quantity and price submodel according to the second observation sample corresponding to each preset charging station, so as to determine the quantity and price submodel for the current time period, the quantity and price parameters of the current time period include: the price elasticity parameter of the current time period; determine the correlation between the charging demand and price elasticity of each preset charging station in the current time period according to the benchmark charging demand submodel for the current time period corresponding to each preset charging station and the quantity and price submodel for the current time period.

[0233] Optionally, when determining the optimal pricing of each preset charging station in the current period based on the correlation between the charging demand and pricing of each preset charging station in the current period and a pre-built target pricing model, the determination submodule 832 is specifically used to: transform the target pile effect submodel and the target revenue submodel according to the correlation between the charging demand and pricing of each preset charging station in the current period to obtain the target pricing model of each preset charging station in the current period; solve the target pricing model for the pricing of each preset charging station in the current period to obtain the optimal pricing of each preset charging station in the current period.

[0234] Optionally, the target pricing model is represented by linearly weighting the target pile effect sub-model and the target revenue sub-model.

[0235] Accordingly, when determining that the submodule 832 solves the target pricing model for the current period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current period, it is specifically used to: use a linear weighted method to convert the multi-objective optimization problem corresponding to the target pricing model for solving the current period pricing of each preset charging station into a single-objective optimization problem; obtain the optimal solution to the single-objective optimization problem of each preset charging station, and determine each optimal solution as the optimal pricing of each preset charging station in the current period.

[0236] Optionally, the charging station service fee pricing device provided in this embodiment further includes: a sending module.

[0237] Among them, the sending module is used to determine that after the sub-module 832 determines the optimal pricing of each preset charging station in the current period according to the correlation between the charging demand and the pricing of each preset charging station in the current period and the pre-built target pricing model, the optimal pricing corresponding to the owner type and the optimal pricing corresponding to the non-owner type are sent to the user terminal, so that the user terminal can display the optimal pricing corresponding to the owner type and the optimal pricing corresponding to the non-owner type.

[0238] The charging station service fee pricing device provided in this embodiment can be executed Figures 4 to 5 The specific implementation principles and technical effects of the method embodiments shown are similar and will not be repeated here.

[0239] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown in FIG. Figure 8 As shown, the electronic device 90 provided in this embodiment includes: a processor 91 and a memory 92 communicatively connected to the processor.

[0240] The memory 92 stores computer-executable instructions; the processor 91 executes the computer-executable instructions stored in the memory 92 to implement the charging station service fee pricing method provided by any of the above embodiments. The relevant descriptions can be understood by referring to the relevant descriptions and effects corresponding to the steps in the accompanying drawings, and no further details are given here.

[0241] The program may include program code, and the program code includes computer-executable instructions. The memory 92 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0242] In this embodiment, the memory 92 is connected to the processor 91 via a bus. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0243] The embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the charging station service fee pricing method provided in any of the above embodiments. For example, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device.

[0244] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the charging station service fee pricing method provided in any one of the above embodiments.

[0245] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0246] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0247] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0248] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0249] If the integrated unit / module is implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the artificial intelligence processor can be any appropriate hardware processor, such as CPU, GPU, FPGA, DSP, and ASIC, etc. Unless otherwise specified, the storage unit can be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory RRAM (Resistive Random Access Memory), dynamic random access memory DRAM (Dynamic Random Access Memory), static random access memory SRAM (Static Random-Access Memory), enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), high-bandwidth memory HBM (High-Bandwidth Memory), hybrid memory cube HMC (Hybrid Memory Cube), etc.

[0250] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a memory and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0251] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0252] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0253] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A charging station service fee pricing method, characterized in that: include: Acquire historical charging characteristic data corresponding to at least one preset charging station, wherein the historical charging characteristic data includes: historical charging pile efficiency characteristic data, historical charging user type, historical power type and historical charging price characteristic data; Performing a first-level clustering process on each preset charging station using the historical power type corresponding to each preset charging station to obtain a power category corresponding to each preset charging station; The historical pile efficiency characteristic data corresponding to each preset charging station is used to perform a second-level clustering process on the preset charging stations in each power category to obtain the pile efficiency category of each preset charging station under each power category; Calculate the historical vehicle owner ratio corresponding to each preset charging station based on the historical charging user type corresponding to each preset charging station; The third-level clustering process is performed on the preset charging stations in each charging pile efficiency category using the historical car owner ratio corresponding to each preset charging station, so as to obtain the car owner ratio category of each preset charging station under the charging pile efficiency category; The following operations are performed for the two charging user types respectively to complete the current time period charging pricing corresponding to each charging user type respectively. The charging user types include: car owner type and non-car owner type. The following operations include: Constructing a first observation sample of the charging user type in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the first observation sample includes: the charging amount of the preset charging station in the previous period, the average price of the preset charging station in the previous period in the historical time period, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous period; Constructing a second observation sample of the charging user type in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the second observation sample includes: the average charging amount of the preset charging station in the historical time period, the category corresponding to the preset charging station, and the price change rate between the preset charging station and the surrounding competing charging stations in the previous period; Determine the correlation between the charging demand and pricing of each preset charging station during the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model; Obtaining a target pricing model for each of the preset charging stations during the current period according to the correlation between charging demand and pricing for each of the preset charging stations during the current period; The target pricing model for the current time period pricing of each of the preset charging stations is solved to obtain the optimal pricing of each of the preset charging stations in the current time period.

2. The method according to claim 1, characterized in that The preset counterfactual prediction model includes a preset benchmark charging demand sub-model and a preset quantity-price sub-model; the preset quantity-price sub-model is a model that defines the relationship between charging demand and price elasticity; The determining of the correlation between the charging demand and pricing of each preset charging station in the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model includes: Determine the benchmark parameters of the current time period in the benchmark charging demand submodel according to the first observation samples corresponding to each preset charging station, so as to determine the benchmark charging demand submodel of the current time period; Determine the quantity and price parameters of the current period in the quantity and price sub-model according to the second observation samples corresponding to each preset charging station, so as to determine the quantity and price sub-model of the current period, wherein the quantity and price parameters of the current period include: a price elasticity parameter of the current period; The correlation between the charging demand and price elasticity of each preset charging station in the current period is determined according to the benchmark charging demand sub-model of the current period corresponding to each preset charging station and the quantity-price sub-model of the current period.

3. The method according to claim 1, characterized in that The target pricing model is represented by a linear weighted method of the target pile effect sub-model and the target income sub-model; Solving the target pricing model for the current time period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current time period includes: A linear weighted method is used to convert a multi-objective optimization problem corresponding to a target pricing model for solving the current period pricing of each of the preset charging stations into a single-objective optimization problem; The optimal solution of the single-objective optimization problem of each of the preset charging stations is obtained, and each of the optimal solutions is determined as the optimal pricing of each of the preset charging stations in the current time period.

4. The method according to any one of claims 1 to 3, characterized in that: After solving the target pricing model for the current time period pricing of each preset charging station to obtain the optimal pricing of each preset charging station in the current time period, the method further includes: The optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type are sent to the user terminal, so that the user terminal displays the optimal pricing corresponding to the car owner type and the optimal pricing corresponding to the non-car owner type.

5. A charging station service fee pricing device, characterized in that: include: An acquisition module, used to acquire historical charging characteristic data corresponding to at least one preset charging station, wherein the historical charging characteristic data includes: historical charging pile efficiency characteristic data, historical charging user type, historical power type and historical charging price characteristic data; Cluster processing module for The historical power types corresponding to the preset charging stations are used to perform the first-level clustering processing on the preset charging stations to obtain the power categories corresponding to the preset charging stations; the historical pile efficiency feature data corresponding to the preset charging stations are used to perform the second-level clustering processing on the preset charging stations in each power category to obtain the pile efficiency categories of each preset charging station under each power category; the historical car owner ratio corresponding to each preset charging station is calculated according to the historical charging user type corresponding to each preset charging station; the historical car owner ratio corresponding to each preset charging station is used to perform the third-level clustering processing on the preset charging stations in each pile efficiency category to obtain the car owner ratio category of each preset charging station under the pile efficiency category; An execution module is used to perform the following operations for two charging user types respectively, so as to complete the current time period charging pricing corresponding to each charging user type respectively. The charging user types include: car owner type and non-car owner type. The execution module includes: A prediction submodule is used to construct a first observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the first observation sample includes: the charging amount of the preset charging station in the previous period, the average price of the preset charging station in the previous period in the historical time period, the average charging amount of the station category to which the preset charging station belongs in the historical time period, and the non-price characteristics of the preset charging station in the previous period; construct a second observation sample of the type of charging users in the previous period of the historical time period corresponding to each preset charging station according to the category corresponding to each preset charging station, wherein the second observation sample includes: the average charging amount of the station category to which the preset charging station belongs in the historical time period, the category corresponding to the preset charging station, and the price change rate of the preset charging station and the surrounding competing charging stations in the previous period; determine the correlation between the charging demand and pricing of each preset charging station in the current period according to the first observation sample and the second observation sample corresponding to each preset charging station and the preset counterfactual prediction model; A determination submodule is used to obtain a target pricing model for each of the preset charging stations in the current period according to the correlation between the charging demand and pricing of each of the preset charging stations in the current period; and solve the target pricing model for the pricing of each of the preset charging stations in the current period to obtain the optimal pricing of each of the preset charging stations in the current period.

6. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 4 when executed by a processor.

Images