Charging station site selection method, system and device and storage medium

Abstract

The invention relates to the technical field of electric vehicle charging station site selection, and particularly provides a charging station site selection method, system and device and a storage medium, and the method comprises the steps: dividing a city geographic region, obtaining a city region set, and obtaining a city region feature index; mining association rules of the urban areas based on the urban area characteristic indexes to obtain a site selection judgment rule set, and screening the urban area set based on the site selection judgment rule set to obtain a target area set and target area characteristic indexes; taking the target area characteristic indexes as input, taking the target area periodic order quantity as output, constructing a charging demand prediction model, and obtaining the predicted periodic order quantity of each target area; and sorting based on the predicted periodic order quantity of each target area to obtain a charging station site selection area. According to the invention, the charging station is preferentially arranged in an area with high charging demand and high growth potential, the utilization rate of the charging pile is improved, and resource waste caused by station building in a low-efficiency area is reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric vehicle charging station location, and particularly relates to a charging station location method, system, device and storage medium. Background Art

[0002] With the rapid development of the new energy vehicle industry, electric vehicles are gradually replacing traditional fuel vehicles and becoming an important part of urban transportation. As the core carrier for electric vehicle energy replenishment, the layout of charging stations, whether it is reasonable, sufficient in coverage, and efficient in operation, directly affects the convenience and satisfaction of users' charging, and is one of the key factors for the popularization of electric vehicles.

[0003] However, the current location and construction of charging stations still face many problems in practice. First, the location planning lacks a unified data support platform. Existing methods often rely on manual experience or historical planning paths, and cannot effectively integrate and comprehensively analyze data such as "vehicles, charging piles, power grids, and positions" in the city, resulting in uneven development in different regions. In some regions, the charging piles are dense but the utilization rate is low, while in some regions, users have difficulty charging. Second, there is a lack of a comprehensive evaluation mechanism for the construction sites in the existing technology. Before site selection, the impacts of key factors such as traffic flow, user behavior, business environment, and competition situation are often ignored. There is a lack of dynamic modeling and heat map analysis of the charging demand in the area, and the potential electricity load and usage potential of different regions cannot be truly reflected, which easily leads to blind construction and waste of resources. Third, the existing technology generally does not have the function of economic calculation after the charging station is built, and lacks a profit model related to construction costs, service pricing, and operation efficiency. It is difficult for operators to judge whether the site has the ability of sustainable operation during the investment decision-making stage, resulting in too long a return period for some charging stations after investment and great operation pressure. Summary of the Invention

[0004] In view of the above deficiencies of the prior art, the present invention provides a charging station location method, system, device and storage medium to solve the above technical problems.

[0005] In the first aspect, the present invention provides a charging station location method, including: Dividing the urban geographical area to obtain a set of urban areas, acquiring the charging station location data of each urban area in the set of urban areas, and obtaining urban area characteristic indicators based on the charging station location data; Based on the urban area characteristic indicators, mining the association rules of the urban areas to obtain a set of location discrimination rules, and screening the set of urban areas based on the set of location discrimination rules to obtain a set of target areas and target area characteristic indicators; Taking the target area characteristic indicators as the input and the target area periodic order volume as the output, constructing a charging demand prediction model to obtain the predicted periodic order volume of each target area; Sort based on the predicted periodic order volume of each target area to obtain the site selection areas for charging stations.

[0006] In an alternative embodiment, divide the urban geographical area to obtain a set of urban areas, acquire the site selection data of charging stations for each urban area in the set of urban areas, and obtain urban area characteristic indicators based on the site selection data of charging stations, including: Use the hexagonal grid cutting method to divide the urban geographical area to obtain a number of hexagonal units, denoted as the set of urban areas; Acquire the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data for each urban area in the set of urban areas; Based on the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data, extract the number of potential customers, the number of casual customers, the power demand of potential customers, the power demand of casual customers, the number of fast charging piles, and the POI density.

[0007] In an alternative embodiment, based on the urban area characteristic indicators, mine the association rules of the urban area to obtain a set of site selection discrimination rules, including: Discretize the urban area characteristic indicators to obtain a standardized set of urban area characteristic indicators; Traverse the occurrence frequency of each single feature item in the standardized set of urban area characteristic indicators, and filter out the feature items that meet the preset minimum support threshold to form a single feature set; In the single feature set with k existing features, through the self-connection method, combine the k features with the same first k - 1 feature items into a new (k + 1)-item candidate feature item set; Traverse the standardized set of urban area characteristic indicators, count the occurrence frequency of each candidate feature item set, retain the candidate feature item sets with support degrees greater than or equal to the preset threshold as new frequent item sets, and apply the pruning principle to eliminate the candidate feature item sets that do not meet the subset frequency requirements; Loop through the generation, screening, and pruning of candidate feature item sets until no new frequent item sets can be generated to complete the frequent item set mining stage; Based on the frequent item sets, divide the antecedent and consequent, calculate the confidence of each association rule, filter out the valid association rules with confidence greater than the preset value, and summarize them to form a set of site selection discrimination rules.

[0008] In an alternative embodiment, use the target area characteristic indicators as the input and the target area periodic order volume as the output to construct a charging demand prediction model to obtain the predicted periodic order volume of each target area, including: Use the random forest regression model as the structure of the charging demand prediction model, set the number of decision trees in the model, the maximum depth of a single decision tree, the minimum number of leaf node samples, and use the mean square error as the training loss function; Taking the characteristic indicators of the target area as input and the periodic order volume of the target area as output, the supervised learning method is used for model training. The k-fold cross-validation method is used to divide the training data set into m non-overlapping subsets. Each time, one of the subsets is selected as the validation set, and the remaining subsets are used as the training set. The training and validation are repeated m times, and finally the average of all validation errors is taken as the model performance indicator.

[0009] In an optional implementation, the charging station site selection area is obtained by sorting the predicted periodic order volume of each target area, including: Based on the charging demand prediction model, a predicted charging order volume for each target area is obtained, where the predicted charging order volume is the total number of charging orders predicted to be generated in the target area within a set prediction period; Taking the predicted charging order volume as the sorting basis, the target areas are sorted in descending order from high to low according to the predicted charging order volume to form the target area sorting result; The predicted growth rate of charging order volume in the target area is used as an auxiliary sorting criterion to rank the target areas with higher predicted growth rate of charging order volume; According to preset screening conditions, a number of top-ranked target areas are selected as recommended charging station areas, wherein the screening conditions include the top N% of the top-ranked target areas or a set number of the top-ranked target areas; The number of the recommended charging station area, the predicted charging order volume, the predicted charging order volume growth rate and the corresponding target area characteristic indicator information are output to form a charging station recommendation list.

[0010] In a second aspect, the present invention provides a charging station site selection system, comprising: A characteristic index acquisition module is used to divide the urban geographical area to obtain an urban area set, obtain the charging station site selection data of each urban area in the urban area set, and obtain the urban area characteristic index based on the charging station site selection data; The target indicator construction module is used to mine the association rules of urban areas based on the characteristic indicators of urban areas, obtain the site selection discrimination rule set, and screen the urban area set based on the site selection discrimination rule set to obtain the target area set and the characteristic indicators of the target area; A prediction model building module is used to build a charging demand prediction model with the target area characteristic index as input and the target area cycle order volume as output, so as to obtain the predicted cycle order volume of each target area; A site selection area generation module, configured to sort based on the predicted periodic order volume of each target area to obtain a charging station site selection area.

[0011] In a third aspect, there is provided a device, including: A memory, configured to store a charging station site selection program; A processor, configured to implement the steps of the charging station site selection method provided in the first aspect when executing the charging station site selection program.

[0012] In a fourth aspect, there is provided a computer-readable storage medium, on which a charging station site selection program is stored. When the charging station site selection program is executed by a processor, the steps of the charging station site selection method provided in the first aspect are implemented.

[0013] The beneficial effects of the present invention are as follows. The charging station site selection method, system, device and storage medium provided by the present invention standardize the grid division of urban geographical areas, and extract urban area characteristic indicators based on charging pile ledger information, vehicle charging behavior data, traffic flow data and POI geographical information, optimizing the charging demand environment modeling process and improving the comprehensiveness and accuracy of regional characteristic expression. By mining the association rules in the urban area characteristic indicators and establishing a site selection discrimination rule set, high-potential areas can be quickly screened, optimizing the screening efficiency of target areas and reducing the blindness of resource investment. Further, a charging demand prediction model with target area characteristic indicators as input and periodic order volume as output is adopted, and through random forest regression and k-fold cross-validation training, the response ability of the prediction model to future charging demand change trends is improved, and the prediction accuracy is increased. Through sorting based on the predicted periodic order volume and introducing the order volume growth rate as an auxiliary sorting criterion, priority site selection for high-demand and growth-potential areas is realized, optimizing the forward-looking and sustainable development capabilities of the station layout. Finally, by outputting a charging station recommendation list with comprehensive information, the quantitative basis for site selection decisions is enhanced, the return on investment and operation efficiency of charging infrastructure construction are improved, and overall, the scientificity, rationality and resource allocation level of charging network construction are enhanced.

[0014] In addition, the design principle of the present invention is reliable and the structure is simple, having a very wide application prospect. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0016] Figure 1It is a schematic flow chart of the method according to an embodiment of the present invention.

[0017] Figure 2 It is a schematic block diagram of the system according to an embodiment of the present invention.

[0018] Figure 3 It is a schematic structural diagram of a device provided by an embodiment of the present invention. Detailed implementation manners

[0019] In order to enable those skilled in the art of the present technology to better understand the technical solutions in the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

[0021] The charging station site selection method provided by the embodiment of the present invention is executed by a computer device. Correspondingly, the charging station site selection system runs in the computer device.

[0022] Figure 1 It is a schematic flow chart of the method according to an embodiment of the present invention. Among them, Figure 1 The execution subject may be a charging station site selection system. According to different requirements, the order of the steps in this flow chart can be changed, and some can be omitted.

[0023] Such as Figure 1 shown, the method includes: S1. Divide the urban geographical area to obtain an urban area set, obtain the charging station site selection data of each urban area in the urban area set, and obtain urban area characteristic indicators based on the charging station site selection data.

[0024] S2. Based on the urban area characteristic indicators, mine the association rules of the urban area to obtain a site selection discrimination rule set. Based on the site selection discrimination rule set, screen the urban area set to obtain a target area set and target area characteristic indicators.

[0025] S3. Use the target area characteristic indicators as input and the target area periodic order volume as output to construct a charging demand prediction model, and obtain the predicted periodic order volume of each target area.

[0026] S4. Sort based on the predicted periodic order volume of each target area to obtain the site selection area for the charging stations.

[0027] In an embodiment of the present invention, based on step S1, a possible embodiment will be given below to non - restrictively elaborate on its specific implementation.

[0028] Adopt the hexagonal grid segmentation method to divide the urban geographical area, obtaining a number of hexagonal units, denoted as the urban area set. Obtain the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data for each urban area in the urban area set. Based on the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data, extract the number of potential customers, the number of casual customers, the electricity demand of potential customers, the electricity demand of casual customers, the number of fast - charging piles, and the POI density.

[0029] Specifically, first, adopt the hexagonal grid segmentation method to divide the urban geographical area. Select a hexagonal unit with a side length of about 900 meters as the basic segmentation unit, so that each hexagonal unit covers an area of about 0.72 square kilometers, ensuring the uniformity and continuity of the area division, and forming the urban area set. This hexagonal division method can better reduce the boundary effect and improve the consistency and accuracy of regional attribute data statistics compared with the traditional square grid division. After completing the urban area division, for each urban area in the urban area set, collect and organize the ledger information of existing charging piles, including the charging pile number, pile type (fast - charging pile or slow - charging pile), rated power, operator, and geographical location information; collect vehicle charging behavior data, including vehicle ID, charging time, charging duration, charging amount, start - end SOC (state of charge), and order amount, etc.; collect traffic flow data, including the density of main roads in the area, the density of road nodes, the number of intersections, and traffic flow intensity change data; collect geographical location POI data, including the location and quantity information of different types of interest points such as commercial facilities, residential communities, office buildings, and transportation hubs. Based on the above - mentioned multi - source heterogeneous data, extract characteristic indicators for each urban area, specifically including: The number of potential customers refers to the number of active vehicles with frequent non - charging stay behavior and no private charging piles in the area; the number of casual customers refers to the number of vehicles with occasional stay and no private piles in the area; the electricity demand of potential customers is calculated based on the difference between the monthly total energy consumption of potential customer vehicles and the charging amount in the area, reflecting the potential unmet charging demand in the area; the electricity demand of casual customers is accumulated based on the actual charging records of casual customer vehicles in the area, reflecting the scale of temporary charging demand; the number of fast - charging piles is obtained by counting the fast - charging pile types in the existing charging pile ledger information; the POI density is obtained by calculating the ratio of the number of POI points in the area to the area of the area, reflecting the commercial activity and population aggregation degree of the area.

[0030] In one embodiment of the present invention, based on step S2, a possible embodiment will be given below to non-restrictively elaborate on its specific implementation.

[0031] Discretize the urban area characteristic indicators to obtain a standardized set of urban area characteristic indicators. Traverse the occurrence frequencies of each single feature item in the standardized set of urban area characteristic indicators, and filter out the feature items that meet the preset minimum support threshold to form a single feature set. In the single feature set with k existing features, through self-joining, combine the k features with the same first k - 1 feature items into a new (k + 1)-item candidate feature item set. Traverse the standardized set of urban area characteristic indicators, count the occurrence frequencies of each candidate feature item set, retain the candidate feature item sets with support greater than or equal to the preset threshold as new frequent item sets, and apply pruning principles to eliminate the candidate feature item sets that do not meet the subset frequency requirements. Loop through the generation, screening, and pruning of candidate feature item sets until no new frequent item sets can be generated to complete the frequent item set mining stage. Based on the frequent item sets, divide the antecedents and consequents, calculate the confidence of each association rule, filter out the valid association rules with confidence greater than the preset value, and summarize to form a site selection discrimination rule set.

[0032] Specifically, first, obtain a set of urban area characteristic indicators, and the characteristic indicators include but are not limited to the number of potential customers, the number of walk-in customers, the power demand of potential customers, the power demand of walk-in customers, the number of fast charging piles, and the POI density. For continuous features, use the quantile division method to divide each feature into three levels: low, medium, and high, to form a standardized set of urban area characteristic indicators, so that all features are represented by the discretized level labels, which is convenient for subsequent logical combination and statistical analysis. Subsequently, traverse each single feature item in the standardized set of urban area characteristic indicators (such as "number of potential customers = high"), and count its occurrence frequency in all urban area samples. If the occurrence frequency of a single feature item is greater than or equal to the preset minimum support threshold (such as 10%), then filter and include this feature item in the single feature set to provide a basis for subsequent feature combination expansion.

[0033] In the single feature set with k formed features, use the self-joining method (Self-Join) for expansion, that is, combine two k-item features with the same first k - 1 feature items to generate a new (k + 1)-item candidate feature item set. For example, if there are two feature groups ["number of potential customers = high", "number of fast charging piles = low"] and ["number of potential customers = high", "POI density = high"], because the first k - 1 feature item "number of potential customers = high" is the same, a new candidate feature item set ["number of potential customers = high", "number of fast charging piles = low", "POI density = high"] can be generated.

[0034] For each generated candidate feature item set, traverse the standardized urban area feature index set and count its occurrence frequency in the urban area samples. If the occurrence frequency of the candidate feature item set is greater than or equal to the preset minimum support threshold, the candidate feature item set is retained as a new frequent item set; if it does not meet the support requirement, it is eliminated.

[0035] Meanwhile, during the generation process of candidate feature item sets, apply the Apriori pruning principle, that is, if any k-item subset of a candidate feature item set is not a frequent item set, directly eliminate the candidate feature item set to avoid invalid expansion and improve the mining efficiency.

[0036] By repeatedly performing operations such as candidate feature item set generation, frequency statistics, screening, and pruning, gradually increase the length of the feature item set until no new frequent item sets can be generated in a certain round, thus completing the frequent item set mining stage.

[0037] After the frequent item set mining is completed, based on each frequent item set, divide the antecedent and the consequent. The antecedent consists of several discrete feature items, and the consequent is the attribute label with high charging demand in the target area.

[0038] Based on the division of the antecedent and the consequent, calculate the confidence of each association rule, that is, the probability of simultaneously satisfying the consequent attribute in the urban area containing the antecedent feature combination. Screen out the effective association rules with a confidence greater than the preset threshold (such as 70%), and eliminate the weak association rules with insufficient confidence.

[0039] Finally, summarize all the screened effective association rules to form a site selection discrimination rule set. The site selection discrimination rule set is used for subsequent urban area screening, assisting in identifying target areas with high charging demand potential, and supporting the optimization decision-making of charging station site selection.

[0040] In an embodiment of the present invention, based on step S3, a possible embodiment will be given below to non-restrictively elaborate on its specific implementation scheme.

[0041] Use a random forest regression model as the structure of the charging demand prediction model, set the number of decision trees in the model, the maximum depth of a single decision tree, and the minimum number of samples in the leaf nodes. The training loss function uses the mean squared error. Use the target area feature index as the input and the target area periodic order volume as the output, and perform model training in a supervised learning manner. Adopt the k-fold cross-validation method, divide the training data set into m non-overlapping subsets, each time select one of the subsets as the validation set, and the remaining subsets as the training set, repeat the training and validation m times, and finally take the average value of all validation errors as the model performance index.

[0042] Specifically, first, a random forest regression model is selected as the structure of the charging demand prediction model. The model consists of multiple regression decision trees, and the prediction stability and accuracy of the model are improved through the integrated learning method. According to the business characteristics of charging demand prediction, key hyperparameters in the random forest regression model are set as follows: the number of decision trees n_estimators is set to 500 to enhance the model's learning ability for complex feature patterns; the maximum depth max_depth of a single decision tree is set to unlimited, allowing the decision tree to grow adaptively according to the data characteristics and improving the model's fitting ability for the non-linear relationship of charging demand; the minimum number of samples in a leaf node min_samples_leaf is set to 10 to control the minimum splitting scale of the decision tree and prevent overfitting. During the training process, the mean squared error (MSE) is used as the loss function, and the goal is to minimize the mean squared error between the model's predicted output and the actual periodic order volume.

[0043] In the model training stage, the target area feature indicators are used as the input, and the historical periodic order volume corresponding to the target area is used as the output, and the model is trained in a supervised learning manner.

[0044] To improve the robustness of the model under different data partitions, the k-fold cross-validation method is used for training evaluation. The entire training data set is divided into m non-overlapping subsets, where m can be set to 10.

[0045] In each round of training, one of the subsets is randomly selected as the validation set, and the remaining m - 1 subsets are used as the training set. The model is trained based on the training set, and the model performance is evaluated on the validation set, and the validation mean squared error is calculated.

[0046] Repeat the training and validation for m rounds, rotate different validation sets each time, and finally average the m validation mean squared errors as the comprehensive performance index of the model.

[0047] Based on the comprehensive evaluation results of cross-validation, the hyperparameters of the random forest regression model can be further dynamically adjusted to obtain a charging demand prediction model with the minimum mean squared error and the optimal generalization ability on the validation set.

[0048] After the model training is completed, the trained charging demand prediction model is used in the subsequent stage. The latest target area feature indicators are input to predict the charging order volume in the future set period for each target area, providing a scientific basis for the location selection and resource layout of charging stations.

[0049] In an embodiment of the present invention, based on step S4, a possible embodiment will be given below to non-restrictively elaborate on its specific implementation scheme.

[0050] Based on the charging demand prediction model, the predicted charging order volume of each target area is obtained, and the predicted charging order volume is the total number of charging orders predicted to be generated in the target area within the set prediction period. Using the predicted charging order volume as the sorting basis, the target areas are arranged in descending order from high to low according to the predicted charging order volume to form the sorting result of the target area. Using the predicted charging order volume growth rate of the target area as the auxiliary sorting standard, the target areas with higher predicted charging order volume growth rate are arranged. According to the preset screening conditions, several target areas with the highest ranking are selected as the recommended charging station areas, and the screening conditions include the top N% of the target areas in the sorting or the set number of target areas before the sorting. Output the number of the recommended charging station area, the predicted charging order volume, the predicted charging order volume growth rate and the corresponding target area characteristic indicator information to form a recommended charging station list.

[0051] Specifically, after obtaining the predicted charging order volume of all target areas, the predicted charging order volume is used as the sorting basis, and each target area is sorted in descending order from high to low according to the predicted charging order volume value to form the target area sorting result. This sorting method gives priority to the target areas with larger charging demand scale, providing a basic basis for subsequent site selection decisions.

[0052] During the sorting process, if the predicted charging order volume values ​​of multiple target areas are similar, in order to further optimize the priority site selection decision, the predicted charging order volume growth rate is introduced as an auxiliary sorting criterion. The predicted charging order volume growth rate is calculated by modeling the historical charging order volume change trend to reflect the growth potential of charging demand in the target area. For target areas with similar predicted charging order volumes, the target areas with higher predicted charging order volume growth rates are prioritized to take into account the current demand scale and future growth potential, and improve the sustainable development capacity of the site selection area.

[0053] After the sorting is completed, according to the preset screening conditions, several target areas with the highest ranking are selected from the target area sorting results as recommended charging station areas. The screening conditions include but are not limited to: selecting the top N% of the target areas, or selecting a set number of target areas before sorting (such as the top 50 or top 100). The specific screening criteria can be flexibly set according to the actual construction plan or investment budget.

[0054] Finally, the recommended charging station areas obtained by screening are output to form a charging station recommendation list. The recommendation list contains the number of each recommended area, the corresponding predicted charging order volume, the predicted charging order volume growth rate, and the corresponding target area characteristic indicator information (such as the number of potential customers, POI density, number of fast charging piles, etc.), providing quantitative and systematic data support for subsequent charging station construction decisions, and optimizing the layout efficiency of charging infrastructure and resource allocation effects.

[0055] In some embodiments, the charging station site selection system may include multiple functional modules composed of computer program segments. The computer programs of each program segment in the charging station site selection system may be stored in the memory of a computer device and executed by at least one processor to perform (see Figure 1 description) the function of charging station site selection.

[0056] In this embodiment, according to the functions it performs, the charging station site selection system can be divided into multiple functional modules, as Figure 2 shown. The module referred to in the present invention means a series of computer program segments that can be executed by at least one processor and can complete fixed functions, and are stored in the memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.

[0057] Feature index acquisition module, which is used to divide the urban geographical area to obtain a set of urban areas, acquire the charging station site selection data of each urban area in the set of urban areas, and obtain the urban area feature indexes based on the charging station site selection data; Target index construction module, which is used to mine the association rules of urban areas based on the urban area feature indexes to obtain a set of site selection discrimination rules, and screen the set of urban areas based on the set of site selection discrimination rules to obtain a set of target areas and target area feature indexes; Prediction model construction module, which is used to construct a charging demand prediction model with the target area feature indexes as the input and the target area periodic order volume as the output, and obtain the predicted periodic order volume of each target area; Site selection area generation module, which is used to sort based on the predicted periodic order volume of each target area to obtain the charging station site selection area.

[0058] In an implementable manner, the feature index acquisition module includes: Area division unit, which is used to divide the urban geographical area by using the hexagonal grid segmentation method to obtain a number of hexagonal units, denoted as the set of urban areas; Data acquisition unit, which is used to acquire the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data of each urban area in the set of urban areas; Index extraction unit, which is used to extract the number of potential customers, the number of casual customers, the power demand of potential customers, the power demand of casual customers, the number of fast charging piles, and the POI density based on the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data, and geographical location POI data.

[0059] In an implementable manner, the target index construction module includes: A discrete unit for discretizing the characteristic indicators of urban areas to obtain a set of standardized urban area characteristic indicators; A first traversal unit for traversing the occurrence frequencies of each single feature item in the set of standardized urban area characteristic indicators, screening out feature items that meet a preset minimum support threshold to form a single feature set; A merging unit for, in the single feature set with k features, combining k features with the same first k - 1 feature items into a new (k + 1)-item candidate feature item set through a self-joining method; A second traversal unit for traversing the set of standardized urban area characteristic indicators, counting the occurrence frequencies of each candidate feature item set, retaining the candidate feature item sets with a support degree greater than or equal to a preset threshold as new frequent item sets, and applying a pruning principle to eliminate candidate feature item sets that do not meet the subset frequency requirements; A pruning unit for repeatedly performing candidate feature item set generation, screening, and pruning until no new frequent item sets can be generated, completing the frequent item set mining stage; A rule generation unit for dividing the antecedent and consequent based on the frequent item set, calculating the confidence of each association rule, screening out valid association rules with a confidence greater than a preset value, and summarizing them to form a site selection discrimination rule set.

[0060] Figure 3 The charging station site selection method provided by the embodiments of the present application can be applied to devices. Those skilled in the art can understand that the device structure involved in the embodiments of the present invention does not constitute a limitation on the device. The device may include more or fewer components than shown in the figure, or combine some components, or have different component arrangements. In the embodiments of the present invention, the device includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown in the figure, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the embodiments of the present application described and / or claimed herein.

[0061] Among them, the device 300 may include: a processor 310, a memory 320, and a communication unit 330. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the server shown in the figure does not constitute a limitation on the present invention. It can be a bus structure, a star structure, and may also include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0062] Among them, the memory 320 can be used to store the execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disc. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 can execute some or all of the steps in the above method embodiments.

[0063] The processor 310 is the control center of the storage device, connecting various parts of the entire electronic device through various interfaces and lines. By running or executing the software programs and / or modules stored in the memory 320, and calling the data stored in the memory, it executes various functions of the electronic device and / or processes data. The processor can be composed of an integrated circuit (IC). For example, it can be composed of a single packaged IC, or can be composed of multiple packaged ICs with the same or different functions connected together. For example, the processor 310 can include only a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single arithmetic core or can include multiple arithmetic cores.

[0064] The communication unit 330 is used to establish a communication channel so that the storage device can communicate with other devices. It receives user data sent by other devices or sends user data to other devices.

[0065] The present invention also provides a computer storage medium. Among them, the computer storage medium can store a program, and when the program is executed, it can include some or all of the steps in the various embodiments provided by the present invention. The storage medium can be a magnetic disk, an optical disc, a read-only memory (ROM) or a random access memory (RAM), etc.

[0066] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc, etc., various media that can store program codes, including several instructions for causing a computer device (which can be a personal computer, a server, or a second device, a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0067] For the same or similar parts among the various embodiments in this specification, reference can be made to each other. In particular, for the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the descriptions in the method embodiments.

[0068] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of systems or modules can be in electrical, mechanical, or other forms.

[0069] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical modules, that is, they can be located in one place, or they can be distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0070] In addition, in each embodiment of the present invention, the functional modules can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0071] Although the present invention has been described in detail by reference to the accompanying drawings and in conjunction with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and all such modifications or substitutions should fall within the scope of the present invention / Any person skilled in the art within the technical scope disclosed by the present invention can easily conceive of changes or substitutions, which should all be covered by the protection scope of the present invention.

Claims

1. A method for charging station site selection, characterized in that, include: Divide the urban geographical areas to obtain an urban area set, obtain the charging station site selection data of each urban area in the urban area set, and obtain the urban area characteristic index based on the charging station site selection data; Based on the characteristic indicators of urban areas, the association rules of urban areas are mined to obtain a set of site selection discrimination rules. Based on the set of site selection discrimination rules, the urban area set is screened to obtain a target area set and target area characteristic indicators. Taking the characteristic index of the target area as input and the periodic order volume of the target area as output, a charging demand prediction model is constructed to obtain the predicted periodic order volume of each target area; The charging station location area is obtained by sorting the predicted periodic order volume of each target area.

2. The method according to claim 1, wherein The urban geographical areas are divided to obtain an urban area set, and the charging station site selection data of each urban area in the urban area set is obtained. Based on the charging station site selection data, the urban area characteristic indicators are obtained, including: The urban geographical area is divided by using the hexagonal gridding method to obtain a number of hexagonal units, which are recorded as urban area sets; Obtain the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data and geographic location POI data for each urban area in the urban area concentration; Based on the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data and geographic location POI data, the number of potential customers, the number of individual customers, the power required by potential customers, the power required by individual customers, the number of fast charging piles and POI density are extracted.

3. The method according to claim 1, wherein Based on the characteristic indicators of urban areas, the association rules of urban areas are mined to obtain a set of site selection rules, including: Discretize the urban area characteristic indicators to obtain a standardized urban area characteristic indicator set; Traverse the occurrence frequency of each single feature item in the standardized urban area feature index set, filter out the feature items that meet the preset minimum support threshold, and form a single feature set; In a single feature set with k features, k features with the same first k-1 features are combined into a new (k+1) candidate feature set through self-connection; Traverse the standardized urban area feature index set, count the occurrence frequency of each candidate feature item set, retain the candidate feature item sets with support greater than or equal to the preset threshold as new frequent item sets, and apply the pruning principle to eliminate the candidate feature item sets that do not meet the subset frequency requirements; The candidate feature item set generation, screening and pruning are executed cyclically until no new frequent item sets can be generated, thus completing the frequent item set mining phase; Based on the frequent item sets, the antecedents and consequents are divided, the confidence of each association rule is calculated, and the valid association rules with confidence greater than the preset value are screened out and summarized to form a set of site selection discrimination rules.

4. The method according to claim 1, wherein Taking the characteristic index of the target area as input and the periodic order volume of the target area as output, a charging demand prediction model is constructed to obtain the predicted periodic order volume of each target area, including: Use the random forest regression model as the structure of the charging demand prediction model, set the number of decision trees in the model, the maximum depth of a single decision tree, the minimum number of leaf node samples, and use the mean square error as the training loss function; Taking the characteristic indicators of the target area as input and the periodic order volume of the target area as output, the supervised learning method is used for model training. The k-fold cross-validation method is used to divide the training data set into m non-overlapping subsets. Each time, one of the subsets is selected as the validation set, and the remaining subsets are used as the training set. The training and validation are repeated m times, and finally the average of all validation errors is taken as the model performance indicator.

5. The method according to claim 1, wherein Based on the predicted periodic order volume of each target area, the charging station site selection area is obtained, including: Based on the charging demand prediction model, a predicted charging order volume for each target area is obtained, where the predicted charging order volume is the total number of charging orders predicted to be generated in the target area within a set prediction period; Taking the predicted charging order volume as the sorting basis, the target areas are sorted in descending order from high to low according to the predicted charging order volume to form the target area sorting result; The predicted growth rate of charging order volume in the target area is used as an auxiliary sorting criterion to rank the target areas with higher predicted growth rate of charging order volume; According to preset screening conditions, a number of top-ranked target areas are selected as recommended charging station areas, wherein the screening conditions include the top N% of the top-ranked target areas or a set number of the top-ranked target areas; The number of the recommended charging station area, the predicted charging order volume, the predicted charging order volume growth rate and the corresponding target area characteristic indicator information are output to form a charging station recommendation list.

6. A charging station site selection system, characterized in that, include: A characteristic index acquisition module is used to divide the urban geographical area to obtain an urban area set, obtain the charging station site selection data of each urban area in the urban area set, and obtain the urban area characteristic index based on the charging station site selection data; The target indicator construction module is used to mine the association rules of urban areas based on the characteristic indicators of urban areas, obtain the site selection discrimination rule set, and screen the urban area set based on the site selection discrimination rule set to obtain the target area set and the characteristic indicators of the target area; A prediction model building module is used to build a charging demand prediction model with the target area characteristic index as input and the target area cycle order volume as output, so as to obtain the predicted cycle order volume of each target area; The site selection area generation module is used to sort the predicted periodic order volume of each target area to obtain the charging station site selection area.

7. The system according to claim 6, wherein The feature index acquisition module includes: The regional division unit is used to divide the urban geographical area by using a hexagonal grid segmentation method to obtain a number of hexagonal units, which are recorded as an urban area set; A data acquisition unit, used to acquire the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data and geographic location POI data of each urban area in the urban area concentration; The indicator extraction unit is used to extract the number of potential customers, the number of individual customers, the power required by potential customers, the power required by individual customers, the number of fast charging piles and the POI density based on the ledger information of existing charging piles, vehicle charging behavior data, traffic flow data and geographic location POI data.

8. The system according to claim 6, wherein The target indicator building blocks include: Discrete units are used to discretize the urban area characteristic indicators to obtain a set of standardized urban area characteristic indicators; The first traversal unit is used to traverse the occurrence frequencies of each single feature item in the standardized urban area feature index set, screen out the feature items that meet the preset minimum support threshold, and form a single-item feature set; The merging unit is used to, in the single-item feature set with k existing features, combine k features with the same first k-1 feature items into a new (k+1)-item candidate feature item set through a self-joining method; The second traversal unit is used to traverse the standardized urban area feature index set, count the occurrence frequencies of each candidate feature item set, retain the candidate feature item sets with support greater than or equal to the preset threshold as new frequent item sets, and apply the pruning principle to eliminate the candidate feature item sets that do not meet the subset frequency requirements; The pruning unit is used to repeatedly execute the generation, screening, and pruning of candidate feature item sets until no new frequent item sets can be generated, completing the frequent item set mining stage; The rule generation unit is used to divide the antecedent and consequent based on the frequent item set, calculate the confidence of each association rule, screen out the valid association rules with confidence greater than the preset value, and summarize them to form a site selection discrimination rule set.

9. A charging station site selection device, characterized in that, Including: A memory for storing the charging station site selection program; A processor for implementing the steps of the charging station site selection method according to any one of claims 1-5 when executing the charging station site selection program.

10. A computer-readable storage medium storing a computer program, characterized in that, The charging station site selection program is stored on the readable storage medium, and when the charging station site selection program is executed by the processor, the steps of the charging station site selection method according to any one of claims 1-5 are implemented.

Images