Method and device for constructing energy supplement network of vehicle

Abstract

The application provides a method and device for constructing an energy supplement network of a vehicle, and applies to the technical field of intelligent transportation systems. The method comprises the following steps: obtaining a data set of a historical driving route of a target vehicle; calculating a route comprehensive similarity between any two routes in the historical driving route of the target vehicle based on the historical driving route data set; clustering the historical driving route of the target vehicle according to the route comprehensive similarity, and generating at least one cluster; and constructing an energy supplement network of the target vehicle according to the at least one cluster. The method avoids directly taking historical discrete trajectory points as analysis units to identify local stay hot spots and scattered trajectory aggregation areas. Therefore, the application can more accurately and completely identify high-frequency transport trunk lines from the historical driving route trajectory of the vehicle, and construct a vehicle energy supplement network that matches actual operation requirements.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation systems technology, specifically to a method and apparatus for constructing a vehicle energy replenishment network. Background Technology

[0002] Electric commercial vehicles, with their outstanding advantages such as zero emissions, low energy consumption, and low operating costs, have been widely used in scenarios such as trunk logistics, port short-haul transportation, urban distribution, and mining operations. Compared with electric passenger vehicles, electric commercial vehicles have significant characteristics such as large load capacity, high energy consumption, fixed operating routes, concentrated charging time periods, and high dependence on dedicated high-power charging piles. Therefore, in order to ensure vehicle range safety and improve the utilization efficiency of charging resources, higher requirements are placed on the accuracy of transportation trunk line identification, the special compatibility of charging stations, and the rationality of the distribution of charging nodes along the route.

[0003] In existing technologies, historical discrete GPS points of vehicles are typically used as analysis units to identify local hotspots and scattered trajectory clusters. This approach tends to result in fragmented trunk line identification, significant deviations from actual traffic flow distribution, and difficulty in accurately delineating high-frequency logistics trunk lines. Therefore, how to accurately identify high-frequency transportation trunk lines of vehicles and, based on this, construct a dedicated power replenishment network that matches their operational characteristics is a pressing technical problem to be solved in this field. Summary of the Invention

[0004] In view of this, this application aims to provide a method and apparatus for constructing a vehicle refueling network. It solves the problems of fragmented trunk line identification, large deviations from actual traffic flow distribution, and difficulty in accurately classifying high-frequency logistics trunk lines in existing technologies.

[0005] In a first aspect, embodiments of this application provide a method for constructing a vehicle's energy replenishment network, comprising: Obtain a dataset of the target vehicle's historical driving routes; Based on the historical driving route dataset, calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle; Based on the comprehensive similarity of the routes, the historical driving routes of the target vehicle are clustered to generate at least one cluster. Based on the at least one cluster, construct the refueling network for the target vehicle.

[0006] Optionally, calculating the comprehensive route similarity between any two routes in the target vehicle's historical driving routes based on the historical driving route dataset includes: Based on the historical driving route dataset, calculate the spatial path overlap, driving direction overlap, and runtime matching degree of any two routes in the historical driving route of the target vehicle. The overall route similarity between any two routes is determined based on the spatial path overlap, the driving direction overlap, and the runtime matching degree.

[0007] Optionally, the step of clustering the historical driving routes of the target vehicle based on the comprehensive route similarity to generate at least one cluster includes: Based on the comprehensive similarity of the routes, the core routes, boundary routes of the core routes, and noisy routes in the historical driving routes of the target vehicle are determined. Based on the core routes and boundary routes of the target vehicle's historical driving routes, the historical driving routes of the target vehicle are clustered to generate at least one cluster, wherein each cluster corresponds to a candidate high-frequency transportation trunk line.

[0008] Optionally, determining the core route, the boundary route of the core route, and the noisy route in the target vehicle's historical driving routes based on the comprehensive route similarity includes: Based on the comprehensive route similarity, the number of neighboring similar routes for each route in the target vehicle's historical driving routes is determined, wherein the neighboring similar routes represent routes whose comprehensive route similarity with each route is greater than or equal to a neighborhood threshold. Based on the comprehensive similarity of the routes, the number of neighboring similar routes for each route, and the route frequency weight for each route, the core routes, boundary routes of the core routes, and noisy routes in the historical driving routes of the target vehicle are determined. The route frequency weight for each route is the ratio of the average daily number of vehicles passing through each route to the average frequency of all routes in the historical driving routes of the target vehicle.

[0009] Optionally, when at least two clusters exist, the vehicle refueling network construction method described in the first aspect further includes: Calculate the overall route similarity between any two clusters; When the route similarity of any two clusters is greater than or equal to the neighborhood threshold, the two clusters are merged. Calculate the total number of routes and total frequency of travel contained in each merged cluster; When the total number of routes in the merged first cluster is less than a preset total number threshold, and / or the total frequency of travel of the routes in the first cluster is less than a preset total frequency threshold, the merged first cluster is removed.

[0010] Optionally, constructing the refueling network for the target vehicle based on the at least one cluster includes: Based on the dataset containing routes in each cluster, calculate the comprehensive score for each cluster; Based on the comprehensive score of each cluster, the level of the candidate high-frequency transport trunk line corresponding to the cluster is determined; Based on the grade of the candidate high-frequency transport trunk lines corresponding to the clusters, a refueling network for the target vehicles is constructed.

[0011] Optionally, based on the dataset containing routes for each cluster, a comprehensive score is calculated for each cluster, including: Based on the dataset containing routes in each cluster, a traffic intensity score, a transport attribute score, and a road network adaptability score are calculated for each cluster. The traffic intensity score is used to characterize the traffic volume of the routes contained in the cluster, the transport attribute score is used to characterize the route value of the routes contained in the cluster, and the road network adaptability score is used to characterize the road network quality of the routes contained in the cluster. The comprehensive score for each cluster is determined based on the traffic intensity score, transport attribute score, and road network adaptability score of the routes contained in each cluster.

[0012] Optionally, determining the level of the candidate high-frequency transport trunk line corresponding to each cluster based on the comprehensive score of each cluster includes: When the comprehensive score of a cluster is greater than or equal to the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a first-level high-frequency trunk line. When the comprehensive score of a cluster is greater than or equal to the second comprehensive score threshold and less than the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a secondary high-frequency trunk line. When the comprehensive score of a cluster is less than the second comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a low-frequency scattered trunk line.

[0013] Optionally, constructing the energy replenishment network for the target vehicle based on the grade of the candidate high-frequency transport trunk lines corresponding to the cluster includes: Obtain the current charging station information for each candidate high-frequency transport trunk line corresponding to each cluster. The current charging station information includes the location, charging capacity, operating status, and number of the current charging station. The candidate high-frequency transport trunk lines corresponding to each cluster are spatially associated with the current charging station to construct the energy replenishment network for the target vehicle. Based on the grade of the candidate high-frequency transport trunk line corresponding to the cluster, determine the target charging station information of the candidate high-frequency transport trunk line corresponding to the cluster; Based on the target charging station information of the candidate high-frequency transport trunk lines corresponding to the clusters, the charging station information corresponding to the locations that need to be recharged in the candidate high-frequency transport trunk lines corresponding to the clusters is labeled, and a recharge network dataset is formed.

[0014] Secondly, embodiments of this application provide a vehicle energy replenishment network construction device, including: a data acquisition module, used to acquire a dataset of historical driving routes of a target vehicle; The calculation module is used to calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle based on the historical driving route dataset. The route clustering module is used to cluster the historical driving routes of the target vehicle based on the comprehensive similarity of the routes, and generate at least one cluster. A power replenishment network construction module is used to construct a power replenishment network for the target vehicle based on the at least one cluster.

[0015] Based on the above, this application provides a method for constructing a vehicle refueling network. This method obtains a dataset of historical driving routes of the target vehicle, calculates the comprehensive similarity between any two routes in the historical driving routes based on this dataset, clusters the historical driving routes of the target vehicle according to the comprehensive similarity, generates at least one cluster, and constructs a refueling network for the target vehicle based on at least one cluster. This application uses a complete dataset of historical driving routes as the analysis object, calculates the comprehensive similarity between any two historical routes, and clusters the routes to generate clusters representing high-frequency transportation trunk lines. This avoids directly using historical discrete trajectory points as analysis units to identify local hotspots and scattered trajectory clusters. Therefore, it can more accurately and completely identify high-frequency transportation trunk lines from the vehicle's historical driving route trajectories and construct a vehicle refueling network that matches actual operational needs. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for constructing a vehicle's energy replenishment network, as provided in an embodiment of this application.

[0018] Figure 2 This is a flowchart illustrating another method for constructing a vehicle's energy replenishment network, as provided in an embodiment of this application.

[0019] Figure 3 This is a flowchart illustrating another method for constructing a vehicle's energy replenishment network, as provided in an embodiment of this application.

[0020] Figure 4 This is a flowchart illustrating another method for constructing a vehicle's energy replenishment network, as provided in an embodiment of this application.

[0021] Figure 5 This is a structural block diagram of a vehicle energy replenishment network construction device provided in an embodiment of this application. Detailed Implementation

[0022] Unless otherwise defined, the technical or scientific terms used in the embodiments of this specification shall have the ordinary meaning understood by one of ordinary skill in the art to which this specification pertains. The terms "first," "second," and similar terms used in the embodiments of this specification do not indicate any order, quantity, or importance, but are merely used to avoid confusion of constituent elements.

[0023] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two," and "including" is interpreted as open-ended or encompassing, that is, "including, but not limited to." In the description of this specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.

[0024] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the execution order in the drawings is merely for convenience of explanation and does not represent the actual execution order. The described embodiments are only a part of the embodiments in this specification, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in this specification without creative effort are within the scope of protection of this specification.

[0025] Exemplary methods like Figure 1 As shown in this exemplary embodiment, a method for constructing a vehicle's energy replenishment network is provided. This method can be executed by a computer system and may include the following steps: S10: Obtain the dataset of historical driving routes of the target vehicle.

[0026] Specifically, the target vehicle refers to one or more electric commercial vehicles that require the construction of a power replenishment network, such as electric heavy-duty trucks, electric passenger vehicles, and electric special-purpose vehicles. The dataset of the target vehicle's historical driving routes can be a sequence of multiple complete driving trajectories experienced by the target vehicle within a past operating cycle. The operating cycle can be preset, such as one week, one month, or one quarter. Historical trajectory data of all vehicles can be collected, which can be represented as T={T1,T2,…T…}. N}, where any trajectory sequence T i It contains trajectory points in the i-th trajectory sequence arranged by timestamp, for example, the j-th trajectory point T in the i-th trajectory sequence. ij ={(lon i,j ,lat i,j ,t i,j ,v i,j ,τ i,j ,ρ i,j ,s i,j ,c i,j Each trajectory point records the longitude of the vehicle during its journey. i,j latitude i,j timestamp t i,j Instantaneous vehicle speed v i,j Chassis number τ i,j Vehicle mileage ρ i,j Vehicle status s i,j and charging status c i,j Information such as...

[0027] It is understood that not all trajectory points or trajectory sequences are valid data. In order to obtain valid driving trajectories and reconstruct complete driving routes, in some embodiments, this application preprocesses these original trajectory points. Preprocessing may include cleaning and standardization, and filtering a set of driving trajectory points that meet preset conditions from the historical driving trajectory point set as the dataset of the target vehicle's historical driving routes. The preset conditions include instantaneous vehicle speed greater than zero, vehicle mileage greater than or equal to a first preset mileage, vehicle status as running, and charging status as not charging. For example, the first preset mileage is 5 kilometers, but it can also be set to other values ​​according to vehicle type, sampling period, or operating scenario. In addition, optionally, the above preprocessing may also include: calculating the distance of the trajectory point from the road network baseline using the spherical distance formula. When the distance is greater than the first preset distance, the trajectory point is determined to be a drift point and is discarded. The first preset distance can be set to 1km. After completing the above preprocessing, multiple valid driving trajectories can be reconstructed into a complete driving route. Specifically, trajectory features are extracted from the obtained multiple valid driving trajectories to obtain the starting point, ending point, road level sequence, route heading angle, total route mileage, average route speed, and time interval of the route, forming a multi-dimensional feature vector of the complete route. Finally, a dataset T'={T1',T2',…T...} of the historical driving routes of the target vehicle is obtained. M '}, where M is less than or equal to N, and the dataset of the target vehicle's historical driving routes is a vector dataset.

[0028] S20: Based on the historical driving route dataset, calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle.

[0029] Route similarity refers to a comprehensive index used to quantitatively evaluate the degree of similarity between any two routes in the historical driving routes of a target vehicle in multiple dimensions, such as spatial morphology, driving direction, and time pattern. This avoids judging the similarity of routes from a single dimension, thus making it more conducive to identifying routes that truly serve the same type of transportation task.

[0030] S30: Based on the comprehensive similarity of the routes, cluster the historical driving routes of the target vehicle to generate at least one cluster.

[0031] In this application, the goal of cluster analysis is to identify candidate transport routes with high-frequency traffic characteristics from a large number of historical travel routes. Unlike clustering based solely on discrete point density, this application uses the comprehensive similarity of routes as the neighborhood criterion to cluster the historical travel routes of target vehicles, generating at least one cluster, thereby distinguishing between high-frequency routes and random routes.

[0032] S40: Construct a refueling network for the target vehicle based on at least one cluster.

[0033] Clustering is a digital representation of high-frequency transport routes, with one cluster corresponding to one candidate high-frequency transport route. The process of constructing the charging network for the target vehicle involves spatially associating the high-frequency transport routes corresponding to the clusters with charging station information.

[0034] In some embodiments of this application, a dataset of historical driving routes of the target vehicle is obtained. Based on this dataset, the comprehensive similarity between any two routes in the target vehicle's historical driving routes is calculated. The historical driving routes of the target vehicle are then clustered according to the comprehensive similarity, generating at least one cluster. Based on this cluster, a power replenishment network for the target vehicle is constructed. This application uses a complete dataset of historical driving routes as the analysis object. By calculating the comprehensive similarity between any two historical routes and clustering the routes, clusters representing high-frequency transportation trunk lines are generated. This avoids directly using discrete historical trajectory points as analysis units to identify local hotspots and scattered trajectory clusters. Therefore, it can more accurately and completely identify high-frequency transportation trunk lines from the vehicle's historical driving route trajectories, and construct a vehicle power replenishment network that matches actual operational needs.

[0035] To better identify routes that truly serve similar transportation tasks, this application uses complete routes as the comparison object, fusing spatial, directional, and temporal dimensions to obtain the comprehensive route similarity between any two routes in the target vehicle's historical travel routes. In some embodiments of this application, such as... Figure 2 As shown in Figure S20, based on the historical driving route dataset, the comprehensive similarity between any two routes in the target vehicle's historical driving routes is calculated. Specific steps may include: S21: Based on the historical driving route dataset, calculate the spatial path overlap, driving direction overlap, and runtime matching degree of any two routes in the historical driving route of the target vehicle.

[0036] Using a complete historical driving route dataset of a single target vehicle as the basic analysis unit, the comprehensive similarity between any two driving routes is quantitatively calculated from three dimensions: spatial path overlap, driving direction overlap, and runtime matching degree. Based on the route heading angle, driving time interval, and total route mileage in the historical driving route dataset, and combined with the vehicle's mileage ρ at each trajectory point on the route... i,jIt can calculate the overlapping mileage of any two routes, the average heading angle of any two routes, and the overlapping time period of any two routes. Among them, the overlapping mileage represents the total length of the road segments that overlap or are traversed by the two routes in space; the average heading angle represents the overall direction of the route from the starting point to the end point; and the overlapping time period represents the overlapping operating duration of the two routes in the time dimension.

[0037] Specifically, the mileage of overlapping road segments can be determined by spatial geometric algorithms, such as by buffer analysis, intersection operation or other route overlay analysis methods; the average driving heading angle can be determined by route heading angle calculation and difference comparison; the overlapping time period can be determined by time interval intersection and duration statistics, and this application does not impose any restrictions on this.

[0038] The spatial path overlap degree of any two routes is calculated based on the length of the overlapping road segment and the maximum total mileage of any two routes. Specifically, this can be calculated by dividing the length of the overlapping road segment of any two routes by the F value of any two routes. i F j The maximum total mileage is used to calculate the spatial path overlap between any two routes, with values ​​ranging from 0 to 1. A higher spatial path overlap value indicates a greater degree of spatial overlap between the two routes. The formula for calculating spatial path overlap is: (1) Among them, Sim spa This represents the spatial path overlap degree, with a value range of [0,1]. F represents i F j The length of the overlapping segment of the two travel route vectors. F respectively i F j The total distance traveled on the two complete routes For F i F j The maximum total mileage between the two routes.

[0039] The degree of overlap between any two routes is determined based on their average heading angles. Specifically, the absolute value of the difference between the average heading angles of any two routes is first calculated, and then this absolute value is normalized to obtain a degree of overlap ranging from 0 to 1. If the difference between the two heading angles is 0 degrees, the degree of overlap is 1; if the difference is 180 degrees, the degree of overlap approaches 0. By introducing this index, routes traveling in opposite directions along the same road can be effectively distinguished. The formula for calculating the degree of overlap is: (2) Among them, Sim dir (F) i ,F j ) represents F i F j The degree of overlap between the driving directions of the two driving routes, with a value ranging from [0,1]. , F respectively i F j The average heading angle over the entire journey of the two routes.

[0040] The runtime matching degree of any two routes is calculated based on the maximum of the overlapping time periods and the total runtime of those two routes. Specifically, the runtime matching degree is obtained by dividing the duration of the overlapping time period by the maximum total runtime of those two routes, resulting in a value between 0 and 1. A higher value indicates that the two routes are more likely to operate concurrently and serve the same or similar transportation tasks. The formula for calculating the runtime matching degree is: (3) Among them, Sim time (F i ,F j ) represents F i F j The runtime matching degree between the two travel routes ranges from [0,1], t ij F represents i F j The duration corresponding to the overlapping time period of the two routes, max(|t i |,|t j |) is F i F j The maximum total travel time between the two routes.

[0041] S22: Determine the overall route similarity between any two routes based on spatial path overlap, driving direction overlap, and runtime matching degree.

[0042] Specifically, spatial path overlap, driving direction overlap, and runtime matching degree can be assigned weights α, β, and γ, respectively, where α, β, and γ are all greater than 0, and α + β + γ = 1. Subsequently, a weighted summation unit is used to weight and fuse these three factors to obtain the overall route similarity. The weights α, β, and γ are not limited to fixed values ​​but can be flexibly adjusted according to the actual operational scenario. For example, in scenarios emphasizing path consistency, the weight corresponding to spatial path overlap can be appropriately increased; in scenarios emphasizing time period consistency, the weight corresponding to runtime matching degree can be appropriately increased. The formula for calculating the overall route similarity is: (4) Among them, Sim total (F i ,F j ) represents F i F j The overall similarity between the two routes ranges from [0,1].

[0043] This embodiment comprehensively judges the overall similarity of any two routes in the target vehicle's historical driving route from three aspects: spatial path overlap, driving direction overlap, and operating time matching. It can not only identify routes that are highly overlapping in space, but also exclude routes with opposite driving directions or completely different operating times, thereby providing a more accurate neighborhood determination basis for subsequent clustering analysis and reducing the risk of false clustering.

[0044] In some embodiments of this application, such as Figure 3 As shown, in step S30, the historical driving routes of the target vehicle are clustered based on the comprehensive route similarity to generate at least one cluster. Specific steps may include: S31: Based on the comprehensive similarity of the routes, determine the core routes, boundary routes of the core routes, and noise routes in the historical driving routes of the target vehicle.

[0045] Specifically, the type of each route in the historical driving route of the target vehicle can be determined in a variety of ways. For example, a similarity threshold corresponding to different route types can be set, and the relationship between the comprehensive similarity of the routes and the similarity threshold can be calculated according to the above formula (4). This directly determines the type of each route in the historical driving route of the target vehicle. The types of routes can include core routes, boundary routes of core routes, and noise routes.

[0046] S32: Based on the core routes and boundary routes of the target vehicle's historical driving routes, cluster the target vehicle's historical driving routes to generate at least one cluster.

[0047] In one implementation, any core route in the target vehicle's historical driving route is used as the starting cluster center. The boundary routes of any core route are clustered to generate an initial cluster. All core routes in the initial cluster are traversed, and the boundary routes corresponding to all core routes in the initial cluster are assigned to the initial cluster. This process continues until all core routes in the target vehicle's historical driving route have been clustered to obtain at least one cluster. Each cluster corresponds to a candidate high-frequency transportation trunk line.

[0048] Specifically, the boundary routes of the core routes can be divided into core routes and non-core routes (associated boundary routes). After classifying the historical travel routes of the target vehicle, any unclassified core route can be used as the starting center for clustering. The boundary routes of this core route are then clustered, with all boundary routes assigned to the same cluster to generate an initial cluster. This process is repeated for all core routes within the initial cluster, assigning all corresponding boundary routes to the initial cluster to obtain the cluster corresponding to that core route. This process is repeated for the remaining unclassified routes until all core routes in the target vehicle's historical travel routes are clustered, resulting in at least one cluster. Each cluster corresponds to a candidate high-frequency transport trunk line. Noisy routes can be directly eliminated and not included in trunk line generation. It is understood that the above implementation method is only one clustering approach; other known clustering methods can also be used to cluster the historical travel routes of the target vehicle, which will not be elaborated upon here.

[0049] To clearly understand this embodiment, an example is used for illustration. The target vehicle's historical driving route includes core routes A and B, boundary routes a1, a2, a3, and B of core route A, boundary routes b1, b2, b3, and b4 of core route B, and a noisy route C. Core route A is used as the initial clustering center. All boundary routes of this core route are assigned to the same cluster, generating an initial cluster {A, a1, a2, a3, B}. All core routes A and B in the initial cluster are traversed. All boundary routes corresponding to core route B in the initial cluster are assigned to the initial cluster, resulting in the cluster {A, a1, a2, a3, B, b1, b2, b3, b4} corresponding to this core route. Noisy route C is then removed.

[0050] To prioritize the identification of high-frequency transport trunk lines and suppress the formation of invalid clusters of low-frequency scattered routes, this application adds two parameters: the number of similar routes in the neighborhood and the route frequency weight, to determine the type of each route in the target vehicle's historical travel routes. In some embodiments of this application, step S31 determines the core route, the boundary routes of the core route, and noisy routes in the target vehicle's historical travel routes based on the comprehensive route similarity. Specific steps may include: S311: Based on the comprehensive route similarity, determine the number of neighboring similar routes for each route in the target vehicle's historical driving routes.

[0051] Among them, the neighborhood similarity route represents a route whose overall route similarity to a given route is greater than or equal to a neighborhood threshold. Specifically, all routes in the target vehicle's historical driving routes are paired, and the overall route similarity is used as the basis for neighborhood determination: (5) If equation (5) is satisfied, then route F is determined. j For route F i Find similar routes in the neighborhood and count each route F in the historical driving routes of the target vehicle. i The number of similar routes in the neighborhood, i.e., counting each route F i The total number of neighborhoods that satisfy the conditions is denoted as N. neigh (F i ),in, This represents the neighborhood threshold.

[0052] S312: Based on the comprehensive similarity of routes, the number of similar routes in the neighborhood of each route, and the route frequency weight of each route, determine the core routes, boundary routes of the core routes, and noisy routes in the historical driving routes of the target vehicle.

[0053] The comprehensive similarity of any two routes is determined according to formula (4) above, and the number of neighboring similar routes of each route is determined according to formula (5) above. The route frequency weight of each route is obtained, where the route frequency weight of each route is the ratio of the average daily number of vehicles passing through each route to the average frequency of all routes in the target vehicle's historical driving route. The relationship between the comprehensive similarity of routes, the number of neighboring similar routes of each route, and the route frequency weight of each route can be preset. By comparing the comprehensive similarity of routes, the number of neighboring similar routes of each route, and the route frequency weight of each route with the preset relationship, the historical driving routes of the target vehicle are classified to obtain the core routes, the boundary routes of the core routes, and the noise routes in the historical driving routes of the target vehicle.

[0054] In some implementations, step S312 determines the core routes, boundary routes, and noisy routes in the target vehicle's historical driving routes based on the comprehensive route similarity, the number of neighboring similar routes for each route, and the route frequency weight for each route. Specific steps may include: S3121: When the product of the number of similar routes in the neighborhood of the first route and the route frequency weight of the first route is greater than or equal to the first preset value, the first route is determined as the core route.

[0055] Specifically, the first preset value is used to determine whether a route can become the core route of a cluster. It represents the minimum number of similar neighboring routes required to become a core route. For example, the first preset value is set to 5, the first route is any route in the target vehicle's historical driving routes, and the route frequency weight W of the first route is... i f is the average daily number of vehicles on the first route. i The average frequency of all routes in the target vehicle's historical travel routes. The ratio, i.e., W i =f i / .

[0056] In this way, even if the number of routes around a certain route that meet the neighborhood threshold condition is relatively small, as long as the route itself is a high-frequency route, it can still be identified as a core route.

[0057] S3122: When the first route is a non-core route and the comprehensive similarity between the first route and the second route is greater than or equal to the neighborhood threshold, the first route is the boundary route of the second route.

[0058] Specifically, when the first route is a non-core route, that is, the product of the number of similar routes in the neighborhood of the first route and the route frequency weight of the first route is less than a first preset value, the relationship between the comprehensive similarity of the first route and the second route and the neighborhood threshold is determined. If the comprehensive similarity of the first route and the second route is greater than or equal to the neighborhood threshold, the first route is the boundary route of the second route, where the second route is a core route determined by S3121.

[0059] S3123: When the first route is a non-core route and the first route is a boundary route that is not the second route, the first route is a noisy route.

[0060] When the first route is a non-core route and is a boundary route of the second route, that is, when the product of the number of similar routes in the neighborhood of the first route and the route frequency weight of the first route is less than the first preset value, and the comprehensive route similarity between the first route and the second route is less than the neighborhood threshold, the first route is a noisy route.

[0061] Based on the comprehensive similarity of the routes, the historical driving routes of the target vehicles are clustered to generate at least one cluster. That is, there may be one or more clusters generated. When there is only one cluster, the cluster can directly correspond to a candidate high-frequency transportation trunk line.

[0062] When there are at least two clusters, to address the issue of redundant and noisy clusters mixed in the clustering results, this application further includes a post-processing procedure for merging and removing clusters. In some embodiments of this application, the above-mentioned method for constructing a vehicle's energy replenishment network may further include the following steps: S301: Calculate the overall route similarity between any two clusters.

[0063] Specifically, a representative core route can be selected from each of the two clusters as the central route, and then the route comprehensive similarity of the corresponding central routes of any two clusters can be calculated according to the above formula (4).

[0064] S302: When the route similarity of any two clusters is greater than or equal to the neighborhood threshold, merge the two clusters.

[0065] When the calculated route similarity between the corresponding central routes of any two clusters is greater than or equal to the neighborhood threshold, the two clusters are merged. This step is repeated to obtain at least one merged cluster.

[0066] S303: Calculate the total number of routes and total frequency of traffic in each merged cluster.

[0067] Each merged cluster contains multiple routes. Calculate the total number of routes and the total frequency of traffic within each merged cluster.

[0068] S304: When the total number of routes in the merged first cluster is less than a preset total number threshold, and / or the total frequency of travel of the routes in the first cluster is less than a preset total frequency threshold, the merged first cluster is removed.

[0069] The first cluster is any one or more clusters within the merged clusters. When the total number of routes contained in the merged first cluster is less than a preset total number threshold, the merged first cluster is removed; when the total frequency of travel on the routes contained in the first cluster is less than a preset total frequency threshold, the merged first cluster is removed; when both the total number of routes contained in the merged first cluster and the total frequency of travel on the routes contained in the first cluster are less than the preset total number threshold, the merged first cluster is removed.

[0070] By merging and removing clusters in the post-processing procedure, the problem of redundant and noisy clusters in the clustering results is effectively solved, the accuracy of high-frequency trunk line identification results is improved, and a more accurate data foundation is provided for the subsequent construction of a power supplementation network that coordinates with the transportation trunk lines.

[0071] In some implementations, when at least two clusters exist, after the application completes the cluster merging and post-processing steps, the above-mentioned step S40, which constructs the target vehicle's energy replenishment network based on at least one cluster, includes: Construct a refueling network for the target vehicle based on at least one merged cluster.

[0072] After completing the cluster merging and post-processing process, at least one merged cluster is obtained, which is a route cluster with independent structure, similar spatial morphology, and concentrated traffic flow. Each merged cluster corresponds to a candidate high-frequency transportation trunk line. Based on at least one merged cluster, the high-frequency transportation trunk line is identified more accurately, and the energy replenishment network for the target vehicles constructed is more in line with the actual operational needs.

[0073] In some embodiments of this application, such as Figure 4 As shown, in step S40, the refueling network for the target vehicle is constructed based on at least one cluster. Specific steps may include: S41: Calculate the overall score for each cluster based on the dataset containing routes in each cluster.

[0074] Each cluster contains multiple routes. The dataset containing multiple routes in each cluster is treated as a whole. The common indicators of all routes in the corresponding cluster are aggregated and statistically analyzed based on the dataset. Based on the relationship between the common indicators and the pre-set scoring rules, the comprehensive score value of each cluster is calculated.

[0075] S42: Determine the grade of the candidate high-frequency transport trunk line corresponding to each cluster based on the comprehensive score of each cluster.

[0076] The correspondence between the comprehensive score of a cluster and the level of a candidate high-frequency transport line can be preset. Based on the comprehensive score of each cluster, the level of the candidate high-frequency transport line corresponding to each cluster can be determined.

[0077] S43: Construct a refueling network for the target vehicles based on the grade of the candidate high-frequency transport trunk lines corresponding to the clusters.

[0078] The classification of candidate high-frequency transport trunk lines corresponding to clusters aims to provide a quantitative basis for the differentiated construction of subsequent energy replenishment networks and achieve optimal resource allocation. For example, primary high-frequency trunk lines, as the main national transport corridors, prioritize the deployment of high-power, high-density energy replenishment stations; secondary high-frequency trunk lines, as regional supplementary corridors, adopt a construction strategy of moderate configuration and gradual improvement. The classification results can be directly applied to determine the priority of charging station matching and the classification and spacing of energy replenishment nodes. Primary trunk lines have the highest traffic flow and the highest trunk line value, and their energy replenishment station construction is initiated first; secondary trunk lines serve as a supplement, and are gradually improved after the primary trunk line network is basically formed. Priority is given to addressing the energy replenishment gaps on primary trunk lines to avoid impacting the main transport arteries; gaps on secondary trunk lines can be addressed gradually based on regional economic development.

[0079] In some embodiments of this application, step S41 calculates a comprehensive score for each cluster based on the dataset containing routes for each cluster. Specific steps may include: S411: Based on the dataset containing routes in each cluster, calculate the traffic intensity score, transport attribute score, and transport attribute score for each cluster.

[0080] Common indicators for all routes within a cluster can include: traffic intensity, transport attributes, and road network adaptability. Similarly, the datasets containing multiple routes in each cluster are treated as a whole for data analysis to obtain data corresponding to the above common indicators, such as the average daily total number of routes in each cluster, the maximum average daily total number of routes in all clusters, the average daily traffic frequency of routes in each cluster, the maximum average daily traffic frequency of routes in all clusters, average mileage, the average driving mileage of routes in each cluster, the maximum driving mileage of routes in all clusters, the proportion of vehicles using dedicated routes in routes in each cluster, the road grade score of routes in each cluster, and the congestion level of routes in each cluster.

[0081] The traffic intensity score for each cluster is calculated based on the average daily total number of routes in each cluster, the maximum average daily total number of routes in all clusters, the average daily frequency of routes in each cluster, and the maximum average daily frequency of routes in all clusters.

[0082] The traffic intensity score characterizes the traffic volume of routes within a cluster, allowing for horizontal comparisons between different route clusters. The formula for calculating the traffic intensity score is as follows: (6) Where Score1 represents the intensity rating, n i n represents the average daily total number of routes included in the current cluster.max q represents the maximum daily average total number of routes included in all clusters. i q represents the average daily travel frequency of routes included in the current cluster. max This represents the maximum daily average traffic frequency across all clusters. λ1 and μ1 are weights, with λ1 + μ1 = 1. The larger the traffic flow and the higher the frequency within a cluster containing routes, the higher the Score1 score.

[0083] The transportation attribute score of each cluster containing routes is calculated based on the average mileage of routes contained in each cluster, the maximum mileage of routes contained in all clusters, and the proportion of vehicles running dedicated routes in each cluster containing routes.

[0084] The transport attribute score is used to characterize the route value of a cluster, distinguishing between long-distance trunk lines and short-distance feeder lines, and between stable fixed routes and temporary scattered routes. The formula for calculating the transport attribute score is as follows: (7) Where Score2 represents the transportation attribute score, L i L represents the average mileage of all routes within the current route cluster. max R represents the maximum average mileage of routes included in all clusters. stable This represents the proportion of vehicles running dedicated routes within a cluster, with values ​​ranging from [0,1]. λ² and μ² are weights, where λ² + μ² = 1. A longer average mileage and a higher proportion of vehicles running dedicated routes result in a higher transport attribute score and greater trunk line value.

[0085] Based on the road grade score and congestion level of the routes contained in each cluster, the road network suitability score of the routes contained in each cluster is calculated.

[0086] The road network adaptability score characterizes the road network quality of routes within a cluster, indicating whether it is predominantly high-speed and whether traffic flow is smooth. The formula for calculating the road network adaptability score is: (8) Where Score3 represents the road network adaptability score, G road This indicates the road grade score of the routes contained in the cluster, with fixed scores assigned according to grade: expressway > national highway > provincial highway > county road > township road; Q flowThe congestion level of the routes contained in a cluster is represented by the road level score, with the capacity ranging from [0,1] based on the congestion level. λ3 and μ3 are the weights, with λ3 + μ3 = 1. Road networks with high road level scores and route clusters with high traffic flow are more suitable as long-distance transport and energy replenishment trunk lines. For example, using a standard of 100 points for expressways, 80 points for national highways, 60 points for provincial highways, 30 points for county roads, and 10 points for rural and urban roads, the road level score of the routes contained in the cluster is calculated with road segment mileage as the weight. The congestion level of a route can be calculated based on third-party traffic situation data or proprietary trajectory speed data; a higher value indicates smoother traffic flow.

[0087] S412: Determine the comprehensive score for each cluster based on the traffic intensity score, transport attribute score, and road network adaptability score of the routes contained in each cluster.

[0088] The three dimensions of traffic intensity, transport attributes, and road network adaptability are integrated into a single comprehensive score to achieve standardized quantitative evaluation of each cluster. The formula for calculating the comprehensive score of a cluster is as follows: (9) Among them, Score total The comprehensive score of the cluster is represented by a, b, and c, which represent the weights of traffic intensity, transportation attributes, and road network adaptability, respectively. a+b+c=1, and the total score ranges from [0,100].

[0089] It should be noted that the various weight parameters in this application are dynamically determined by combining expert experience with cross-validation of business scenarios, ensuring that the weights not only conform to common sense in the electric commercial vehicle transportation industry, but also adapt to the differentiated needs of different application scenarios.

[0090] In some embodiments of this application, step S42 determines the level of the candidate high-frequency transport trunk line corresponding to each cluster based on the comprehensive score of each cluster. Specific steps may include: S421: When the comprehensive score of a cluster is greater than or equal to the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a first-level high-frequency trunk line.

[0091] Among them, the primary high-frequency trunk line can be characterized as the core main channel with the highest comprehensive score, large traffic volume and better road network conditions.

[0092] S422: When the comprehensive score of a cluster is greater than or equal to the second comprehensive score threshold and less than the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a secondary high-frequency trunk line.

[0093] Among them, secondary high-frequency trunk lines can represent corridors with moderate overall performance, undertaking regional or auxiliary transportation tasks. The first and second comprehensive scoring thresholds can be set according to specific circumstances; for example, the first comprehensive scoring threshold can be set to 80 points, and the second comprehensive scoring threshold can be set to 60 points.

[0094] S423: When the comprehensive score of a cluster is less than the second comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a low-frequency scattered trunk line.

[0095] Among them, low-frequency scattered trunk lines can represent temporary or short-distance delivery routes with relatively low transportation value and traffic volume.

[0096] In some embodiments of this application, step S43 involves constructing a refueling network for the target vehicle based on the grade of the candidate high-frequency transport trunk lines corresponding to the clusters. Specific steps may include: S431: Obtain the current charging station information for each candidate high-frequency transport trunk line corresponding to each cluster.

[0097] Obtain charging station information for each candidate high-frequency transport trunk line corresponding to each cluster. The current charging station information includes the current location, charging capacity, operating status, and number of charging stations. The location of the charging station can be the geographical coordinates (longitude and latitude), and the charging capacity of the charging station is the charging speed. Optionally, charging stations with charging capacity corresponding to charging power less than the preset charging power and / or operating status of being out of service are excluded.

[0098] S432: Spatially associate the candidate high-frequency transport trunk lines corresponding to each cluster with the current charging station to construct the energy replenishment network for the target vehicle.

[0099] Using the trunk line centerline as a reference, a spatial buffer analysis method is adopted to delineate a circular radiation range with a preset radius, such as a preset radius of 5km. Through spatial overlay analysis, the dedicated charging stations of target vehicles within the radiation range of each trunk line are retrieved in batches. The location, charging capacity, spatial distribution and number of charging stations corresponding to each candidate high-frequency transport trunk line are obtained, and the location information of the charging stations is unified with the location coordinates of the candidate high-frequency transport trunk lines.

[0100] Additionally, the spatial distribution of charging stations can be obtained. This spatial distribution can be calculated by recording the corresponding position of each charging station on the main line and calculating the average distance between adjacent stations. Furthermore, the charging capacity can be obtained by summing the rated charging power of all stations along the line to obtain the total power, and then taking the average value to obtain the average charging speed.

[0101] The charging stations are screened again to remove those that exceed the preset spatial distribution or have abnormal operating status (such as temporary shutdown), ensuring that the matched stations can meet the actual charging needs of the target vehicles.

[0102] S433: Determine the target charging station information for the candidate high-frequency transport trunk line corresponding to the cluster based on the level of the candidate high-frequency transport trunk line corresponding to the cluster.

[0103] The correspondence between the candidate high-frequency transport trunk lines' levels and target charging station information can be pre-defined to determine station density, spatial distribution, and charging capacity. Based on the levels of the candidate high-frequency transport trunk lines corresponding to clusters, a quantitative basis is provided for the differentiated construction of the subsequent energy replenishment network, achieving optimal resource allocation. High-frequency trunk lines are spatially associated with heavy-duty truck charging stations along their routes, integrating them into a "trunk line + charging station" energy replenishment network. This determines the geographical coverage of the network and the distribution of core energy replenishment points, thus identifying the target charging station information for the candidate high-frequency transport trunk lines corresponding to the clusters.

[0104] S434: Based on the target charging station information of the candidate high-frequency transport trunk lines corresponding to the clusters, the charging station information corresponding to the locations that need to be recharged in the candidate high-frequency transport trunk lines corresponding to the clusters is labeled, and a recharge network dataset is formed.

[0105] Each candidate high-frequency transport route is traversed segment by segment, and the length of the road segment without a continuous effective heavy truck charging station is calculated. If the length exceeds the average range of the target vehicle, it is marked as a charging blank road segment and a key optimization area, forming a charging network dataset. The charging network dataset includes vector information of candidate high-frequency transport routes, detailed attributes of charging stations along the route, charging node annotation results, and the location and length of charging blank road segments.

[0106] Exemplary device In some embodiments of this application, such as Figure 5 As shown, this application provides a vehicle refueling network construction device, including: a data acquisition module 501, a calculation module 502, a route clustering module 502, and a refueling network construction module 504, wherein, The data acquisition module 501 is used to acquire a dataset of the historical driving routes of the target vehicle.

[0107] The calculation module 502 is used to calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle based on the historical driving route dataset.

[0108] The route clustering module 503 is used to cluster the historical driving routes of the target vehicle based on the comprehensive similarity of the routes, and generate at least one cluster.

[0109] The energy replenishment network construction module 504 is used to construct an energy replenishment network for the target vehicle based on at least one cluster.

[0110] The vehicle refueling network construction device provided in this application acquires a dataset of the target vehicle's historical driving routes through a data acquisition module 501, and transmits it to a calculation module 502. The calculation module 502 calculates the comprehensive similarity of any two routes in the target vehicle's historical driving routes based on the historical driving route dataset, and transmits it to a route clustering module 503. The route clustering module 503 clusters the target vehicle's historical driving routes according to the comprehensive similarity of the routes, generating at least one cluster. The refueling network construction module 504 constructs the target vehicle's refueling network based on the at least one cluster generated by the route clustering module 503.

[0111] The vehicle refueling network construction device provided in this application uses a complete dataset of historical driving routes as the analysis object. By calculating the comprehensive similarity between any two historical routes and clustering the routes, it generates clusters representing high-frequency transportation trunk lines. This avoids directly using historical discrete trajectory points as analysis units to identify local hotspots and scattered trajectory clusters. Therefore, it can more accurately and completely identify high-frequency transportation trunk lines from the historical driving route trajectories of vehicles, and construct a vehicle refueling network that matches actual operational needs accordingly.

[0112] The vehicle refueling network construction device provided in this embodiment belongs to the same application concept as the vehicle refueling network construction method provided in the above embodiments of this application. It can execute the vehicle refueling network construction method provided in any of the above embodiments of this application and has the corresponding functional units and beneficial effects of the vehicle refueling network construction method. Technical details not described in detail in this embodiment can be found in the specific processing content of the vehicle refueling network construction method provided in the above embodiments of this application, and will not be repeated here.

[0113] Exemplary electronic devices In some embodiments of this application, an electronic device is provided, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the vehicle power replenishment network construction method described above.

[0114] Specifically, the aforementioned electronic device may further include: a bus, a communication interface, input devices, and output devices. The processor, memory, communication interface, input devices, and output devices are interconnected via the bus. The bus may include a pathway for transmitting information between various components of the computer system.

[0115] The processor can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0116] The processor may include the main processor, as well as baseband chips, modems, etc.

[0117] The memory stores a computer program that executes the technical solution of this application, and may also store an operating system and other critical business functions. Specifically, the program may include program code, which includes computer operation instructions. More specifically, the memory may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, etc.

[0118] The computer program can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this specification. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0119] Input devices may include devices that receive data and information input by the user, such as keyboards, mice, cameras, scanners, light pens, voice input devices, touch screens, pedometers, or gravity sensors.

[0120] Output devices may include devices that allow information to be output to the user, such as displays, printers, speakers, etc.

[0121] The communication interface may include any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0122] The processor executes the program stored in the memory and calls other devices, which can be used to implement the various steps of the vehicle energy replenishment network construction method provided in the above embodiments of this application.

[0123] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are 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.

[0124] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in hardware or as a software functional module. This disclosure is not limited to any particular combination of hardware and software.

[0125] Unless otherwise defined, all terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. It should also be understood that terms such as those defined in a common dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not as having an idealized or highly formalized meaning, unless expressly defined herein.

[0126] The foregoing description is intended to illustrate the present disclosure and should not be construed as limiting it. While several exemplary embodiments of the present disclosure have been described, those skilled in the art will readily understand that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present disclosure. Therefore, all such modifications are intended to be included within the scope of the present disclosure as defined by the claims. It should be understood that the foregoing description is intended to illustrate the present disclosure and should not be construed as limiting it to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present disclosure is defined by the claims and their equivalents.

Claims

1. A method for constructing a vehicle energy replenishment network, characterized in that, include: Obtain a dataset of the target vehicle's historical driving routes; Based on the dataset of historical driving routes, calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle. Based on the comprehensive similarity of the routes, the historical driving routes of the target vehicle are clustered to generate at least one cluster. Based on the at least one cluster, construct the refueling network for the target vehicle.

2. The method for constructing a vehicle energy replenishment network according to claim 1, characterized in that, The step of calculating the comprehensive route similarity between any two routes in the target vehicle's historical driving routes based on the historical driving route dataset includes: Based on the historical driving route dataset, calculate the spatial path overlap, driving direction overlap, and runtime matching degree of any two routes in the historical driving route of the target vehicle. The overall route similarity between any two routes is determined based on the spatial path overlap, the driving direction overlap, and the runtime matching degree.

3. The method for constructing a vehicle energy replenishment network according to claim 1, characterized in that, The step of clustering the historical driving routes of the target vehicle based on the comprehensive route similarity to generate at least one cluster includes: Based on the comprehensive similarity of the routes, the core routes, boundary routes of the core routes, and noisy routes in the historical driving routes of the target vehicle are determined. Based on the core routes and boundary routes of the target vehicle's historical driving routes, the historical driving routes of the target vehicle are clustered to generate at least one cluster, wherein each cluster corresponds to a candidate high-frequency transportation trunk line.

4. The method for constructing a vehicle's energy replenishment network according to claim 3, characterized in that, The step of determining the core route, the boundary route of the core route, and the noisy route in the target vehicle's historical driving route based on the comprehensive route similarity includes: Based on the comprehensive route similarity, the number of neighboring similar routes for each route in the target vehicle's historical driving routes is determined, wherein the neighboring similar routes represent routes whose comprehensive route similarity with each route is greater than or equal to a neighborhood threshold. Based on the comprehensive similarity of the routes, the number of neighboring similar routes for each route, and the route frequency weight for each route, the core routes, boundary routes of the core routes, and noisy routes in the historical driving routes of the target vehicle are determined.

5. The method for constructing a vehicle's energy replenishment network according to claim 3, characterized in that, When at least two clusters exist, the method further includes: Calculate the overall route similarity between any two clusters; When the route similarity of any two clusters is greater than or equal to the neighborhood threshold, the two clusters are merged. Calculate the total number of routes and total frequency of travel contained in each merged cluster; When the total number of routes in the merged first cluster is less than a preset total number threshold, and / or the total frequency of travel of the routes in the first cluster is less than a preset total frequency threshold, the merged first cluster is removed.

6. The method for constructing a vehicle's energy replenishment network according to claim 3, characterized in that, The step of constructing the refueling network for the target vehicle based on the at least one cluster includes: Based on the dataset containing routes in each cluster, calculate the comprehensive score for each cluster; Based on the comprehensive score of each cluster, the level of the candidate high-frequency transport trunk line corresponding to the cluster is determined; Based on the grade of the candidate high-frequency transport trunk lines corresponding to the clusters, a refueling network for the target vehicles is constructed.

7. The method for constructing a vehicle energy replenishment network according to claim 6, characterized in that, Based on the dataset containing routes for each cluster, a comprehensive score is calculated for each cluster, including: Based on the dataset containing routes in each cluster, a traffic intensity score, a transport attribute score, and a road network suitability score are calculated for each cluster. The traffic intensity score is used to characterize the traffic volume of the routes contained in the cluster, the transport attribute score is used to characterize the route value of the routes contained in the cluster, and the road network suitability score is used to characterize the road network quality of the routes contained in the cluster. The comprehensive score for each cluster is determined based on the traffic intensity score, transport attribute score, and road network adaptability score of the routes contained in each cluster.

8. The method for constructing a vehicle energy replenishment network according to claim 6, characterized in that, The step of determining the level of the candidate high-frequency transport trunk line corresponding to each cluster based on the comprehensive score of each cluster includes: When the comprehensive score of a cluster is greater than or equal to the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a first-level high-frequency trunk line. When the comprehensive score of a cluster is greater than or equal to the second comprehensive score threshold and less than the first comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a secondary high-frequency trunk line. When the comprehensive score of a cluster is less than the second comprehensive score threshold, the candidate high-frequency transport trunk line corresponding to the cluster is a low-frequency scattered trunk line.

9. The method for constructing a vehicle's energy replenishment network according to claim 6, characterized in that, The step of constructing the energy replenishment network for the target vehicle based on the grade of the candidate high-frequency transport trunk lines corresponding to the clusters includes: Obtain the current charging station information for each candidate high-frequency transport trunk line corresponding to each cluster. The current charging station information includes the location, charging capacity, operating status, and number of the current charging station. The candidate high-frequency transport trunk lines corresponding to each cluster are spatially associated with the current charging station to construct the energy replenishment network for the target vehicle. Based on the grade of the candidate high-frequency transport trunk line corresponding to the cluster, determine the target charging station information of the candidate high-frequency transport trunk line corresponding to the cluster; Based on the target charging station information of the candidate high-frequency transport trunk lines corresponding to the clusters, the charging station information corresponding to the locations that need to be recharged in the candidate high-frequency transport trunk lines corresponding to the clusters is labeled, and a recharge network dataset is formed.

10. A device for constructing a vehicle's energy replenishment network, characterized in that, include: The data acquisition module is used to acquire a dataset of the target vehicle's historical driving routes. The calculation module is used to calculate the comprehensive similarity between any two routes in the historical driving routes of the target vehicle based on the historical driving route dataset. The route clustering module is used to cluster the historical driving routes of the target vehicle based on the comprehensive similarity of the routes, and generate at least one cluster. A power replenishment network construction module is used to construct a power replenishment network for the target vehicle based on the at least one cluster.

Images