A method, device, equipment, medium and product for dynamically identifying a physical boundary of a charging station

Abstract

Embodiments of the present application provide a charging station physical boundary dynamic identification method, device, equipment, medium and product, which utilizes the method to perform charging station physical boundary dynamic identification based on residence characteristics and charging piles, fuses multi-source space-time data including user residence behavior characteristics, charging pile space distribution, road network topology structure, building contour and other multi-dimensional information, aggregates effective user behavior data through a pile-driven residence point clustering algorithm, performs concave boundary generalization through a contour extraction algorithm, realizes physical element fitting in combination with road network cutting and building contour adsorption, and ensures boundary geometry effectiveness and business compliance through topological optimization and business rule checking. The embodiments form a full-process capability of data perception-space calculation-intelligent decision-making-closed loop verification, realize dynamic identification of charging station physical boundary, improve charging station boundary identification precision, and break through the limitations of traditional schemes relying on manual experience, rigid boundary and lagging update.

Description

Technical Field

[0001] This invention relates to the field of charging station technology, and in particular to a method, device, equipment, medium and product for dynamic identification of the physical boundaries of a charging station. Background Technology

[0002] Physical boundary identification of charging stations is an essential requirement for the refined and intelligent management of large-scale and complex charging stations following the widespread adoption of electric vehicles. Clearly defining the boundaries and internal layout of charging stations is fundamental for asset maintenance, utilization analysis, and future expansion planning. Traditional methods relying on manual management or simple pile signal detection are no longer sufficient to meet the challenges of large-scale, efficient operation. Therefore, how to systematically and "automatically and intelligently" sense and identify the physical boundaries of charging stations has become a key technical issue.

[0003] The main existing methods for charging station boundary identification are as follows: 1) static identification methods based on fixed geometric boundaries; 2) methods based on manual polygon boundary drawing; and 3) methods based on high-precision Global Navigation Satellite System (GNSS) / Real-Time Kinematic (RTK) mapping. These methods all have shortcomings in terms of application scope, data accuracy, and cost. Therefore, a method is needed to address these shortcomings. Summary of the Invention

[0004] This invention provides a method, device, equipment, medium, and product for dynamic identification of the physical boundaries of charging stations. It realizes dynamic identification of the physical boundaries of charging stations, improves the accuracy of charging station boundary identification, and breaks through the limitations of traditional solutions that rely on human experience, have rigid boundaries, and are slow to update.

[0005] Firstly, this embodiment provides a method for dynamic identification of the physical boundaries of a charging station, the method comprising: Construct multidimensional dwelling feature data for each dwelling point within the effective identification range of the charging station, and perform clustering processing on the multidimensional dwelling feature data to obtain the dwelling behavior cluster of the charging station; Contour extraction is performed on the cluster of dwelling behaviors to generate the initial physical boundary of the charging station; Based on the road network information within the preset range of the initial physical boundary, the initial physical boundary is corrected to obtain the first corrected boundary of the charging station after road adaptation. Based on the building information within the preset range of the first correction boundary, the first correction boundary is corrected to obtain a second correction boundary of the charging station that conforms to the building layout. The second modified boundary is subjected to topology optimization processing to obtain the final physical boundary of the charging station with complete topology and simple structure; The final physical boundary is subjected to multi-dimensional compliance verification to obtain the verification result of the final physical boundary of the charging station.

[0006] Secondly, this embodiment provides a dynamic identification device for the physical boundary of a charging station, the device comprising: The clustering module is used to construct multi-dimensional dwell feature data of each dwelling point within the effective identification range of the charging station, and to perform clustering processing on the multi-dimensional dwell feature data to obtain the dwell behavior clusters of the charging station. An initial boundary determination module is used to extract the contours of the dwelling behavior clusters and generate the initial physical boundary of the charging station. The first boundary correction module is used to correct the initial physical boundary based on the road network information within the preset range of the initial physical boundary, so as to obtain the first corrected boundary of the charging station after road adaptation. The second boundary correction module is used to correct the first correction boundary based on the building information within the preset range of the first correction boundary, so as to obtain the second correction boundary of the charging station after conforming to the building layout. The third boundary correction module is used to perform topology optimization processing on the second corrected boundary to obtain the final physical boundary of the charging station with complete topology and simple structure. The verification module is used to perform multi-dimensional compliance verification on the final physical boundary and obtain the verification result of the final physical boundary of the charging station.

[0007] Thirdly, this embodiment provides an electronic device, including: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the dynamic identification method for the physical boundary of a charging station according to any embodiment of the present invention.

[0008] Fourthly, this embodiment provides a computer-readable storage medium storing computer instructions that cause a processor to execute and implement the dynamic identification method for physical boundaries of charging stations as described in any embodiment of the present invention.

[0009] Fifthly, embodiments of the present invention also provide a computer program product, the computer program product including a computer program, which, when executed by a processor, implements the dynamic identification method for the physical boundary of a charging station as described in any embodiment of the present invention.

[0010] This invention provides a method, apparatus, device, medium, and product for dynamic identification of the physical boundary of a charging station. The method includes: constructing multi-dimensional dwelling feature data of each dwelling point within the effective identification range of the charging station, and performing clustering processing on the multi-dimensional dwelling feature data to obtain dwelling behavior clusters of the charging station; extracting contours from the dwelling behavior clusters to generate an initial physical boundary of the charging station; correcting the initial physical boundary based on road network information within a preset range of the initial physical boundary to obtain a first corrected boundary of the charging station after road adaptation; correcting the first corrected boundary based on building information within a preset range of the first corrected boundary to obtain a second corrected boundary of the charging station after conforming to the building layout; performing topology optimization processing on the second corrected boundary to obtain a final physical boundary of the charging station with complete topology and simple structure; and performing multi-dimensional compliance verification on the final physical boundary to obtain the verification result of the final physical boundary of the charging station. The aforementioned technical solution dynamically identifies the physical boundaries of charging stations based on dwell characteristics and charging pile locations. By fusing multi-source spatiotemporal data, including user dwell behavior characteristics, charging pile spatial distribution, road network topology, and building outlines, it aggregates effective user behavior data through a dwell point clustering algorithm driven by charging piles. A contour extraction algorithm is used for concave boundary generalization, and road network segmentation and building outline adsorption are combined to achieve physical element alignment. Topology optimization and business rule verification are employed to ensure the geometric validity and business compliance of the boundaries. This forms a complete process capability encompassing data perception, spatial computation, intelligent decision-making, and closed-loop verification, enabling dynamic identification of charging station physical boundaries. It supports meter-level practical accuracy in charging station boundary identification, improving accuracy and overcoming the limitations of traditional solutions that rely on human experience, have rigid boundaries, and suffer from delayed updates. Furthermore, it achieves dynamic boundary identification based on real user behavior data, automatically adapting to spatial changes in the charging station's service area. While ensuring boundary accuracy and physical alignment, it significantly reduces labor costs and update delays, providing an efficient, accurate, and traceable boundary identification solution for charging station operation management, expansion planning, and compliance supervision.

[0011] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a method for dynamic identification of the physical boundary of a charging station according to Embodiment 1 of the present invention. Figure 2 This is a flowchart illustrating another method for dynamic identification of the physical boundary of a charging station provided in Embodiment 2 of the present invention. Figure 3 This is a schematic diagram of the structure of a charging station physical boundary dynamic identification device provided in Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0014] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0015] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0016] It should be noted that the main existing methods for charging station boundary identification are as follows: Existing technical solutions include: 1. Static identification method based on fixed geometric boundaries: A circular buffer zone method is used, drawing a circular service area centered on the charging station with a fixed radius (e.g., 500m-2km). This method is simple to implement but cannot reflect the spatial heterogeneity of the actual service area. 2. Method based on manual polygon boundary drawing: An irregular boundary drawing method based on geographic elements and human experience, combining administrative divisions, road network structures, or human experience to draw irregular polygonal boundaries. This method can consider topographic constraints but lacks dynamism. 3. Method based on high-precision GNSS / RTK mapping: Using high-precision global navigation satellite systems (such as BeiDou, Global Positioning System), especially real-time dynamic differential (RTK) technology, to collect boundary point coordinates. This method offers extremely high accuracy (centimeter-level), is an industry standard method, and provides authoritative and reliable results. It is suitable for statutory surveying, property rights demarcation, and high-precision map production requiring centimeter-level accuracy. However, it requires specialized equipment and personnel, is time-consuming, costly, and cannot be automatically updated.

[0017] Several existing technical solutions have shortcomings in terms of application scope, data accuracy, and cost, specifically: Existing technical solution 1 – a static identification method based on fixed geometric boundaries – cannot accurately identify the actual physical boundaries of existing charging stations. The fixed radius assumes that the physical boundaries of the charging station are circular in all directions, but in reality, the physical boundaries of charging stations are often irregular in shape. For example, a charging station might be located on one side of a road, with a parking lot to the east, buildings to the west, a green belt to the south, and a road to the north. The actual physical boundary should follow these actual geographical features, not a simple circle. This method provides insufficient accuracy in identifying physical boundaries. The fixed radius method provides boundaries that are too coarse to meet the needs of refined management. The basic characteristics of this solution can be summarized as: inaccurate identification of actual physical boundaries, neglect of geographical environmental constraints, lack of boundary change identification, and insufficient boundary accuracy.

[0018] Existing technical solution 2 - a method based on manual polygon boundary drawing: This method relies heavily on manual drawing, which is highly subjective. Different people may have different identifications of the physical boundaries of the same charging station, lacking objective standards. Human judgment is easily affected by factors such as personal cognition, experience level, and comprehension bias, leading to inconsistent boundary recognition results and making it difficult to guarantee the accuracy and reliability of the recognition results; this method requires manual boundary drawing and updating, which is costly. The basic characteristics of this solution can be summarized as follows: strong subjectivity of manual drawing, high cost of updating and maintenance, lack of data support for verification, poor adaptability to dynamic changes, insufficient standardization, difficulty in accuracy control, and low efficiency in large-scale application.

[0019] Existing technical solutions are based on high-precision GNSS / RTK mapping methods. These methods require specialized equipment such as high-precision GNSS receivers, RTK devices, and data processing software, resulting in high equipment costs. Furthermore, they require skilled technicians for operation and maintenance, leading to high labor costs. This method is essentially a "contact" survey, requiring technicians to visit each boundary feature point with a mobile station to collect data. For large-scale or complex-boundary charging stations, field data collection is time-consuming. Moreover, if the station layout changes, a complete set of field measurements must be performed again, making automated, real-time boundary updates and monitoring impossible. The basic characteristics of this solution can be summarized as: strong dependence on external equipment, high cost, insufficient real-time performance, and complex data processing. Therefore, a method is needed to address these shortcomings.

[0020] Example 1 Figure 1 This is a flowchart illustrating a method for dynamically identifying the physical boundary of a charging station according to Embodiment 1 of the present invention. This method is applicable to situations where the physical boundary of a charging station is dynamically identified. This method can be executed by a dynamic identification device for the physical boundary of a charging station, which can be implemented in hardware and / or software and is generally integrated into an electronic device.

[0021] like Figure 1 As shown, the method for dynamic identification of the physical boundary of a charging station provided in this embodiment may specifically include the following steps: S101. Construct multi-dimensional dwelling feature data for each dwelling point within the effective identification range of the charging station, and perform clustering processing on the multi-dimensional dwelling feature data to obtain the dwelling behavior cluster of the charging station.

[0022] This step is used to aggregate charging station-driven dwell points. Charging station information may include charging station identifiers, station locations, etc. Dwelling behavior data may include time, charging station identifiers, station locations, user identifiers, dwelling duration, dwelling frequency, and other auxiliary fields. In this embodiment, using known charging station information and basic user dwelling day details, the dwelling behavior of users within a specific charging station area is statistically analyzed and studied to understand the dwelling characteristics, behavioral habits, and usage patterns of users in that area. In the data preparation and feature data construction stage, the data used is not only the original list of dwelling point latitude and longitude coordinates, but also associates each dwelling point with known charging station locations. In addition to spatial coordinates, behavioral characteristics such as dwelling duration and dwelling frequency are also included in the clustering analysis considerations, and the associated data is recorded as multidimensional dwelling feature data for each dwelling point. Subsequently, the multidimensional dwelling feature data containing multidimensional factors is used to identify the physical boundaries of the charging station. For example, a density-based clustering algorithm (DBSCAN) is used to scan all dwell points for dwell point clustering using two core parameters: neighborhood radius and minimum number of points required for core points. Based on the spatial and behavioral distribution density of data points, it automatically divides the data into different categories, namely dwell behavior clusters. Dwell behavior clusters can be further divided into core charging behavior clusters and associated auxiliary behavior clusters.

[0023] Before clustering, valid dwell points can be selected based on dwell time exceeding a set time and dwell frequency exceeding a set frequency, excluding noise points within the road buffer zone. Valid dwell points are then selected from these valid dwell points, and subsequent clustering is performed based on their multidimensional dwell characteristic data. Furthermore, the dwell time of each valid dwell point is used as a weight to weight each valid dwell point, obtaining cluster centers. Then, adjacent clusters with small center distances are merged to obtain various dwell behavior clusters, which serve as the dwell behavior clusters for the charging station.

[0024] S102. Extract the contours of the dwell behavior cluster to generate the initial physical boundary of the charging station.

[0025] This step is used to transform the discrete set of dwelling points into preliminary boundary polygons based on the dwelling behavior cluster data from the previous steps, thereby generalizing the boundary and forming the initial outline of the physical boundary of the charging station, which is denoted as the initial physical boundary of the charging station.

[0026] Understandably, the aforementioned obtaining of dwell behavior clusters is equivalent to a point set. This step is used to enhance the point set, i.e., to obtain enhanced dwell behavior clusters. First, the geometric center of each dwell behavior cluster is calculated, denoted as the cluster center, and added as a new, mandatory point to the dwell behavior cluster. The cluster center represents the "center of gravity" of the entire cluster. Even if there is a small amount of noise or fluctuation at the edge points, introducing the cluster center can act as an "anchor" to ensure that the generated polygonal core area does not shift excessively, enhancing the stability of the boundary and preventing the boundary from missing important core areas due to uneven distribution of discrete points. Adding the cluster center serves as a key anchor point. Then, virtual boundary points are inserted to prevent excessive concavity. At the edge of the dwell behavior cluster, some virtual points are intelligently inserted as a "catch-all" based on the distribution density of the dwell behavior cluster. In real data, the points located at the very edge may be noise points or extremely rare behaviors. If concave boundaries are directly generated based on these points, it may lead to unreasonable, extreme "concavities" or "jagged edges" on the boundary. Inserting virtual points is equivalent to setting a smoothing constraint, preventing the algorithm from "overfitting" to noisy points and making the boundaries more natural and reasonable while preserving details. The parameters involved in this step include: the augmented point set, anchor point positions, virtual boundary points, and point set density.

[0027] Then, the enhanced dwell behavior clusters are triangulated using a triangulation algorithm to form a triangular mesh. A contour extraction algorithm is then used to extract the contours of the triangular mesh, generating concave polygons, denoted as polygons. Finally, the charging piles are forcibly included within the physical boundary. Specifically, the convex hull region of each charging pile is determined, and the intersection of the polygons and the convex hull regions is taken to ensure that all charging piles are within the boundary, obtaining the initial boundary polygon, denoted as the initial physical boundary of the charging station.

[0028] The parameters involved in this step mainly include: cluster ID, boundary polygon, preset radius threshold, boundary precision, and boundary type. The goal is not to draw a simple box, but to generate a fine-grained boundary that reflects real-world spatial usage and may include "recesses."

[0029] S103. Based on the road network information within the preset range of the initial physical boundary, the initial physical boundary is corrected to obtain the first corrected boundary of the charging station after road adaptation.

[0030] In this embodiment, the preset range can be set according to actual conditions. The boundary obtained by correcting the initial physical boundary based on road network information is denoted as the first corrected boundary of the charging station. This step is used for road network cutting. Based on road network information, the boundary is made to fit the actual road isolation, the boundary shape is corrected, and the first corrected boundary of the charging station after road adaptation is obtained. First, the road network information within the preset range of the initial physical boundary is obtained. The road is classified according to the road level and function in the road network information to obtain each road category. The roads are divided into three main levels: hard barrier main roads, soft barrier secondary arterial roads, and mergeable branch roads.

[0031] For main roads with physical barriers, the approach is to cut them off. These include highways, urban expressways, arterial roads, and major roads with central medians. These roads have high vehicle speeds and heavy traffic, and are typically physically separated. Users cannot cross highways to reach charging stations, and the physical service area of ​​charging stations cannot extend to the other side of the road. Therefore, they constitute natural and absolute spatial boundaries. Extending the boundary of a charging station to the other side of the road is meaningless and would introduce a large amount of irrelevant area, severely compromising the accuracy of the boundary. Therefore, it is necessary to remove them from the initial physical boundary.

[0032] For secondary arterial roads with soft barriers, the approach is offsetting. This includes types such as urban secondary arterial roads, regional arterial roads, and roads with high traffic volume but no central median. To simulate realistic access routes and safety buffer zones, the road is not directly cut off; instead, the boundary is offset parallel to the inward direction of the charging station by a set distance. This acknowledges the road's barrier effect while maintaining the integrity and usability of the charging station area.

[0033] For roads that can be integrated, a smoothing approach is used. These include urban side roads, internal roads within residential areas, and auxiliary roads primarily for non-motorized vehicles. These roads have low traffic volume and slow speeds, and may themselves be entrance roads to charging stations or their extensions. They do not constitute spatial barriers, and the service area of ​​a charging station can fully cover both sides of the road. Cutting or shifting these side roads would disrupt what should be a continuous charging area, resulting in fragmented boundaries. Therefore, the areas on both sides of the side road are preserved, and only the boundaries crossing the side road are smoothed to ensure the integrity of the entire service area.

[0034] The parameters involved in this step mainly include: road network, post-cut boundary, road grade, and cutting parameters.

[0035] S104. Based on the building information within the preset range of the first correction boundary, the first correction boundary is corrected to obtain a second correction boundary of the charging station that conforms to the building layout.

[0036] The building classification includes charging-related buildings and obstacle buildings. For example, charging-related buildings include charging distribution rooms / rest rooms. The preset range can be set according to actual conditions; for example, the preset range can be 50m. The boundary after correcting the first correction boundary based on the building information is denoted as the second correction boundary.

[0037] This step is used to achieve building outline snapping, ensuring the boundary conforms to the physical building edge based on the building outline information. Using the spatial query function of the Shapely library, all buildings intersecting with or within this preset buffer zone are quickly identified. For the selected buildings, a pre-trained building classification model is used for functional identification. The geometry library's fusion or merging operation is used to directly merge the polygons of "charging-related buildings" with the boundary polygons of the current first correction boundary. For obstacle buildings, the nearest edge from the first correction boundary to the obstacle building is calculated and offset along the normal direction to the obstacle building edge plus a set safety distance. Based on the above operations, a second correction boundary for the charging station, conforming to the building layout, is obtained. The parameters involved in this step mainly include: building information, snapped boundary, building type, and safety distance.

[0038] S105. Perform topology optimization processing on the second modified boundary to obtain the final physical boundary of the charging station with complete topology and simple structure.

[0039] This step optimizes the topology of the charging station boundary. The boundary obtained after topology optimization of the second corrected boundary is denoted as the final physical boundary. This step eliminates redundant vertices by correcting geometric errors and simplifying the boundary, ensuring the geometric correctness and simplicity of the boundary. Geometric defect repair, redundant vertex elimination, and fragment regularization are performed on the second corrected boundary to obtain the final physical boundary of the charging station with a complete topology and simple structure. For example, this step can use the Douglas-Peucker algorithm for topology optimization, which is used to reduce the number of curve points to achieve an approximate representation. The parameters involved in this step mainly include: geometric repair, simplified boundary, topology state, and optimization parameters.

[0040] S106. Perform multi-dimensional compliance verification on the final physical boundary to obtain the verification result of the final physical boundary of the charging station.

[0041] This step verifies the physical characteristics of the charging station to ensure that the boundaries conform to these characteristics, providing final quality control and verification. Multi-dimensional compliance verification includes, but is not limited to: area range verification, charging pile quantity verification, thermal density verification, and connectivity verification. The final physical boundaries undergo multi-dimensional compliance verification, and based on the verification results, corrections, alarms, or processing suggestions are generated. The parameters involved in this step mainly include: verification rules, verification results, alarm information, and processing suggestions.

[0042] The aforementioned technical solution dynamically identifies the physical boundaries of charging stations based on dwell characteristics and charging pile locations. By fusing multi-source spatiotemporal data, including user dwell behavior characteristics, charging pile spatial distribution, road network topology, and building outlines, it aggregates effective user behavior data through a dwell point clustering algorithm driven by charging piles. A contour extraction algorithm is used for concave boundary generalization, and road network segmentation and building outline adsorption are combined to achieve physical element alignment. Topology optimization and business rule verification are employed to ensure the geometric validity and business compliance of the boundaries. This forms a complete process capability encompassing data perception, spatial computation, intelligent decision-making, and closed-loop verification, enabling dynamic identification of charging station physical boundaries. It supports meter-level practical accuracy in charging station boundary identification, improving accuracy and overcoming the limitations of traditional solutions that rely on human experience, have rigid boundaries, and suffer from delayed updates. Furthermore, it achieves dynamic boundary identification based on real user behavior data, automatically adapting to spatial changes in the charging station's service area. While ensuring boundary accuracy and physical alignment, it significantly reduces labor costs and update delays, providing an efficient, accurate, and traceable boundary identification solution for charging station operation management, expansion planning, and compliance supervision.

[0043] As an optional embodiment of the present invention, based on the above embodiments, this optional embodiment can optimize the method before constructing the multi-dimensional dwelling feature data of each dwelling point within the effective identification range of the charging station, and further includes: a1) Determine the initial buffer zone of the charging pile according to the initial search range corresponding to the type of charging pile in the charging station.

[0044] The initial search range can be understood as the initial search radius. Specifically, the initial search range is set according to the type of charging pile in the charging station, determining the search area corresponding to the charging pile, which is denoted as the initial buffer zone. For example, the initial search radius for fast charging piles is set to 300m, and the initial search radius for slow charging piles is set to 150m.

[0045] b1) Overlay the initial buffer zone with the road buffer zone and remove invalid areas to determine the effective identification range of the charging station.

[0046] In this embodiment, a road buffer zone is superimposed on the initial buffer zone to eliminate invalid areas isolated by roads, thus obtaining the effective identification range of the charging station. The parameters involved in this step include the charging station identifier, charging station type, initial radius, road buffer zone, and effective area range.

[0047] The above technical solution adds a step to determine the effective identification range of the charging station, providing basic data for subsequent identification of the physical boundary of the charging station.

[0048] Example 2 Figure 2 This is a flowchart illustrating another dynamic identification method for the physical boundary of a charging station provided in Embodiment 2 of the present invention. This embodiment is a further optimization of the above embodiment. In this embodiment, the following optimizations are made: "constructing multi-dimensional dwelling feature data of each dwelling point within the effective identification range of the charging station, and performing clustering processing on the multi-dimensional dwelling feature data to obtain the dwelling behavior cluster of the charging station"; "extracting contours from the dwelling behavior cluster to generate the initial physical boundary of the charging station"; "correcting the initial physical boundary based on road network information within a preset range of the initial physical boundary to obtain a first corrected boundary of the charging station after road adaptation"; "correcting the first corrected boundary based on building information within a preset range of the first corrected boundary to obtain a second corrected boundary of the charging station after conforming to the building layout"; "performing topology optimization processing on the second corrected boundary to obtain a final physical boundary of the charging station with complete topology and simple structure"; and "performing multi-dimensional compliance verification on the final physical boundary to obtain the verification result of the final physical boundary of the charging station".

[0049] like Figure 2 As shown in the figure, this embodiment 2 provides a method for dynamic identification of the physical boundary of a charging station, which specifically includes the following steps: S201. Obtain information on each charging pile within the effective identification range of the charging station and dwelling behavior data of each dwelling point, and perform association processing on the charging pile information and the dwelling behavior data to obtain multi-dimensional dwelling feature data of each dwelling point.

[0050] The charging pile information may include charging pile identification and location. Dwelling behavior data may include time, charging pile identification, location, user identification, dwell time, dwell frequency, and other auxiliary fields. In this embodiment, using known charging pile information and detailed user dwell day data, the user dwell behavior within a specific charging station area is statistically analyzed and studied to understand the dwell characteristics, behavioral habits, and usage patterns of users in that area. During the data preparation and feature data construction phase, the data used is not only the original list of dwell point latitude and longitude coordinates, but also associates each dwell point with known charging pile locations. In addition to spatial coordinates, behavioral characteristics such as dwell time and dwell frequency are also considered in cluster analysis, and the associated data is recorded as multi-dimensional dwell feature data for each dwell point.

[0051] S202. Based on the dwell time and dwell frequency in the multidimensional dwell feature data of each dwell point, select effective dwell points from each dwell point.

[0052] In this embodiment, valid dwell points are selected based on their dwell time exceeding a set duration and dwell frequency exceeding a set frequency, while noise points within the road buffer zone are excluded. Valid dwell points are then selected from among these, and subsequent clustering is performed based on the multidimensional dwell feature data of these valid dwell points. The set duration and set frequency can be set according to actual conditions. For example, dwell points with a stay of >30 minutes and a weekly frequency of ≥3 times are considered valid dwell points, while noise points within the road buffer zone are excluded. The parameters involved in this step mainly include: user identifier, dwell point location, dwell time, dwell frequency, and validity identifier.

[0053] S203. Perform clustering processing on the multidimensional dwelling feature data of the effective dwelling points to obtain the dwelling behavior cluster of the charging station.

[0054] In this embodiment, the dwell time of the effective dwell points is used as the weight to perform weighted processing on each effective dwell point to obtain each cluster center. Then, adjacent clusters with small distance between cluster centers are merged to obtain each dwell behavior cluster, which is used as the dwell behavior cluster of the charging station.

[0055] As a specific implementation, the step of clustering the multidimensional dwelling feature data of the effective dwelling points to obtain the dwelling behavior clusters of the charging station can be optimized, including: a2) Using the residence time of the effective residence points in the multidimensional residence feature data of the effective residence points as weights, perform weighted processing on each of the effective residence points to obtain each cluster center.

[0056] In this embodiment, the residence time of the effective residence points in the multidimensional residence feature data of the effective residence points is used as the weight to perform weighted processing on each effective residence point, and the centroid of each cluster is obtained as the cluster center.

[0057] b2) Based on each cluster center, generate multiple initial behavior clusters from the multidimensional residency feature data of the effective residency points.

[0058] Specifically, after determining the cluster centers, the density-based clustering algorithm scans all the dwell points in the multidimensional dwell feature data of the effective dwell points using two core parameters (i.e., the neighborhood radius and the minimum number of points required for the core points) to perform dwell point clustering, generating multiple behavioral clusters, denoted as the initial behavioral clusters.

[0059] c2) Merge the initial behavior clusters whose spacing between the cluster centers is less than a set distance threshold to obtain the dwell behavior clusters of the charging station.

[0060] The dwell behavior clusters include core charging behavior clusters and associated auxiliary behavior clusters. Specifically, initial behavior clusters where the distance between cluster centers is less than a set distance threshold are merged, and the merged behavior clusters are divided into core charging behavior clusters and associated auxiliary behavior clusters. For example, the set distance threshold can be 100m. Core charging behavior clusters (high density, long duration, high frequency): These points are closely clustered around the charging piles (high density), and the dwell time at these points is generally long (matching the time required for charging), and the frequency of the same user's appearance is also high (possibly a regular customer or an operational vehicle). This high spatial density clustering and long, high-frequency dwell characteristics strongly suggest that the behavioral purpose is "charging". This is the most important behavioral pattern to be identified, directly pointing to the core service area of ​​the charging station. Associated auxiliary behavior clusters (medium density, medium to short duration): These points are adjacent to the core clusters but at a certain distance, with relatively low density. The dwell time is medium or short, and these points may correspond to other activities that users perform during charging, such as going to the nearby restroom, toilet, convenience store, etc. Identifying these clusters helps to more completely define the "physical boundary" of the charging station, as it encompasses the range of user activity resulting from charging. Noise / abnormal behavior points (low density, isolated, short duration): These points are isolated and do not meet the density requirements of any cluster. Their dwell time is usually extremely short, or their location is significantly off-center from the charging station area (e.g., in the center of a main road). These points are classified as "atypical behavior" or "noise," and may represent signal drift, vehicles briefly passing by, waiting at red lights, or users' social activities in unrelated areas around the charging station. Filtering out these points using the DBSCAN algorithm can greatly purify the dataset and prevent these "false signals" from interfering with subsequent boundary generalization calculations. The DBSCAN algorithm is a density-based clustering algorithm that can discover clusters of arbitrary shapes and automatically identify noise points without pre-setting the number of clusters. The main parameters involved in this step include cluster ID, core charging station points, dwell point set, and heat index.

[0061] The above technical solution specifies the steps for obtaining the dwell behavior clusters of charging stations through clustering.

[0062] S204. Determine the cluster center of the dwell behavior cluster and the virtual boundary point inserted into the edge of the dwell behavior cluster, and add the cluster center and the virtual boundary point to the dwell behavior cluster to obtain the enhanced dwell behavior cluster.

[0063] The virtual boundary points are determined based on the spatial distribution density of the dwelling behavior cluster.

[0064] The aforementioned obtained clusters of dwelling behaviors are equivalent to a set of points. This step is used to enhance these point sets, i.e., to obtain enhanced clusters of dwelling behaviors. Specifically, the geometric center (centroid) of each dwelling behavior cluster is calculated and denoted as the cluster center. This center is then added as a new, forced point to the dwelling behavior cluster. The cluster center represents the "centroid" of the entire cluster. Even if there is a small amount of noise or fluctuation at the edge points, introducing the cluster center acts like an "anchor," ensuring that the generated polygonal core region does not shift excessively. This enhances the stability of the boundary and prevents important core regions from being missed due to uneven distribution of discrete points. The cluster center serves as a key anchor point.

[0065] Then, virtual boundary points are inserted to prevent excessive concavity. At the edges of clusters of resident behaviors, virtual points are intelligently inserted as a "catch-all" based on the distribution density of the clusters. In real data, the points at the very edge may be noise points or extremely isolated behaviors. Generating concave boundaries directly based on these points could lead to unreasonable, extreme "concavities" or "jagged edges." Inserting virtual boundary points is equivalent to setting a smoothing constraint, preventing the algorithm from "overfitting" to noise points, making the boundaries more natural and reasonable while preserving detail. The parameters involved in this step mainly include: augmented point set, anchor point position, virtual boundary points, and point set density.

[0066] S205. The enhanced dwell behavior cluster is processed using a triangulation algorithm to construct multiple triangular meshes, and the contour extraction algorithm is used to extract the contours of each triangular mesh to generate polygons.

[0067] In this embodiment, the Delaunay triangulation algorithm can be used, and the Alpha Shape algorithm can be used for contour extraction. Specifically, the enhanced dwell behavior clusters are triangulated using the triangulation algorithm to form a triangular mesh. Then, the contour extraction algorithm is used to extract the contours of the triangular mesh to generate concave polygons, denoted as polygons.

[0068] As a specific implementation method, the step of extracting the contours of each triangular mesh using a contour extraction algorithm to generate polygons can be optimized, including: a3) Traverse each of the triangulations and determine the circumcircle radius of each edge in the triangulation.

[0069] In this embodiment, each triangulation is traversed to determine the radius of the outer circle of each edge in the triangulation.

[0070] b3) Compare the radius of the circumcircle of the edge with a preset radius threshold.

[0071] In this embodiment, the preset radius threshold can be set according to actual conditions. Specifically, the denser the enhanced dwell behavior cluster, the smaller the preset radius threshold. The preset radius threshold can be dynamically adjusted, using different thresholds based on the density level of the point set (high-density area / low-density area). For example, in high-density areas, the preset radius threshold α = 0.5 (fine boundary): dense point sets indicate highly concentrated and predictable user behavior (e.g., core parking areas). Using a smaller α value generates a "fine boundary," depicting more complex indentation details with higher accuracy. In low-density areas, the preset radius threshold α = 0.8 (smooth boundary): sparse point sets indicate dispersed or uncertain user behavior (e.g., site edges). Using a larger α value generates a "smooth boundary," avoiding excessive jagged edges and trivial indentations in areas with unreliable data, thus enhancing the robustness of the boundary.

[0072] This step compares the circumcircle radius of the edges with a preset radius threshold. By judging the relationship between the circumcircle radius of the edges in the triangular mesh and the preset radius threshold, edges are filtered or removed, ultimately forming polygons.

[0073] c3) If the radius of the circumcircle of the edge is less than or equal to the preset radius threshold, then the edge is retained.

[0074] Specifically, if the radius of the circumcircle of an edge is less than or equal to a preset radius threshold, the edge will be retained.

[0075] d3) If the radius of the circumcircle of the edge is greater than the preset radius threshold, then the edge is removed.

[0076] Specifically, if the radius of the circumcircle of an edge is greater than a preset radius threshold, the edge will be removed.

[0077] e3) Generate the polygon based on the retained edges.

[0078] This step generates polygons based on the retained edges. The main parameters involved in this step include: preset radius threshold, boundary polygons, density level, and boundary precision.

[0079] The above technical solution specifies the steps of using a contour extraction algorithm to extract the contours of each triangular mesh and generate polygons.

[0080] S206. Determine the convex hull region of each charging pile, and perform intersection processing between the polygon and the convex hull region of the charging pile to generate the initial physical boundary of the charging station.

[0081] This step is used to force the charging piles to be included within the physical boundary. Specifically, the convex hull region of each charging pile is determined, and the intersection of the polygon and the convex hull region is taken to ensure that all charging piles are within the boundary, thus obtaining the initial boundary polygon, which is denoted as the initial physical boundary of the charging station. The parameters involved in this step mainly include: the convex hull of the charging piles, the intersection boundary, and the inclusion status of the charging piles.

[0082] S207. Obtain road network information within the preset range of the initial physical boundary, and classify each road according to the road network information to obtain the road classification.

[0083] The road classification includes hard-barrier main roads, soft-barrier secondary roads, and merging branch roads. The preset range can be set according to actual conditions. This step is used for road classification processing. First, road network information within the preset range of the initial physical boundary is obtained. Then, the roads are classified according to their grade and function in the road network information to obtain each road classification. The parameters involved in this step mainly include: road grade, road type, road width, and traffic flow.

[0084] In this embodiment, roads are divided into three main levels: hard-barrier main roads, soft-barrier secondary roads, and merging secondary roads. Hard-barrier main roads include types such as highways, urban expressways, main traffic arteries, and major roads with central medians. Soft-barrier secondary roads include types such as urban secondary roads, regional traffic arteries, and roads without central medians but with high traffic volume. Merging secondary roads include types such as urban secondary roads, internal roads within residential areas, and auxiliary roads primarily consisting of non-motorized vehicle lanes.

[0085] S208. For the hard barrier main road, cut off the hard barrier main road from the initial physical boundary.

[0086] In this embodiment, for main roads with physical barriers, the processing method is cutting, including types such as highways, urban expressways, main traffic arteries, and major roads with central medians. These roads have high vehicle speeds and large traffic volumes, and are usually physically separated. Users cannot cross highways to reach charging stations, and the physical service area of ​​charging stations can never extend to the other side of the road. Therefore, they constitute natural and absolute spatial boundaries. Extending the boundary of charging stations to the other side of the road is meaningless and would introduce a large number of irrelevant areas, seriously impairing the accuracy of the boundary. Therefore, they must be removed from the initial physical boundary. Specifically, for main roads with physical barriers, the road buffer zone of the main road with physical barriers is removed from the initial physical boundary.

[0087] S209. For the soft-barrier secondary trunk road, the initial physical boundary is offset by a set distance in parallel towards the charging station to construct a safe buffer boundary.

[0088] In this embodiment, the set distance can be set according to the actual situation, for example, the set distance can be 3~5m. For soft-barrier secondary arterial roads, the processing method is offset, including types such as: urban secondary arterial roads, regional traffic arterial roads, and roads without a central median but with high traffic volume. In order to simulate real access channels and safety buffer zones, the boundary is not directly cut off, but is offset parallel to the inward direction of the station by a set distance. This acknowledges the barrier effect of the road while maintaining the integrity and practicality of the charging station area. Specifically, for soft-barrier secondary arterial roads, the initial physical boundary is offset parallel to the road by a set distance to construct a safety buffer boundary. The parameters involved in this step mainly include: cutting operation, offset distance, and boundary after cutting.

[0089] S210. For the mergeable branch, retain the areas on both sides of the road of the mergeable branch, and smooth the initial physical boundary that crosses the mergeable branch.

[0090] In this embodiment, for mergeable branch roads, the processing method is smoothing, including types such as urban branch roads, internal roads of residential areas, and auxiliary roads mainly consisting of non-motorized vehicle lanes. These roads have low traffic volume and slow speeds, and may themselves be the entrance roads of charging stations or their extensions. They do not constitute spatial barriers, and the service range of the charging station may fully cover both sides of the road. If the branch road is cut or offset, it will disrupt a charging area that should be continuous, resulting in fragmented boundaries. Therefore, the areas on both sides of the branch road are preserved, and only the boundaries crossing the branch road are smoothed to ensure the integrity of the entire service area. Specifically, for mergeable branch roads, the areas on both sides of the branch road are preserved, and only smoothing is performed. The parameters involved in this step mainly include: branch road processing, smoothing parameters, and final boundaries.

[0091] S211. Obtain the corrected initial physical boundary as the first corrected boundary of the charging station after road adaptation.

[0092] By correcting the initial physical boundary through the above steps S207~S210, the boundary after road adaptation can be obtained, which is denoted as the first corrected boundary of the charging station.

[0093] S212. Obtain building information within the preset range of the first correction boundary, and input the building information into the pre-trained classification model to obtain building classification.

[0094] The building classification includes charging-related buildings and obstacle buildings. For example, charging-related buildings include charging distribution rooms / rest rooms. The preset range can be set according to actual conditions; for example, the preset range can be 50m. The pre-trained classification model can be understood as a trained neural network model used to classify buildings. For example, the classification model can use a gradient boosting machine (LightGBM) decision tree.

[0095] This step is used for building association analysis. Specifically, it filters relevant buildings within a preset range of the first correction boundary, obtains their building information, and then inputs this information into a pre-trained classification model to classify the buildings as charging-related buildings or obstacle buildings. For example, it filters relevant buildings less than 50m from the first correction boundary and classifies them as charging-related buildings or obstacle buildings using the pre-trained classification model. The parameters involved in this step include: building location, building type, association distance, and classification label.

[0096] S213. For the charging-related buildings, the outline of the charging-related buildings is merged with the first modified boundary.

[0097] This step is used for differentiated adsorption. For charging-related buildings, they are directly incorporated into the first modified boundary, preserving the original outline, that is, the outline of the charging-related buildings is merged with the first modified boundary.

[0098] S214. For the obstacle structure, the boundary associated with the obstacle structure in the first correction boundary is shifted to a predetermined position outside the edge of the obstacle structure.

[0099] In this embodiment, the "set position" can be understood as setting a safety distance, for example, a safety distance of 3m. The boundary associated with the obstacle building in the first correction boundary can be understood as the nearest edge of the first correction boundary to the obstacle building. Specifically, for the obstacle building, the nearest edge of the first correction boundary to the obstacle building is calculated and offset along the normal direction to the edge of the obstacle building plus the set safety distance; that is, the boundary associated with the obstacle building in the first correction boundary is offset to a set position outside the edge of the obstacle building. The parameters involved in this step mainly include: adsorption type, offset distance, safety distance, and the boundary after adsorption.

[0100] S215. Based on the contour features of each building, perform contour optimization processing on the first modified boundary.

[0101] This step involves optimizing the first modified boundary based on the contour features of each building, thereby obtaining the physical boundary after the buildings are fitted. For example, contour optimization can preserve the right-angle features of the buildings and avoid rounded deformation, thus obtaining the physical boundary after the buildings are fitted. The parameters involved in this step mainly include: right-angle features, contour optimization, and final boundary.

[0102] S216. The modified first boundary is used as the second modified boundary of the charging station after conforming to the building layout.

[0103] By correcting the first correction boundary through the above steps S212~S215, a boundary that fits the building layout can be obtained, which is denoted as the second correction boundary of the charging station.

[0104] S217. Perform geometric defect repair processing, redundant vertex elimination processing, and fragment regularization processing on the second modified boundary to obtain the final physical boundary of the charging station with complete topology and simple structure.

[0105] In this embodiment, geometric defects are repaired on the second correction boundary, for example, by automatically repairing self-intersecting polygons (buffer 0 operation), where the filled area is smaller than a set area (e.g., 50 m²). 2 The parameters involved in this step, which are internal holes, mainly include: self-intersecting repair, hole filling, and geometric state.

[0106] Further redundant vertex elimination is performed on the second corrected boundary using a boundary simplification algorithm (such as the Douglas-Peucker algorithm). First, the start and end points of a curve are retained. Then, the point on the original curve furthest from this straight line is found. If the offset of this point is greater than a set threshold (e.g., retaining 5m accuracy), the point is retained as a critical node. The algorithm recursively executes this process until the offsets of all points are less than the threshold. Parameters involved in this step include: simplification algorithm, accuracy parameter, and simplified boundary.

[0107] Further fragment processing: The second modified boundary is processed through geometric fusion or boundary bridging, and the merged area is smaller than the preset area (e.g., 100 m²). 2 The fragmented polygons are processed into the main region, and the narrow peninsula with an aspect ratio greater than the preset ratio (such as 5:1) is trimmed to obtain a topologically clean polygon. The parameters involved in this step mainly include: fragment merging, peninsula trimming, and final boundary.

[0108] By correcting the second modified boundary based on the above steps, the final physical boundary of the charging station with a complete topology and simple structure can be obtained.

[0109] S218. Determine the area contained within the final physical boundary. If the area exceeds a preset area threshold, then determine that the final physical boundary is invalid.

[0110] This step is used to verify the area range. Using the area attribute of a geometry library (such as Shapely), it checks whether the area contained within the final physical boundary is within a preset area threshold range. If it exceeds this range, the final physical boundary is deemed invalid, discarded, and an alarm is triggered. For example, the preset area threshold can be set to 100-20,000 m². 2 The parameters involved in this step include: area value, range limit, verification result, and alarm information.

[0111] S219. Count the number of charging piles contained within the final physical boundary. If the number of charging piles is less than a set threshold, expand the initial search range and return to re-execute the step of determining the effective identification range of the charging station.

[0112] This step verifies the number of charging stations. Using a spatial database or geometric library's spatial inclusion method, it checks if the number of charging stations contained within the final physical boundary is greater than or equal to a set threshold. If insufficient, the initial search range is expanded, and the step of determining the effective identification range of charging stations is re-executed to regenerate the boundary. For example, the set threshold can be set to 3. The main parameters involved in this step include: the number of charging stations, the set threshold, and the processing method.

[0113] S220. Determine the residence heat density of the charging station based on the total residence time and the area contained within the final physical boundary. If the residence heat density is less than a preset heat density threshold, mark the charging station as a low-activity charging station.

[0114] This step is used to verify the thermal density. The thermal density is calculated as total dwell time divided by the area contained within the final physical boundary. It checks whether the thermal density is greater than or equal to a preset thermal density threshold. If it is insufficient, the site is marked as low-activity. For example, the preset thermal density threshold could be 0.5 minutes / m². 2 The parameters involved in this step include: thermal density, thermal density threshold, and activity level indicator.

[0115] S221. Detect the spatial connectivity of multiple components within the final physical boundary, and merge boundary components whose spatial connectivity is independent and whose spacing is less than a preset spacing threshold.

[0116] This step performs connectivity verification, checking the connectivity of multiple components within the final physical boundary. It calculates the distance between components using a distance metric from the geometry library. If the spatial connectivity is independent and the component spacing is less than a preset spacing threshold, components are automatically bridged or forcibly merged. For example, the preset spacing threshold can be 30m. The main parameters involved in this step include: connectivity status, component spacing, and processing method.

[0117] Understandably, the above solution first aggregates real user vehicle parking locations with known charging station locations. Using clustering algorithms, it distinguishes between dense, long-term parking clusters (core charging behavior) and short-term, noisy locations (such as temporary roadside parking), accurately delineating "hot zones" of user activity. An initial concave boundary is generated using a contour extraction algorithm, naturally encompassing all parking spaces and even reconstructing the internal turning area recess, avoiding the problem of convex hulls including invalid areas. Subsequently, road data from OpenStreetMap (OSM) is used for intelligent segmentation: main municipal roads crossing the station are removed from the boundary to ensure the boundary doesn't extend to the opposite side of the road; simultaneously, secondary roads at the station entrance are offset by a set distance to simulate a safe buffer zone. Furthermore, a classification model is used to incorporate charging-related buildings into the boundary, and finally, quality verification and loop closure are performed.

[0118] As described above, in the data preprocessing stage, clustering algorithms are used to assist in filtering noise at residence points and identifying outliers; spatial data processing technology is used to achieve boundary generalization / road cutting; and a geometric object operation library is used to achieve core calculations such as buffer generation, polygon cutting, and topology repair; when selecting building adsorption strategies, LightGBM classification is used to identify building functions (charging-related / non-related).

[0119] The aforementioned technical solution integrates multi-source spatiotemporal data, including multi-dimensional information such as user dwell behavior characteristics, charging pile spatial distribution, road network topology, and building outlines. It aggregates effective user behavior data through a dwell point clustering algorithm driven by charging piles, employs a contour extraction algorithm for concave boundary generalization, and combines road network segmentation and building outline snapping to achieve physical element alignment. Furthermore, it utilizes topology optimization and business rule verification to ensure the geometric validity of the boundaries and business compliance. On the other hand, compared to the limitations of existing fixed geometric boundary methods (lacking spatial heterogeneity), manual drawing methods (lacking dynamism), and GNSS mapping methods (high cost and inability to automatically update), this technical solution achieves dynamic boundary recognition based on real user behavior data. It can automatically adapt to spatial changes in the service area of ​​charging stations, significantly reducing labor costs and update delays while ensuring boundary accuracy and physical alignment. This provides an efficient, accurate, and traceable boundary recognition solution for charging station operation management, expansion planning, and compliance supervision.

[0120] To verify the effectiveness of this technical solution, the dynamic identification of the physical boundary of charging stations was evaluated based on both the traditional method and this technical solution. The evaluation dimensions were: 1) Boundary integrity: 82% for the traditional method and 100% for this technical solution, representing an 18% improvement over the traditional method; 2) Charging station coverage: the traditional method only covers the area within 50m of the charging station, while this technical solution covers 95%+ of the actual serviceable area, representing a 45% improvement over the traditional method; 3) Computational efficiency: the traditional method requires manual inspection every 2 weeks per station, while this technical solution generates automatic data in less than 2 hours, representing a 98% improvement over the traditional method.

[0121] The aforementioned technical solution specifies the steps for determining multi-dimensional dwelling feature data, clustering processing, generating initial physical boundaries, and adapting to road conditions, building layouts, and compliance. Based on the dynamic identification of the charging station's physical boundaries using dwelling features and charging piles, it achieves automated and accurate definition of a charging station's true physical service area. By integrating multi-source information such as user dwelling behavior data, charging pile spatial distribution, road network topology, and building outlines, it employs a pile-driven dwelling point clustering algorithm to aggregate effective user behavior features, uses a triangulation algorithm for concave boundary generalization, combines road network cutting and building outline snapping to achieve physical element fitting, and utilizes topology optimization and business rule verification to ensure the geometric validity and business compliance of the boundaries. This solution can automatically identify the actual physical boundaries of charging stations and dynamically update them, effectively solving the technical challenges of existing fixed geometric boundary methods lacking spatial heterogeneity, manual drawing methods lacking dynamism, and GNSS mapping methods being costly and unable to update automatically.

[0122] Example 3 Figure 3 This is a schematic diagram of a dynamic identification device for the physical boundary of a charging station according to Embodiment 3 of the present invention. This device is applicable to situations requiring dynamic identification of the physical boundary of a charging station. The dynamic identification device can be implemented in hardware and / or software and is generally integrated into an electronic device. Figure 3 As shown, the device includes: a clustering module 31, an initial boundary determination module 32, a first boundary correction module 33, a second boundary correction module 34, a third boundary correction module 35, and a verification module 36, wherein... Clustering module 31 is used to construct multi-dimensional dwell feature data of each dwelling point within the effective identification range of the charging station, and to perform clustering processing on the multi-dimensional dwell feature data to obtain the dwell behavior cluster of the charging station. The initial boundary determination module 32 is used to extract the contours of the dwelling behavior cluster and generate the initial physical boundary of the charging station. The first boundary correction module 33 is used to correct the initial physical boundary based on the road network information within the preset range of the initial physical boundary, so as to obtain the first corrected boundary of the charging station after road adaptation. The second boundary correction module 34 is used to correct the first correction boundary according to the building information within the preset range of the first correction boundary, so as to obtain the second correction boundary of the charging station after conforming to the building layout. The third boundary correction module 35 is used to perform topology optimization processing on the second corrected boundary to obtain the final physical boundary of the charging station with complete topology and simple structure. The verification module 36 is used to perform multi-dimensional compliance verification on the final physical boundary and obtain the verification result of the final physical boundary of the charging station.

[0123] The aforementioned technical solution dynamically identifies the physical boundaries of charging stations based on dwell characteristics and charging pile locations. By fusing multi-source spatiotemporal data, including user dwell behavior characteristics, charging pile spatial distribution, road network topology, and building outlines, it aggregates effective user behavior data through a dwell point clustering algorithm driven by charging piles. A contour extraction algorithm is used for concave boundary generalization, and road network segmentation and building outline adsorption are combined to achieve physical element alignment. Topology optimization and business rule verification are employed to ensure the geometric validity and business compliance of the boundaries. This forms a complete process capability encompassing data perception, spatial computation, intelligent decision-making, and closed-loop verification, enabling dynamic identification of charging station physical boundaries. It supports meter-level practical accuracy in charging station boundary identification, improving accuracy and overcoming the limitations of traditional solutions that rely on human experience, have rigid boundaries, and suffer from delayed updates. Furthermore, it achieves dynamic boundary identification based on real user behavior data, automatically adapting to spatial changes in the charging station's service area. While ensuring boundary accuracy and physical alignment, it significantly reduces labor costs and update delays, providing an efficient, accurate, and traceable boundary identification solution for charging station operation management, expansion planning, and compliance supervision.

[0124] Optionally, clustering module 31 includes: The feature construction unit is used to acquire information on each charging pile within the effective identification range of the charging station and dwelling behavior data of each dwelling point, and to associate the charging pile information with the dwelling behavior data to obtain multi-dimensional dwelling feature data of each dwelling point. An effective filtering unit is used to filter out effective stations from each of the stations based on the station duration and station frequency in the multidimensional station characteristic data of each station. The clustering processing unit is used to perform clustering processing on the multidimensional dwelling feature data of the effective dwelling points to obtain the dwelling behavior clusters of the charging station.

[0125] Optionally, the clustering processing unit is specifically used for: The residence time of the effective residence points in the multidimensional residence feature data of the effective residence points is used as the weight to perform weighted processing on each of the effective residence points to obtain each cluster center. Based on each cluster center, the multidimensional residency feature data of the effective residency points are used to generate multiple initial behavior clusters; The initial behavior clusters whose spacing between the cluster centers is less than a set distance threshold are merged to obtain the dwell behavior cluster of the charging station. The dwell behavior cluster includes the core charging behavior cluster and the associated auxiliary behavior cluster.

[0126] Optionally, the device includes a range identification module, which, before the multidimensional dwelling feature data of each dwelling point within the effective identification range of the constructed charging station, is used to: The initial buffer zone of the charging pile is determined based on the initial search range corresponding to the type of charging pile in the charging station; The initial buffer zone is superimposed with the road buffer zone, and invalid areas are removed to determine the effective identification range of the charging station.

[0127] Optionally, the initial boundary determination module 32 includes: An enhancement processing unit is used to determine the cluster center of the dwell behavior cluster and the virtual boundary point inserted into the edge of the dwell behavior cluster, and add the cluster center and the virtual boundary point to the dwell behavior cluster to obtain an enhanced dwell behavior cluster. The virtual boundary point is determined according to the spatial distribution density of the dwell behavior cluster. The subdivision processing unit is used to process the enhanced dwell behavior clusters using a triangulation algorithm to construct multiple triangular meshes, and to extract the contours of each triangular mesh using a contour extraction algorithm to generate polygons. An intersection processing unit is used to determine the convex hull region of each of the charging piles, and to perform intersection processing between the polygon and the convex hull region of the charging pile to generate the initial physical boundary of the charging station.

[0128] Optionally, the partitioning processing unit is specifically used for: Traverse each of the aforementioned triangulations to determine the circumcircle radius of each edge in the triangulation; Compare the radius of the circumcircle of the edge with a preset radius threshold. If the radius of the circumcircle of the edge is less than or equal to the preset radius threshold, then the edge is retained; If the radius of the circumcircle of the edge is greater than the preset radius threshold, then the edge is removed. The polygon is generated based on each of the retained edges; The denser the enhanced dwell behavior cluster, the smaller the value of the preset radius threshold.

[0129] Optionally, the first boundary correction module 33 is specifically used for: Obtain road network information within the preset range of the initial physical boundary, and classify each road according to the road network information to obtain each road classification. The road classification includes hard barrier main roads, soft barrier secondary arterial roads, and merging branch roads. For the hard-barrier main road, the hard-barrier main road is cut off from the initial physical boundary; For the soft-barrier secondary trunk road, the initial physical boundary is offset by a set distance in parallel towards the charging station to construct a safe buffer boundary; For the mergeable branch, the areas on both sides of the road of the mergeable branch are preserved, and the initial physical boundary of the branch crossing the mergeable branch is smoothed. The corrected initial physical boundary is obtained as the first corrected boundary of the charging station after road adaptation.

[0130] Optionally, the second boundary correction module 34 is specifically used for: Obtain building information within a preset range of the first correction boundary, and input the building information into a pre-trained classification model to obtain building classifications, the building classifications including charging-related buildings and obstacle buildings; For the charging-related buildings, the outline of the charging-related buildings is merged with the first corrected boundary; For the obstacle structure, the boundary associated with the obstacle structure in the first correction boundary is shifted to a predetermined position outside the edge of the obstacle structure; Based on the outline features of each building, the first modified boundary is subjected to outline optimization processing; The revised first boundary is used as the second revised boundary of the charging station after conforming to the building layout.

[0131] Optionally, the third boundary correction module 35 is specifically used for: The second modified boundary is subjected to geometric defect repair processing, redundant vertex elimination processing, and fragment regularization processing to obtain the final physical boundary of the charging station with complete topology and simple structure.

[0132] Optionally, the third boundary correction module 35 is specifically used for: Determine the area contained within the final physical boundary; if the area exceeds a preset area threshold, then the final physical boundary is determined to be invalid. The number of charging piles contained within the final physical boundary is counted. If the number of charging piles is less than a set threshold, the initial search range is expanded and the step of determining the effective identification range of the charging station is re-executed. The residence heat density of the charging station is determined based on the total residence time and the area contained within the final physical boundary. If the residence heat density is less than a preset heat density threshold, the charging station is marked as a low-activity charging station. The spatial connectivity of multiple components within the final physical boundary is detected, and boundary components whose spatial connectivity is independent and whose spacing is less than a preset spacing threshold are merged.

[0133] The charging station physical boundary dynamic identification device provided in this embodiment of the invention can execute the charging station physical boundary dynamic identification method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0134] Example 4 Figure 4 This is a schematic diagram of an electronic device provided in Embodiment 4 of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0135] like Figure 4 As shown, the electronic device 40 includes at least one processor 41 and a memory, such as a read-only memory (ROM) 42 or a random access memory (RAM) 43, communicatively connected to the at least one processor 41. The memory stores computer programs executable by the at least one processor. The processor 41 can perform various appropriate actions and processes based on the computer program stored in the ROM 42 or loaded from storage unit 48 into the RAM 43. The RAM 43 may also store various programs and data required for the operation of the electronic device 40. The processor 41, ROM 42, and RAM 43 are interconnected via a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0136] Multiple components in electronic device 40 are connected to I / O interface 45, including: input unit 46, such as keyboard, mouse, etc.; output unit 47, such as various types of monitors, speakers, etc.; storage unit 48, such as disk, optical disk, etc.; and communication unit 49, such as network card, modem, wireless transceiver, etc. Communication unit 49 allows electronic device 40 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0137] Processor 41 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 41 performs the various methods and processes described above, such as the dynamic identification method for the physical boundaries of charging stations.

[0138] In some embodiments, the charging station physical boundary dynamic identification method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 48. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 40 via ROM 42 and / or communication unit 49. When the computer program is loaded into RAM 43 and executed by processor 41, one or more steps of the charging station physical boundary dynamic identification method described above can be performed. Alternatively, in other embodiments, processor 41 can be configured to perform the charging station physical boundary dynamic identification method by any other suitable means (e.g., by means of firmware).

[0139] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0140] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0141] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0142] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0143] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0144] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0145] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the dynamic identification method for the physical boundaries of charging stations as provided in any embodiment of this invention.

[0146] In implementing a computer program product, computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, 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 computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0147] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0148] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for dynamically identifying a physical boundary of a charging station, characterized in that, include: Construct multidimensional dwelling feature data for each dwelling point within the effective identification range of the charging station, and perform clustering processing on the multidimensional dwelling feature data to obtain the dwelling behavior cluster of the charging station; Contour extraction is performed on the cluster of dwelling behaviors to generate the initial physical boundary of the charging station; Based on the road network information within the preset range of the initial physical boundary, the initial physical boundary is corrected to obtain the first corrected boundary of the charging station after road adaptation. Based on the building information within the preset range of the first correction boundary, the first correction boundary is corrected to obtain a second correction boundary of the charging station that conforms to the building layout. The second modified boundary is subjected to topology optimization processing to obtain the final physical boundary of the charging station with complete topology and simple structure; The final physical boundary is subjected to multi-dimensional compliance verification to obtain the verification results of the final physical boundary of the charging station.

2. The method of claim 1, wherein, The method involves constructing multidimensional dwelling feature data for each dwelling point within the effective identification range of the charging station, and performing clustering processing on the multidimensional dwelling feature data to obtain dwelling behavior clusters for the charging station, including: Information on each charging pile within the effective identification range of the charging station and dwell behavior data of each dwelling point are obtained, and the charging pile information and dwell behavior data are correlated to obtain multi-dimensional dwell feature data of each dwelling point. Based on the dwell time and dwell frequency in the multidimensional dwell characteristic data of each dwell point, effective dwell points are selected from each dwell point. Clustering is performed on the multidimensional dwelling feature data of the effective dwelling points to obtain the dwelling behavior clusters of the charging stations.

3. The method according to claim 2, characterized in that, The clustering process performed on the multidimensional dwelling feature data of the effective dwelling points to obtain the dwelling behavior clusters of the charging stations includes: The residence time of the effective residence points in the multidimensional residence feature data of the effective residence points is used as the weight to perform weighted processing on each of the effective residence points to obtain each cluster center. Based on each cluster center, the multidimensional residency feature data of the effective residency points are used to generate multiple initial behavior clusters; The initial behavior clusters whose spacing between the cluster centers is less than a set distance threshold are merged to obtain the dwell behavior cluster of the charging station. The dwell behavior cluster includes the core charging behavior cluster and the associated auxiliary behavior cluster.

4. The method according to claim 1, characterized in that, Before constructing the multidimensional dwelling feature data of each dwelling point within the effective identification range of the charging station, the following is also included: The initial buffer zone of the charging pile is determined based on the initial search range corresponding to the type of charging pile in the charging station; The initial buffer zone is superimposed with the road buffer zone, and invalid areas are removed to determine the effective identification range of the charging station.

5. The method according to claim 1, characterized in that, The step of extracting the contours of the dwell behavior clusters to generate the initial physical boundaries of the charging station includes: Determine the cluster center of the dwell behavior cluster and the virtual boundary point inserted into the edge of the dwell behavior cluster, and add the cluster center and the virtual boundary point to the dwell behavior cluster to obtain an enhanced dwell behavior cluster. The virtual boundary point is determined according to the spatial distribution density of the dwell behavior cluster. The enhanced dwell behavior clusters are processed using a triangulation algorithm to construct multiple triangular meshes, and contour extraction algorithms are used to extract the contours of each triangular mesh to generate polygons. The convex hull region of each charging pile is determined, and the intersection of the polygon and the convex hull region of the charging pile is performed to generate the initial physical boundary of the charging station.

6. The method according to claim 5, characterized in that, The step of extracting contours from each of the triangular meshes using a contour extraction algorithm to generate polygons includes: Traverse each of the aforementioned triangulations to determine the circumcircle radius of each edge in the triangulation; Compare the radius of the circumcircle of the edge with a preset radius threshold. If the radius of the circumcircle of the edge is less than or equal to the preset radius threshold, then the edge is retained; If the radius of the circumcircle of the edge is greater than the preset radius threshold, then the edge is removed. The polygon is generated based on each of the retained edges; The denser the enhanced dwell behavior cluster, the smaller the value of the preset radius threshold.

7. The method according to claim 1, characterized in that, The step of correcting the initial physical boundary based on road network information within a preset range of the initial physical boundary to obtain a first corrected boundary of the charging station after road adaptation includes: Obtain road network information within the preset range of the initial physical boundary, and classify each road according to the road network information to obtain each road classification. The road classification includes hard barrier main roads, soft barrier secondary arterial roads, and merging branch roads. For the hard-barrier main road, the hard-barrier main road is cut off from the initial physical boundary; For the soft-barrier secondary trunk road, the initial physical boundary is offset by a set distance in parallel towards the charging station to construct a safe buffer boundary; For the mergeable branch, the areas on both sides of the road of the mergeable branch are preserved, and the initial physical boundary of the branch crossing the mergeable branch is smoothed. The corrected initial physical boundary is obtained as the first corrected boundary of the charging station after road adaptation.

8. The method according to claim 1, characterized in that, The step of correcting the first correction boundary based on building information within a preset range of the first correction boundary to obtain a second correction boundary for the charging station that conforms to the building layout includes: Obtain building information within a preset range of the first correction boundary, and input the building information into a pre-trained classification model to obtain building classifications, the building classifications including charging-related buildings and obstacle buildings; For the charging-related buildings, the outline of the charging-related buildings is merged with the first corrected boundary; For the obstacle structure, the boundary associated with the obstacle structure in the first correction boundary is shifted to a predetermined position outside the edge of the obstacle structure; Based on the outline features of each building, the first modified boundary is subjected to outline optimization processing; The revised first boundary is used as the second revised boundary of the charging station after conforming to the building layout.

9. The method according to claim 1, characterized in that, The step of performing topology optimization on the second modified boundary to obtain the final physical boundary of the charging station with a complete topology and simple structure includes: The second modified boundary is subjected to geometric defect repair processing, redundant vertex elimination processing, and fragment regularization processing to obtain the final physical boundary of the charging station with complete topology and simple structure.

10. The method according to claim 4, characterized in that, The multi-dimensional compliance verification of the final physical boundary to obtain the verification result of the final physical boundary of the charging station includes: Determine the area contained within the final physical boundary; if the area exceeds a preset area threshold, then the final physical boundary is determined to be invalid. The number of charging piles contained within the final physical boundary is counted. If the number of charging piles is less than a set threshold, the initial search range is expanded and the step of determining the effective identification range of the charging station is re-executed. The residence heat density of the charging station is determined based on the total residence time and the area contained within the final physical boundary. If the residence heat density is less than a preset heat density threshold, the charging station is marked as a low-activity charging station. The spatial connectivity of multiple components within the final physical boundary is detected, and boundary components whose spatial connectivity is independent and whose spacing is less than a preset spacing threshold are merged.

11. A dynamic identification device for the physical boundary of a charging station, characterized in that, include: The clustering module is used to construct multi-dimensional dwell feature data of each dwelling point within the effective identification range of the charging station, and to perform clustering processing on the multi-dimensional dwell feature data to obtain the dwell behavior clusters of the charging station. An initial boundary determination module is used to extract the contours of the dwelling behavior clusters and generate the initial physical boundary of the charging station. The first boundary correction module is used to correct the initial physical boundary based on the road network information within the preset range of the initial physical boundary, so as to obtain the first corrected boundary of the charging station after road adaptation. The second boundary correction module is used to correct the first correction boundary based on the building information within the preset range of the first correction boundary, so as to obtain the second correction boundary of the charging station after conforming to the building layout. The third boundary correction module is used to perform topology optimization processing on the second corrected boundary to obtain the final physical boundary of the charging station with complete topology and simple structure. The verification module is used to perform multi-dimensional compliance verification on the final physical boundary and obtain the verification result of the final physical boundary of the charging station.

12. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the dynamic identification method for the physical boundary of a charging station as described in any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the dynamic identification method for the physical boundary of a charging station as described in any one of claims 1-10.

14. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the dynamic identification method for the physical boundary of a charging station as described in any one of claims 1-10.

Images