An electric vehicle charging station site selection planning method, system, device and medium

Abstract

The application relates to the field of electric vehicle charging station site selection planning, and discloses an electric vehicle charging station site selection planning method, system, device and medium. Trajectory coordinates are obtained by performing coordinate conversion on trajectory data of electric vehicles in a target area. Based on a vortex field optimization algorithm, parameters of a DBSCAN algorithm are optimized through a double force driving mechanism and a vortex splitting mechanism to obtain optimal parameters. The trajectory coordinates are clustered through the DBSCAN algorithm with the optimal parameters to obtain a charging demand point set and a preselected charging station site set to construct a minimum charging facility model and solve the model to obtain a final site selection planning scheme. The vortex field optimization algorithm is used to globally optimize parameters of a clustering algorithm, the subjectivity and limitations of manual trial and error are avoided, the clustering effect is more stable, and the minimum charging facility model is used to eliminate site demand overlap. On the premise of ensuring coverage of all demand points, the number of charging facilities is reduced, and equipment utilization is improved.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging station site selection and planning, and in particular to a method, system, equipment and medium for electric vehicle charging station site selection and planning. Background Technology

[0002] With the increasing popularity of electric vehicles, the layout of charging infrastructure has become a key issue restricting their development. Traditional charging station site selection methods often rely on experience-based judgment or simple spatial interpolation, lacking accurate analysis of users' actual charging needs. This easily leads to a "supply-demand mismatch"—some areas have excessively dense charging stations (high equipment idle rates), while other areas have a severe shortage of charging stations (users queue to charge), resulting in a waste of public resources and a bottleneck in the supply of new infrastructure. Currently, although some studies have attempted to use clustering algorithms (such as K-Means) to identify charging demand areas, the parameters of these algorithms are often determined through manual trial and error, making it difficult to adapt to the differences in demand distribution across different cities. Furthermore, they do not consider the overlap of demand at pre-selected sites, resulting in high construction costs and poor economic efficiency in the final site selection scheme.

[0003] Therefore, how to provide a precise, efficient and economical method for site selection and planning of electric vehicle charging stations has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] This invention provides a method, system, equipment, and medium for electric vehicle charging station site selection and planning. It solves the problems of existing clustering algorithms, which rely heavily on manual trial and error to determine parameters when planning electric vehicle site selection. These algorithms are difficult to adapt to the different demand distributions in different cities and do not consider the overlapping demand of pre-selected sites, resulting in high construction costs and poor economic efficiency in the final site selection scheme.

[0005] To address the aforementioned technical problems, the first aspect of this invention provides a method for site selection and planning of electric vehicle charging stations, comprising: The trajectory data of electric vehicles within the target area is acquired, and the trajectory data is transformed to obtain the trajectory coordinates of the electric vehicles. Based on the vortex field optimization algorithm, the parameters of the DBSCAN algorithm are optimized through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters; By applying the DBSCAN algorithm with the optimal parameters, the trajectory coordinates are clustered to obtain a set of charging demand points and a set of pre-selected charging stations. Based on the set of charging demand points and the set of pre-selected charging stations, a minimum charging facility model is constructed and solved to optimize the site selection layout of the pre-selected charging station set and obtain the final site selection planning scheme.

[0006] As one preferred embodiment, the step of acquiring trajectory data of electric vehicles within the target area and performing coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles includes: The trajectory data of electric vehicles within the target area is acquired, and the trajectory data is preprocessed to obtain the trajectory processing data of the electric vehicles. Based on the Mercator projection algorithm, the trajectory processing data is converted from latitude and longitude coordinates to plane rectangular coordinates to obtain the trajectory coordinates of the electric vehicle.

[0007] As one preferred embodiment, the parameters of the DBSCAN algorithm include the neighborhood radius and density threshold; wherein, the vortex field optimization algorithm optimizes the parameters of the DBSCAN algorithm through a dual-force driving mechanism and a vortex splitting mechanism to obtain the optimal parameters, including: Set the number of particles in the vortex field, and set the search space corresponding to the neighborhood radius and the density threshold respectively. Randomly generate an initial particle swarm based on the search space; each particle corresponds to a set of parameters of the DBSCAN algorithm. For each particle, the trajectory coordinates are clustered by applying the DBSCAN algorithm with the corresponding parameters of the particle to obtain multiple clusters. An optimization objective function is constructed based on the clusters to quantify the optimization objective value of each particle, and the particle with the smallest optimization objective value is taken as the global optimal particle. Based on the aforementioned dual-force driving mechanism, the positions of each particle are updated through the coordinated driving of centripetal force and tangential force; The positional aggregation degree of each particle after the update is quantified, and when the positional aggregation degree exceeds the preset aggregation degree requirement, the vortex splitting mechanism is triggered to introduce Levy flight to randomly generate new particles and obtain a new population. The process of determining the global optimal particle and generating the new population is performed iteratively according to the new population until the maximum preset number of iterations is reached, and the position of the global optimal particle in this iteration is used as the optimal parameter.

[0008] As one preferred embodiment, the updating of the positions of each particle based on the dual-force driving mechanism through the coordinated driving of centripetal force and tangential force includes: Obtain the current iteration number, and determine the centripetal force coefficient that controls the centripetal force intensity and the tangential force coefficient that controls the tangential force disturbance intensity based on the current iteration number; The centripetal force data of each particle is determined based on the centripetal force coefficient and the position of the globally optimal particle, and the tangential force data of each particle is determined based on the tangential force coefficient and the position of the globally optimal particle. Based on the centripetal force data and tangential force data of each particle, the positions of each particle are updated collaboratively to obtain the new positions of each particle.

[0009] As one preferred embodiment, the step of constructing the optimization objective function based on the clusters includes: The DB index of each cluster is quantified, and minimizing the DB index is used as the optimization objective function.

[0010] As one preferred embodiment, the step of clustering the trajectory coordinates using the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations includes: Based on the optimal neighborhood radius and the optimal density threshold, the core object in the trajectory coordinates is determined, and the density reachable sample in the trajectory coordinates is determined based on the core object; Based on the principle of density reachability calculation, all density reachable samples connected by the maximum density are grouped into one cluster, resulting in multiple clusters; The geometric center of each cluster is calculated as a pre-selected charging station, and the remaining samples within the cluster are used as charging demand points to obtain a set of pre-selected charging stations and a set of charging demand points.

[0011] As one preferred embodiment, the step of constructing and solving a minimum charging facility model based on the charging demand point set and the pre-selected charging station set to optimize the site selection layout of the pre-selected charging station set and obtain a final site selection planning scheme includes: Based on the set of charging demand points and the set of pre-selected charging stations, the minimum charging facility model is constructed with the objective of minimizing the number of charging stations and the constraint that each charging demand point is covered by the service range of at least one candidate charging station. The minimized charging facility model is solved using a preset optimization algorithm to optimize the site selection layout of the pre-selected charging site set and obtain the final site selection planning scheme.

[0012] A second aspect of the present invention provides an electric vehicle charging station site selection and planning system, comprising: The coordinate transformation module is used to acquire the trajectory data of electric vehicles within the target area and perform coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles. The parameter optimization module is used to optimize the parameters of the DBSCAN algorithm based on the vortex field optimization algorithm through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters. The trajectory clustering module is used to cluster the trajectory coordinates by applying the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations. The site selection planning module is used to construct and solve a minimum charging facility model based on the set of charging demand points and the set of pre-selected charging stations, so as to optimize the site selection layout of the set of pre-selected charging stations and obtain the final site selection planning scheme.

[0013] A third aspect of the present invention provides an electronic device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the electric vehicle charging station site selection planning method as described above.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium comprising a stored computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the electric vehicle charging station site selection and planning method as described above.

[0015] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows: By performing coordinate transformation on the trajectory data of electric vehicles within the target area, the measurement of electric vehicle distance becomes more accurate and convenient. A vortex field optimization algorithm is used to globally optimize the parameters of the DBSCAN algorithm, avoiding the subjectivity and limitations of manual trial and error, resulting in more stable clustering effects. Minimizing the charging facility model eliminates overlapping site demands, reducing the number of charging facilities while ensuring coverage of all demand points, significantly lowering construction costs, improving equipment utilization, and enhancing site profitability. Based on electric vehicle trajectory data and dynamic parameter optimization, it can adapt to cities of different sizes and demand distributions. Furthermore, it creatively combines algorithms from three different fields—bionic optimization (vortex field optimization algorithm), unsupervised learning (DBSCAN), and operations research optimization (set coverage model)—to form an end-to-end, complete site selection decision chain. Attached Figure Description

[0016] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, the 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.

[0017] Figure 1 This is a flowchart of a site selection and planning method for electric vehicle charging stations provided in a certain embodiment of the present invention; Figure 2 This is a flowchart of a vortex field optimization algorithm provided in a certain embodiment of the present invention; Figure 3 This is a flowchart of the DBSCAN algorithm provided in a certain embodiment of the present invention; Figure 4 This is a structural diagram of an electric vehicle charging station site selection and planning system provided in a certain embodiment of the present invention; Figure 5 This is a structural diagram of an electronic device provided in a certain embodiment of the present invention; Figure label: Among them, 10 is the coordinate transformation module; 20 is the parameter optimization module; 30 is the trajectory clustering module; 40 is the location planning module; 5000 is the electronic equipment; 5001 is the processor; 5002 is the bus; 5003 is the memory; and 5004 is the transceiver. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings and examples. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] In this invention description, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In this invention description, unless otherwise stated, "a plurality of" means two or more. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art will understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] In the description of this invention, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is merely for describing specific embodiments and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] In one embodiment, such as Figure 1 As shown, the first aspect of the present invention provides a method for site selection and planning of electric vehicle charging stations, comprising: S1. Obtain the trajectory data of electric vehicles within the target area, and perform coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles; In one embodiment, step S1 includes: The trajectory data of electric vehicles within the target area is acquired, and the trajectory data is preprocessed to obtain the trajectory processing data of the electric vehicles. Based on the Mercator projection algorithm, the trajectory processing data is converted from latitude and longitude coordinates to plane rectangular coordinates to obtain the trajectory coordinates of the electric vehicle.

[0022] Specifically, trajectory data of electric vehicles within the target area is acquired. This data is continuously collected via a BeiDou satellite positioning system terminal to ensure the accuracy of spatiotemporal information. The collected dataset includes the acquisition time (accurate to the second, used for subsequent trajectory time-series analysis), real-time vehicle latitude and longitude (accurate to six decimal places, ensuring spatial positioning accuracy), and a unique vehicle ID for each electric vehicle (used to distinguish different vehicle trajectories and avoid data confusion). The target area can be a city or a pre-defined area of ​​land, which will not be elaborated further here. Because the raw trajectory data is affected by uncontrollable factors such as signal interference and equipment failure during the acquisition process, resulting in missing, redundant, and duplicate data attributes, these noisy data will affect subsequent analysis. Therefore, the raw trajectory data needs to undergo preprocessing, including handling missing values, outlier handling, out-of-order data processing, and data filtering, to obtain the processed trajectory data of the electric vehicles.

[0023] Because the measurement relationships of precision and dimension differ in geographic coordinate systems, directly using the distance between latitude and longitude as the unit of measurement will result in deviations, necessitating coordinate transformation of latitude and longitude information. This invention employs the Mercator projection to convert the latitude and longitude coordinates in trajectory processing data into Cartesian coordinates, thereby obtaining the trajectory coordinates of the electric vehicle. Specifically, the Mercator projection maps the latitude and longitude grid information located on Earth onto a planar system, making latitude and longitude lines straight lines in the planar coordinate system. In the planar coordinate system, distance measurement is more accurate and convenient. The conversion formula is as follows: In the formula, and These are the projected planar coordinates (unit: kilometers); The radius of the Earth is taken as 6371 kilometers. Longitude (unit: degrees); Latitude (unit: degrees); Longitude converted to radians (unit: radians); The latitude is converted to radians (unit: radians).

[0024] This invention preprocesses the trajectory data of electric vehicles to make subsequent analysis results based on this data more accurate and reliable, avoiding analysis bias caused by problems with the original data. The processed data is then transformed into coordinates based on the Mercator projection algorithm to convert the trajectory data from latitude and longitude coordinates to Cartesian coordinates, which greatly simplifies the data processing and analysis process and helps to more efficiently extract information from the trajectory data of electric vehicles, providing more valuable decision-making basis for the layout of smart charging facilities.

[0025] S2. Based on the vortex field optimization algorithm, the parameters of the DBSCAN algorithm are optimized through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters; wherein, the parameters of the DBSCAN algorithm include the neighborhood radius and the density threshold. In one embodiment, step S2 includes: Set the number of particles in the vortex field, and set the search space corresponding to the neighborhood radius and the density threshold respectively. Randomly generate an initial particle swarm based on the search space; each particle corresponds to a set of parameters of the DBSCAN algorithm. For each particle, the trajectory coordinates are clustered by applying the DBSCAN algorithm with the corresponding parameters of the particle to obtain multiple clusters. An optimization objective function is constructed based on the clusters to quantify the optimization objective value of each particle, and the particle with the smallest optimization objective value is taken as the global optimal particle. Based on the aforementioned dual-force driving mechanism, the positions of each particle are updated through the coordinated driving of centripetal force and tangential force; The positional aggregation degree of each particle after the update is quantified, and when the positional aggregation degree exceeds the preset aggregation degree requirement, the vortex splitting mechanism is triggered to introduce Levy flight to randomly generate new particles and obtain a new population. The process of determining the global optimal particle and generating the new population is performed iteratively according to the new population until the maximum preset number of iterations is reached, and the position of the global optimal particle in this iteration is used as the optimal parameter.

[0026] Specifically, this invention uses DBSCAN (Density-Based Spatial Clustering) to cluster the transformed planar coordinate data to identify areas with dense charging demand. However, the effectiveness of DBSCAN clustering depends on the reasonable setting of the neighborhood radius and density threshold. Traditional manual trial-and-error methods are inefficient and prone to getting trapped in local optima. To address this, this invention proposes a vortex field optimization algorithm to optimize these two parameters. Optimization is achieved by simulating the dynamic characteristics of vortices in fluid mechanics. Vortices can maintain local rotation while also drifting with the flow field. Furthermore, the vortex field emphasizes global interaction, avoiding isolated searches. The core idea is to treat the solution space as a dynamic vortex field, where each solution is a particle in the field, driven by two forces: centripetal force and tangential force. The centripetal force attracts particles to gather in the current optimal region, enhancing the search capability; the tangential force propels particles to drift along the vortex trajectory, avoiding getting trapped in a local center. The core idea is to treat the parameter solution space as a dynamic vortex field, with each parameter combination as a particle in the field. Optimization is achieved by simulating the dual-force driving mechanism of the vortex—centripetal force (attracting particles to gather in the optimal region) and tangential force (propelling particles to avoid local optima)—as well as the vortex splitting mechanism. The implementation process is as follows: (a) Set the number of particles in the vortex field to be (Based on the solution space complexity setting, typically 30-100, 50 is preferred in this invention), maximum number of iterations. (Typically, the value is 50-200, but 100 is preferred in this invention). Next, the search spaces corresponding to the neighborhood radius ε and the density threshold MinPts are defined, including the lower bound LB of ε. ε (e.g., 0.1 km), Upper Boundary UB ε (e.g., 2 km), the lower bound of MinPts is LB MinPts (e.g. 5), Upper Boundary UB MinPts (e.g., 30), forming the lower bound vector LB=[LB ε LB MinPts ] and the upper bound vector UB=[UB ε UB MinPts ]. Subsequently, under the given lower bound vector and upper bound vector The initial population of particles is randomly generated in the search space, and this process is represented by the following equation: In the formula, These are the initial parameter values ​​for individual particles. Each particle corresponds to a set of parameters for the DBSCAN algorithm, including the neighborhood radius and density threshold. The number of particles; For dimensions; For the generated interval [0,1] A random vector is used to ensure that the initial particles are uniformly distributed in the search space.

[0027] (ii) For each particle, the trajectory coordinates are clustered by applying the DBSCAN algorithm with the corresponding parameters of the particle to obtain multiple clusters; that is, each particle is clustered in real time.

[0028] (III) Construct an optimization objective function (which can also be considered a fitness function) based on the clusters obtained from instantaneous clustering. Use this optimization objective function to quantify the optimization objective value of each particle, and take the particle with the smallest optimization objective value as the global optimal particle. In this embodiment, the parameter optimization problem of DBSCAN is to find the optimal combination of parameters that maximizes the clustering quality, which is the optimization objective. This invention uses the DB index as an indicator to evaluate the quality of the DBSCAN clustering algorithm, and uses minimizing the DB index as the optimization objective function. This process is expressed by the following formula: In the formula, For clusters and clusters The similarity between them; For clusters The average distance from all samples to the cluster center; For clusters Centroids and Clusters The distance between the centroids; DB is the optimization objective. A higher DB index indicates a worse clustering algorithm performance, while a lower DB index indicates a better clustering algorithm performance.

[0029] (iv) This invention is based on a dual-force driving mechanism, which updates the position (i.e., parameter combination) of each particle through the coordinated driving of centripetal force and tangential force; in this embodiment, in one embodiment, the updating of the position of each particle based on the dual-force driving mechanism through the coordinated driving of centripetal force and tangential force includes: Obtain the current iteration number, and determine the centripetal force coefficient that controls the centripetal force intensity and the tangential force coefficient that controls the tangential force disturbance intensity based on the current iteration number; The centripetal force data of each particle is determined based on the centripetal force coefficient and the position of the globally optimal particle, and the tangential force data of each particle is determined based on the tangential force coefficient and the position of the globally optimal particle. Based on the centripetal force data and tangential force data of each particle, the positions of each particle are updated collaboratively to obtain the new positions of each particle.

[0030] This invention determines the centripetal force coefficient and tangential force coefficient based on the current iteration number, a process expressed by the following formula: In the formula, α ( t ) is the first t Centripetal force coefficient in the next iteration; α min , α max The minimum (preferably 0.2) and maximum (preferably 0.8) values ​​of the centripetal force coefficient are given. β ( t ) is the first t Tangential force coefficients in the next iteration; β min , β max The minimum (preferably 0.1) and maximum (preferably 0.5) values ​​of the tangential force coefficient are given. t The current iteration number is represented by α(t). The centripetal force aims to pull particles towards the global optimum, and its coefficient is a key control parameter. This coefficient decreases during iteration. In the early stages of iteration, α(t) is large, and the centripetal force dominates, strongly pulling particles towards the global optimum for rapid convergence and large-scale mining. In the later stages of iteration, α(t) decreases, the centripetal force weakens, and tangential force and other mechanisms allow for more refined local searches. The tangential force aims to guide particles in exploratory motion around the global optimum, and its coefficient is a key control parameter. This coefficient increases during iteration. In the early stages of iteration, β(t) is small, the tangential force is weak, and the algorithm focuses on global search and convergence. In the later stages of iteration, β(t) increases, the tangential force strengthens, and particles are pushed to "rotate" and drift around the optimum, conducting a more thorough exploration near the optimum and avoiding overlooking potentially better neighboring solutions. Compared to Particle Swarm Optimization (PSO), which explores through randomness, the vortex field optimization algorithm of this invention guides the exploration direction through this deterministic physical mechanism, resulting in higher efficiency.

[0031] Next, the centripetal force data of each particle is determined based on the centripetal force coefficient and the position of the globally optimal particle, and the tangential force data of each particle is determined based on the tangential force coefficient and the position of the globally optimal particle. This process is expressed by the following formula: In the formula, Centripetal force data; The globally optimal particle; For the first i The particle in the first t The position of the next iteration; This is tangential force data; It is a random unit vector.

[0032] Subsequently, based on the centripetal force data and tangential force data of each particle, their positions are updated collaboratively to obtain the new positions of each particle. This process is represented by the following formula: In the formula, For the first i The particle in the first t+ The new position in the first iteration.

[0033] This invention employs a dual-force driven mechanism to update particle positions. Centripetal force attracts particles to cluster towards the "optimal region in the current iteration," enhancing the algorithm's ability to search for optimal parameter combinations and ensuring particles move towards parameters with better clustering results (smaller DB exponent). Tangential force propels particles to "drift" along a vortex trajectory. Simultaneously, randomness is introduced to prevent excessive particle aggregation in local optima, ensuring the algorithm possesses global exploration capabilities and preventing optimization from getting stuck in local optima. The method of calculating different positions based on different force coefficients provides diverse references for particle position updates, enriching the search strategy and helping particles explore more possible regions in the search space, increasing the chance of finding the global optimum. Furthermore, the collaborative update mechanism combines the effects of centripetal and tangential forces, ensuring that particles are both pulled towards the global optimum (centripetal force) and maintain a certain level of exploration capability (tangential force) when updating their positions. This balances the algorithm's convergence speed and search accuracy, effectively improving search efficiency and enabling the finding of better solutions in a shorter time.

[0034] (v) Calculate the standard deviation of the particles after position update, and characterize the position aggregation degree through the standard deviation. When the calculated position aggregation degree exceeds the preset aggregation degree requirement (i.e., less than the set threshold, such as 0.05), trigger the vortex splitting mechanism to introduce Levy flight to randomly generate new particles to update the population, thus obtaining a new population; wherein, the new particle generation process is represented by the following formula: In the formula, For new particles; The splitting coefficient is set to 0.5. For the Lévy distribution, Let's take the Levi index as 1.5.

[0035] (vi) Based on the new population, the steps for determining the global optimal particle and generating the new population are executed iteratively, that is, steps (ii) to (v) are executed iteratively until the maximum preset number of iterations is reached, and the position of the global optimal particle in this iteration is taken as the optimal parameter.

[0036] In another embodiment, before parameter optimization, the trajectory coordinates can be initially clustered using the DBSCAN algorithm with the applied initialized neighborhood radius and density threshold, and the initial DB index can be calculated based on the clustering results. That is, "manually preset initial parameters → perform DBSCAN clustering once → calculate the initial DB index → ​​start vortex field parameter optimization again", which provides a reference benchmark for the subsequent vortex field optimization algorithm to provide a reference benchmark for poor clustering results caused by improper parameters.

[0037] The flowchart of the vortex field optimization algorithm is as follows: Figure 2 As shown, this invention introduces for the first time a vortex field optimization algorithm simulating the dynamic characteristics of fluid vortices into the parameter optimization process of density clustering algorithms. This solves the industry problem of DBSCAN algorithm's parameter sensitivity and reliance on experience, achieving data-driven precision in parameter optimization and complete automation and optimization of the clustering process. Furthermore, the unique dual-force driving mechanism of the vortex field optimization algorithm can balance convergence efficiency and exploration breadth, improving optimization speed and reliability. At the same time, the vortex splitting mechanism breaks the local optimum trap, ensuring global optimum optimization. This makes its global search capability superior to traditional genetic algorithms, particle swarm optimization, etc., and can more effectively escape local optima and find better clustering parameters.

[0038] S3. Cluster the trajectory coordinates by applying the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations; In one embodiment, step S3 includes: Based on the optimal neighborhood radius and the optimal density threshold, the core object in the trajectory coordinates is determined, and the density reachable sample in the trajectory coordinates is determined based on the core object; Based on the principle of density reachability calculation, all density reachable samples connected by the maximum density are grouped into one cluster, resulting in multiple clusters; The geometric center of each cluster is calculated as a pre-selected charging station, and the remaining samples within the cluster are used as charging demand points to obtain a set of pre-selected charging stations and a set of charging demand points.

[0039] Specifically, this invention uses the transformed planar trajectory coordinates as the feature set of the data to be clustered, ensuring that each data sample corresponds to a vehicle trajectory point. Then, based on the optimal neighborhood radius (representing the search range of a sample point) and optimal density threshold (representing the minimum number of samples within the neighborhood) in the optimal parameters, the core objects in the trajectory coordinates are determined. That is, for a data sample belonging to the feature set, if the number of data samples contained within its optimal neighborhood radius exceeds the optimal density threshold, then the sample is called a core object; otherwise, it is not a core object. For any sample points 1 and 2, if sample 1 is a core object (i.e., it has at least MinPts samples in its ε-neighborhood), and sample 2 is within the ε-neighborhood of sample 1, then sample 2 is said to be directly density-reachable from sample 1. Finally, based on the density reachability calculation principle, the set of samples with the highest density is taken as a cluster, thus forming multiple clusters. Finally, the cluster center (i.e., the geometric center) obtained by DBSCAN clustering is used as a pre-selected charging station, and the remaining sample points are used as charging demand points, thus obtaining the charging demand point set. ( A (Total number of all vehicle trajectory sample points) and pre-selected charging station set ( B (This represents the number of clusters after clustering).

[0040] The flowchart of the DBSCAN algorithm is as follows: Figure 3 As shown, the real-time clustering process of trajectory coordinates in the vortex field optimization algorithm can refer to the application process of the DBSCAN algorithm described above, or the application process of the DBSCAN algorithm with optimal parameters in existing technologies. Additionally, it should be noted that the OPTICS algorithm can also be used for clustering trajectory data, with the neighborhood radius and minimum sample size as the parameters to be optimized; another example is the HDBSCAN algorithm, with the minimum cluster size as the parameter to be optimized. The specific optimization of these parameters can refer to the optimization process of the DBSCAN algorithm parameters in the vortex field optimization algorithm, but corresponding adjustments to the particle positions and initialization parameters are required. The application processes of other optimization algorithms can refer to the application processes of the corresponding algorithms with optimal parameters in existing technologies, and will not be elaborated upon here.

[0041] This invention improves the accuracy and adaptability of clustering algorithms by optimizing parameters, thereby obtaining a more reasonable set of charging demand points and a pre-selected set of charging stations. This enables more efficient solutions to problems such as charging station planning, reduces repetitive work and resource waste caused by inaccurate data clustering, and improves the overall efficiency of problem solving.

[0042] S4. Based on the set of charging demand points and the set of pre-selected charging stations, construct a minimum charging facility model and solve it to optimize the site selection layout of the pre-selected charging station set and obtain the final site selection planning scheme. In one embodiment, step S4 includes: Based on the set of charging demand points and the set of pre-selected charging stations, the minimum charging facility model is constructed with the objective of minimizing the number of charging stations and the constraint that each charging demand point is covered by the service range of at least one candidate charging station. The minimized charging facility model is solved using a preset optimization algorithm to optimize the site selection layout of the pre-selected charging site set and obtain the final site selection planning scheme.

[0043] Specifically, due to the excessive density of charging demand points in real-world scenarios, selecting charging stations can easily lead to high construction costs and low profitability for charging facilities. To eliminate some overlapping charging demands, this invention optimizes the site selection and layout of pre-selected charging stations by constructing a minimization charging facility model. Here, the pre-selected charging station set represents the most suitable candidate locations for building charging stations, while the charging demand point set represents the specific objects and ranges that need to be served. Therefore, based on the charging demand point set and the pre-selected charging station set, this invention constructs a minimization charging facility model with the objective of minimizing the number of selected charging stations and the constraint that each charging demand point is covered by the service range of at least one candidate charging station; the objective function of minimizing the charging facility model is: In the formula, F To minimize the number of charging facilities after addressing charging demand; n This represents the total number of pre-selected charging stations; For the first j The binary decision variable corresponding to the nth pre-selected charging station is set to 1 to represent the nth... j One pre-selected charging station is selected; 0 indicates that it is not selected.

[0044] The constraints for minimizing the charging facility model are: In the formula, To be able to cover charging demand points A collection of charging stations; m The total number of charging points; For charging demand points Is it at the charging station? The service area is defined by 1 representing coverage and 0 representing no coverage. A set of [variables] is determined by solving the minimization model of the charging infrastructure. The value of is chosen such that, assuming all charging demand points are covered by at least one selected station, the total number of selected stations (i.e., (Minimum)

[0045] By using a pre-defined optimization algorithm (such as the branch and bound method) to solve the minimization model of charging facilities, the problem of overlapping charging sites can be eliminated, and the optimal set of candidate charging sites can be obtained, which is the final site selection planning scheme.

[0046] In this invention, existing clustering algorithms for electric vehicle (EV) site selection planning rely heavily on manual trial and error to determine parameters, making it difficult to adapt to varying demand distributions across different cities. Furthermore, they fail to consider overlapping demand at pre-selected sites, leading to high construction costs and poor economic efficiency in the final site selection scheme. This invention proposes an EV charging station site selection planning method. This method achieves more accurate and convenient distance measurement of EVs by performing coordinate transformation on their trajectory data within the target area. A vortex field optimization algorithm is used to globally optimize the parameters of the DBSCAN algorithm, avoiding the subjectivity and limitations of manual trial and error, resulting in more stable clustering effects. By minimizing the charging facility model, overlapping site demand is eliminated, reducing the number of charging facilities while ensuring coverage of all demand points, significantly lowering construction costs, improving equipment utilization, and enhancing site profitability. Based on EV trajectory data and dynamic parameter optimization, this method is adaptable to cities of different sizes and demand distributions.

[0047] It should be noted that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders.

[0048] In another embodiment, such as Figure 4 As shown, a second aspect of the present invention provides an electric vehicle charging station site selection and planning system, comprising: The coordinate transformation module 10 is used to acquire the trajectory data of electric vehicles within the target area and perform coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles. The parameter optimization module 20 is used to optimize the parameters of the DBSCAN algorithm based on the vortex field optimization algorithm through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters. The trajectory clustering module 30 is used to cluster the trajectory coordinates by applying the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations. The site selection planning module 40 is used to construct and solve a minimum charging facility model based on the set of charging demand points and the set of pre-selected charging stations, so as to optimize the site selection layout of the set of pre-selected charging stations and obtain the final site selection planning scheme.

[0049] It should be noted that each module in the aforementioned electric vehicle charging station site selection and planning system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module. For specific limitations regarding the electric vehicle charging station site selection and planning system, please refer to the limitations of the electric vehicle charging station site selection and planning method described above; both have the same function and role, and will not be repeated here.

[0050] A third aspect of the present invention provides an electronic device comprising: Processor, memory, and bus; The bus is used to connect the processor and the memory; The memory is used to store operation instructions; The processor is configured to execute instructions by calling the operation instructions, causing the processor to perform operations corresponding to the electric vehicle charging station site selection and planning method shown in the first aspect of the present invention.

[0051] In one alternative embodiment, an electronic device is provided, such as Figure 5 As shown, Figure 5 The illustrated electronic device 5000 includes a processor 5001 and a memory 5003. The processor 5001 and the memory 5003 are connected, for example, via a bus 5002. Optionally, the electronic device 5000 may also include a transceiver 5004. It should be noted that in practical applications, the transceiver 5004 is not limited to one type, and the structure of this electronic device 5000 does not constitute a limitation on the embodiments of the present invention.

[0052] Processor 5001 may be a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in connection with this disclosure. Processor 5001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0053] Bus 5002 may include a path for transmitting information between the aforementioned components. Bus 5002 may be a PCI bus or an EISA bus, etc. Bus 5002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0054] The memory 5003 may be a ROM or other type of static storage device capable of storing static information and instructions, RAM or other type of dynamic storage device capable of storing information and instructions, or it may be an EEPROM, CD-ROM or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0055] The memory 5003 is used to store application code that executes the present invention, and its execution is controlled by the processor 5001. The processor 5001 is used to execute the application code stored in the memory 5003 to implement the content shown in any of the foregoing method embodiments.

[0056] Among them, electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers.

[0057] The fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the electric vehicle charging station site selection and planning method shown in the first aspect of the present invention.

[0058] Another embodiment of the present invention provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in the foregoing method embodiments.

[0059] Furthermore, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0060] In summary, this invention relates to the field of electric vehicle charging station site selection and planning, and discloses a method, system, equipment, and medium for electric vehicle charging station site selection and planning. It obtains trajectory coordinates by transforming the trajectory data of electric vehicles within a target area; based on the vortex field optimization algorithm, the parameters of the DBSCAN algorithm are optimized through a dual-force driving mechanism and a vortex splitting mechanism to obtain optimal parameters; the trajectory coordinates are clustered using the DBSCAN algorithm with optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations, thereby constructing and solving a minimized charging facility model to obtain the final site selection and planning scheme; the vortex field optimization algorithm globally optimizes the parameters of the clustering algorithm, avoiding the subjectivity and limitations of manual trial and error, resulting in more stable clustering effects; and the minimized charging facility model eliminates overlapping station demands, reducing the number of charging facilities and improving equipment utilization while ensuring coverage of all demand points.

[0061] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0062] The above-described embodiments are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for site selection and planning of electric vehicle charging stations, characterized in that, include: The trajectory data of electric vehicles within the target area is acquired, and the trajectory data is transformed to obtain the trajectory coordinates of the electric vehicles. Based on the vortex field optimization algorithm, the parameters of the DBSCAN algorithm are optimized through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters; By applying the DBSCAN algorithm with the optimal parameters, the trajectory coordinates are clustered to obtain a set of charging demand points and a set of pre-selected charging stations. Based on the set of charging demand points and the set of pre-selected charging stations, a minimum charging facility model is constructed and solved to optimize the site selection layout of the pre-selected charging station set and obtain the final site selection planning scheme.

2. The method for site selection and planning of electric vehicle charging stations according to claim 1, characterized in that, The step of acquiring trajectory data of electric vehicles within the target area and performing coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles includes: The trajectory data of electric vehicles within the target area is acquired, and the trajectory data is preprocessed to obtain the trajectory processing data of the electric vehicles. Based on the Mercator projection algorithm, the trajectory processing data is converted from latitude and longitude coordinates to plane rectangular coordinates to obtain the trajectory coordinates of the electric vehicle.

3. The method for site selection and planning of electric vehicle charging stations according to claim 1, characterized in that, The parameters of the DBSCAN algorithm include the neighborhood radius and the density threshold; wherein, The vortex field optimization algorithm optimizes the parameters of the DBSCAN algorithm through a dual-force driving mechanism and a vortex splitting mechanism to obtain the optimal parameters, including: Set the number of particles in the vortex field, and set the search space corresponding to the neighborhood radius and the density threshold respectively. Randomly generate an initial particle swarm based on the search space; each particle corresponds to a set of parameters of the DBSCAN algorithm. For each particle, the trajectory coordinates are clustered by applying the DBSCAN algorithm with the corresponding parameters of the particle to obtain multiple clusters. An optimization objective function is constructed based on the clusters to quantify the optimization objective value of each particle, and the particle with the smallest optimization objective value is taken as the global optimal particle. Based on the aforementioned dual-force driving mechanism, the positions of each particle are updated through the coordinated driving of centripetal force and tangential force; The positional aggregation degree of each particle after the update is quantified, and when the positional aggregation degree exceeds the preset aggregation degree requirement, the vortex splitting mechanism is triggered to introduce Levi flight to randomly generate new particles and obtain a new population. The process of determining the global optimal particle and generating the new population is performed iteratively according to the new population until the maximum preset number of iterations is reached, and the position of the global optimal particle in this iteration is used as the optimal parameter.

4. The method for site selection and planning of electric vehicle charging stations according to claim 3, characterized in that, The method of updating the position of each particle based on the dual-force driving mechanism, through the coordinated driving of centripetal force and tangential force, includes: Obtain the current iteration number, and determine the centripetal force coefficient that controls the centripetal force intensity and the tangential force coefficient that controls the tangential force disturbance intensity based on the current iteration number; The centripetal force data of each particle is determined based on the centripetal force coefficient and the position of the globally optimal particle, and the tangential force data of each particle is determined based on the tangential force coefficient and the position of the globally optimal particle. Based on the centripetal force data and tangential force data of each particle, the positions of each particle are updated collaboratively to obtain the new positions of each particle.

5. The method for site selection and planning of electric vehicle charging stations according to claim 3, characterized in that, The step of constructing the optimization objective function based on the clusters includes: The DB index of each cluster is quantified, and minimizing the DB index is used as the optimization objective function.

6. The method for site selection and planning of electric vehicle charging stations according to claim 3, characterized in that, The process of clustering the trajectory coordinates using the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations includes: Based on the optimal neighborhood radius and the optimal density threshold, the core object in the trajectory coordinates is determined, and the density reachable sample in the trajectory coordinates is determined based on the core object; Based on the principle of density reachability calculation, all density reachable samples connected by the maximum density are grouped into one cluster, resulting in multiple clusters; The geometric center of each cluster is calculated as a pre-selected charging station, and the remaining samples within the cluster are used as charging demand points to obtain a set of pre-selected charging stations and a set of charging demand points.

7. The method for site selection and planning of electric vehicle charging stations according to claim 1, characterized in that, The step of constructing and solving a minimum charging facility model based on the set of charging demand points and the set of pre-selected charging stations to optimize the site selection layout of the pre-selected charging station set and obtain the final site selection planning scheme includes: Based on the set of charging demand points and the set of pre-selected charging stations, the minimum charging facility model is constructed with the objective of minimizing the number of charging stations and the constraint that each charging demand point is covered by the service range of at least one candidate charging station. The minimized charging facility model is solved using a preset optimization algorithm to optimize the site selection layout of the pre-selected charging site set and obtain the final site selection planning scheme.

8. A site selection and planning system for electric vehicle charging stations, characterized in that, include: The coordinate transformation module is used to acquire the trajectory data of electric vehicles within the target area and perform coordinate transformation on the trajectory data to obtain the trajectory coordinates of the electric vehicles. The parameter optimization module is used to optimize the parameters of the DBSCAN algorithm based on the vortex field optimization algorithm through the dual-force driving mechanism and the vortex splitting mechanism to obtain the optimal parameters. The trajectory clustering module is used to cluster the trajectory coordinates by applying the DBSCAN algorithm with the optimal parameters to obtain a set of charging demand points and a set of pre-selected charging stations. The site selection planning module is used to construct and solve a minimum charging facility model based on the set of charging demand points and the set of pre-selected charging stations, so as to optimize the site selection layout of the set of pre-selected charging stations and obtain the final site selection planning scheme.

9. An electronic device, characterized in that, The device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the electric vehicle charging station site selection planning method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the electric vehicle charging station site selection and planning method as described in any one of claims 1 to 7.

Images