Game-based method and system for siting electric vehicle charging stations

Abstract

The application provides a game-based electric vehicle charging station site selection method and system, and relates to the technical field of charging station site selection. The upper target function is determined based on the game strategy of the charging station operator, the lower target function is determined based on the game strategy of the electric vehicle owner, then the charging station site selection optimization model is constructed based on the upper target function and the lower target function, and the charging station site selection optimization model is solved based on the preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme. The application considers the information interaction between each subject in the electric vehicle charging station site selection process, and the charging station site selection result obtained based on the game between the charging station operator and the electric vehicle owner is more accurate, so that the interests of the charging station operator and the electric vehicle owner are maximized, and the income and resource utilization rate of the charging station operator are improved.

Description

Technical Field

[0001] This invention relates to the field of charging station site selection technology, and specifically to a game-theoretic method and system for selecting the site of electric vehicle charging stations. Background Technology

[0002] The insufficient number of electric vehicle charging infrastructures is one of the major obstacles to the wider adoption of electric vehicles. Because the construction cost of electric vehicle charging stations is relatively high, improper construction can lead to a significant waste of resources and increased costs. Therefore, the scientific and rational selection of electric vehicle charging station locations is of great practical significance.

[0003] Currently, the site selection and planning for electric vehicle charging stations mainly involves allocating demand based on traffic flow. This involves establishing charging demand on a set of origin-destination travel pairs and solving them using precise algorithms, heuristic algorithms, or solvers to obtain a site selection and planning scheme for the charging stations.

[0004] However, when planning the location of charging stations based on the above methods, the interests of charging station operators and electric vehicle owners are often considered as unrelated entities, and the location of charging stations is selected on this basis. Or, it only considers the game behavior of charging station operators aiming to maximize their own interests and electric vehicle owners aiming to complete their trips with the least amount of time and cost, while ignoring the game process in which electric vehicle owners choose deviated routes by adjusting charging prices through charging station operators. This game process does not take into account the real situation under the information interaction environment between charging station operators and electric vehicle owners, which leads to inaccurate location results and failure to improve resource utilization. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a game-theoretic method and system for selecting the location of electric vehicle charging stations, which solves the problem of low accuracy in existing electric vehicle charging station location technologies due to the lack of consideration for information interaction between various stakeholders.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] Firstly, this invention proposes a game-theoretic method for selecting the location of electric vehicle charging stations, the method comprising:

[0010] The upper-level objective function is determined based on the game strategy of charging station operators, and the lower-level objective function is determined based on the game strategy of electric vehicle owners.

[0011] A charging station site selection optimization model is constructed based on the upper-level objective function and the lower-level objective function, and the charging station site selection optimization model is solved based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm.

[0012] Preferably, the upper-level objective function is:

[0013] Maximize C e -C o

[0014]

[0015] C o =C n +C p +C t +C l

[0016]

[0017]

[0018]

[0019]

[0020]

[0021] Among them, C e Indicates the annual revenue of the charging station operator; C o M represents the annual cost of a charging station operator; h represents the set of origin-destination trip pairs; R represents a single origin-destination trip pair. h R represents the set of alternative routes for electric vehicle owners on the origin-destination journey pair h; r represents the set of alternative routes on R. h Alternative paths in; p represents the average daily total energy consumption of an electric vehicle traveling along path r with the origin-destination journey pair h as the starting point and destination; s This indicates the price at which a charging station sells a unit of electricity; For a 0-1 decision variable, if the electric vehicle owner chooses r as the driving route for the origin-destination trip h, then... Select 1 if the value is 1, otherwise select 0; C n C represents the annual depreciation value of the construction cost of a charging station; p C represents the annual depreciation value of the cost of purchasing charging piles for charging stations; t This indicates the annual operating cost of the charging station; C lThe value represents the cost of obtaining electricity for the charging station; N represents the set of candidate charging stations, i represents the candidate charging station, r0 represents the annual depreciation factor of the charging pile, and c represents the annual depreciation factor of the charging pile. z x represents the construction cost of a charging station. i Let x represent a 0-1 decision variable, where x is the decision variable if a charging station is built at candidate site i. i Select 1 otherwise select 0, y e Indicates the service life of the charging station; c r This indicates the purchase cost of the charging station; y s β represents the usable lifespan of the charging station; p represents the annual operating cost of the charging station. e This represents the price per unit of electricity sold by the charging station; r0 represents the annual depreciation factor; g represents a positive constant.

[0022] The upper-level objective function includes the following decision variables and constraint functions:

[0023] a. Path constraints:

[0024] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0025] b. Decision variables:

[0026] This constraint indicates 0-1 decision variables;

[0027] This constraint represents x i Represents 0-1 variables;

[0028] p s The constraint ≥0 indicates that p s It is a non-negative variable;

[0029] The objective function of the lower layer is:

[0030]

[0031] in, f represents the distance of the shortest path s among the alternative paths. h Let γ represent the flow rate over the origin-destination journey h, and let p represent the electricity consumption of the electric vehicle per unit distance traveled. a This indicates the electricity price charged to electric vehicle owners before they changed their routes. p represents the distance of path r among the candidate paths. b θ represents the electricity price after the electric vehicle owner changes their route, θ represents the conversion factor between distance and time, and ε represents the cost per unit time.

[0032] The decision variables and constraint functions of the lower-level objective function are as follows:

[0033] a. Electricity price constraints:

[0034] p a ≥p b This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity.

[0035] b. Path constraints:

[0036] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0037] This constraint means that in the origin-destination trip pair h, if the deviation distance of the deviation path exceeds twice that of the shortest alternative path, it will not be selected by the electric vehicle owner;

[0038] c. Decision variables:

[0039] This constraint indicates 0-1 decision variables;

[0040] This constraint represents x i Represents 0-1 variables;

[0041] p a ,p b The constraint ≥0 indicates that p a ,p b It is a non-negative variable.

[0042] Preferably, the charging station site selection optimization model is constructed based on the upper-level objective function and the lower-level objective function, and the model is solved based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm including:

[0043] S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm;

[0044] S22. Under the condition of satisfying the electric vehicle battery capacity limit, select the initial position of the electric vehicle charging station on the initialized electric vehicle path based on the iterative greedy algorithm;

[0045] S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme.

[0046] S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

[0047] Preferably, step S23, further optimizing the initial location of the electric vehicle charging station based on the adaptive large neighborhood search algorithm to obtain an optimized charging station site selection scheme, and outputting the current charging station site selection scheme as the optimal charging station site selection scheme when the adaptive large neighborhood search algorithm stops iterating and generates a better solution, includes:

[0048] The initial location of electric vehicle charging stations is modified and optimized by introducing deletion and insertion operators into the adaptive large neighborhood search algorithm.

[0049] Preferably, the deletion operator includes: random deletion, worst-case deletion, single-point removal, and two-point removal operator; the insertion operator includes: greedy insertion, k-regret value insertion, and random insertion operator.

[0050] Secondly, the present invention also proposes a game-theoretic-based site selection system for electric vehicle charging stations, the system comprising:

[0051] The objective function acquisition module is used to determine the upper-level objective function based on the game strategy of the charging station operator and the lower-level objective function based on the game strategy of the electric vehicle owner.

[0052] The charging station site selection scheme acquisition module is used to construct a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solve the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm.

[0053] Preferably, the upper-level objective function in the objective function acquisition module is:

[0054] Maximize C e -C o

[0055]

[0056] Co =C n +C p +C t +C l

[0057]

[0058]

[0059]

[0060]

[0061]

[0062] Among them, C e Indicates the annual revenue of the charging station operator; C o M represents the annual cost of a charging station operator; h represents the set of origin-destination trip pairs; R represents a single origin-destination trip pair. h R represents the set of alternative routes for electric vehicle owners on the origin-destination journey pair h; r represents the set of alternative routes on R. h Alternative paths in; p represents the average daily total energy consumption of an electric vehicle traveling along path r with the origin-destination journey pair h as the starting point and destination; s This indicates the price at which a charging station sells a unit of electricity; For a 0-1 decision variable, if the electric vehicle owner chooses r as the driving route for the origin-destination trip h, then... Select 1 if the value is 1, otherwise select 0; C n C represents the annual depreciation value of the construction cost of a charging station; p C represents the annual depreciation value of the cost of purchasing charging piles for charging stations; t This indicates the annual operating cost of the charging station; C l The value represents the cost of obtaining electricity for the charging station; N represents the set of candidate charging stations, i represents the candidate charging station, r0 represents the annual depreciation factor of the charging pile, and c represents the annual depreciation factor of the charging pile. z x represents the construction cost of a charging station. i Let x represent a 0-1 decision variable, where x is the decision variable if a charging station is built at candidate site i. i Select 1 otherwise select 0, y e Indicates the service life of the charging station; c r This indicates the purchase cost of the charging station; y s β represents the usable lifespan of the charging station; p represents the annual operating cost of the charging station. e This represents the price per unit of electricity sold by the charging station; r0 represents the annual depreciation factor; g represents a positive constant.

[0063] The upper-level objective function includes the following decision variables and constraint functions:

[0064] a. Path constraints:

[0065] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0066] b. Decision variables:

[0067] This constraint indicates 0-1 decision variables;

[0068] This constraint represents x i Represents 0-1 variables;

[0069] p s The constraint ≥0 indicates that p s It is a non-negative variable;

[0070] The objective function of the lower layer is:

[0071]

[0072] in, f represents the distance of the shortest path s among the alternative paths. h Let γ represent the flow rate over the origin-destination journey h, and let p represent the electricity consumption of the electric vehicle per unit distance traveled. a This indicates the electricity price charged to electric vehicle owners before they changed their routes. p represents the distance of path r among the candidate paths. b θ represents the electricity price after the electric vehicle owner changes their route, θ represents the conversion factor between distance and time, and ε represents the cost per unit time.

[0073] The decision variables and constraint functions of the lower-level objective function are as follows:

[0074] a. Electricity price constraints:

[0075] p a ≥p b This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity.

[0076] b. Path constraints:

[0077] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0078] This constraint means that in the origin-destination trip pair h, if the deviation distance of the deviation path exceeds twice that of the shortest alternative path, it will not be selected by the electric vehicle owner;

[0079] c. Decision variables:

[0080] This constraint indicates 0-1 decision variables;

[0081] This constraint represents x i Represents 0-1 variables;

[0082] p a ,p b The constraint ≥0 indicates that p a ,p b It is a non-negative variable.

[0083] Preferably, the charging station site selection scheme acquisition module constructs a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solves the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm including:

[0084] S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm;

[0085] S22. Under the condition of satisfying the electric vehicle battery capacity limit, select the initial position of the electric vehicle charging station on the initialized electric vehicle path based on the iterative greedy algorithm;

[0086] S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme.

[0087] S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

[0088] Preferably, step S23, further optimizing the initial location of the electric vehicle charging station based on the adaptive large neighborhood search algorithm to obtain an optimized charging station site selection scheme, and outputting the current charging station site selection scheme as the optimal charging station site selection scheme when the adaptive large neighborhood search algorithm stops iterating and generates a better solution, includes:

[0089] The initial location of electric vehicle charging stations is modified and optimized by introducing deletion and insertion operators into the adaptive large neighborhood search algorithm.

[0090] Preferably, the deletion operator includes: random deletion, worst-case deletion, single-point removal, and two-point removal operator; the insertion operator includes: greedy insertion, k-regret value insertion, and random insertion operator.

[0091] (III) Beneficial Effects

[0092] This invention provides a game-theoretic method and system for selecting the location of electric vehicle charging stations. Compared with existing technologies, it has the following advantages:

[0093] This invention determines the upper-level objective function based on the game strategy of charging station operators and the lower-level objective function based on the game strategy of electric vehicle owners. Then, a charging station site selection optimization model is constructed based on these upper and lower-level objective functions. The model is solved using a pre-defined heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme. This invention considers the information interaction among various stakeholders in the electric vehicle charging station site selection process. Based on a realistic game between charging station operators and electric vehicle owners, the obtained charging station site selection results are more accurate, thereby maximizing the interests of both charging station operators and electric vehicle owners, while simultaneously improving the revenue and resource utilization rate of charging station operators. Attached Figure Description

[0094] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are 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.

[0095] Figure 1 This is an embodiment of a game-theoretic method for selecting the location of electric vehicle charging stations according to the present invention;

[0096] Figure 2 This is a flowchart illustrating how a preset heuristic algorithm is used to solve for the optimal location scheme of an electric vehicle charging station in an embodiment of the present invention. Detailed Implementation

[0097] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0098] This application provides a game-theoretic-based method and system for selecting the location of electric vehicle charging stations. This solves the problem that existing electric vehicle charging station location technologies suffer from low accuracy due to a lack of consideration for information interaction between stakeholders. The method aims to increase the revenue of charging station operators and electric vehicle owners, as well as improve resource utilization.

[0099] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows:

[0100] To address the issue of low accuracy in existing electric vehicle charging station site selection technologies due to the lack of consideration for information exchange between stakeholders, this application first determines the upper-level objective function for charging station operators and the lower-level objective function for electric vehicle owners based on their respective game strategies. Simultaneously, it determines the decision variables and constraints for both the upper and lower-level objective functions based on practical application scenarios. Then, a charging station site selection optimization model is constructed based on the upper and lower-level objective functions and their corresponding decision variables and constraints. When collected traffic network information, charging station data, and relevant data for electric vehicles are input into the model, the electric vehicle paths are first initialized using the k-shortest path algorithm, and the station locations are initialized using an iterative greedy algorithm. Then, an adaptive large neighborhood search algorithm optimizes the charging station site selection from the charging station operator's perspective. Finally, an iterative greedy algorithm optimizes the electric vehicle path selection from the electric vehicle owner's perspective. This application's technical solution provides more accurate site selection results for electric vehicle charging stations, achieving a win-win situation for both charging station operators and electric vehicle owners, while also improving resource utilization.

[0101] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0102] Example 1:

[0103] Firstly, this invention proposes a game-theoretic method for selecting the location of electric vehicle charging stations, see [link to relevant documentation]. Figure 1 The method includes:

[0104] S1. Determine the upper-level objective function based on the game strategy of charging station operators, and determine the lower-level objective function based on the game strategy of electric vehicle owners;

[0105] S2. Construct a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solve the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm.

[0106] As can be seen, this embodiment determines the upper-level objective function based on the game strategy of charging station operators and the lower-level objective function based on the game strategy of electric vehicle owners. Then, a charging station site selection optimization model is constructed based on these upper and lower-level objective functions, and a preset heuristic algorithm is used to solve the model to obtain the optimal electric vehicle charging station site selection scheme. This embodiment considers the information interaction between various stakeholders in the electric vehicle charging station site selection process. Based on a realistic game between charging station operators and electric vehicle owners, the obtained charging station site selection results are more accurate.

[0107] The following is in conjunction with the appendix Figures 1-2 The following details the implementation process of an embodiment of the present invention, including explanations of the specific steps S1-S2.

[0108] S1. The upper-level objective function is determined based on the game strategy of the charging station operator, and the lower-level objective function is determined based on the game strategy of the electric vehicle owner.

[0109] In the site selection process of electric vehicle charging stations, both charging station operators and electric vehicle owners have their own corresponding game strategies based on their own interests. The charging station operator's strategy is to maximize its own revenue, setting net profit as a linear function of revenue and cost, thus incentivizing electric vehicle owners to choose off-ramp paths to reduce fixed costs and increase revenue. The electric vehicle owner's strategy is to minimize path costs as much as possible, by choosing off-ramp paths to incentivize charging station operators to lower electricity prices, thereby reducing driving costs. Based on their respective game strategies, the charging station operator determines the upper-level objective function, and the electric vehicle owner determines the lower-level objective function. Specifically:

[0110] The upper-level objective function is:

[0111] Maximize C e -C o

[0112]

[0113] C o =Cn +C p +C t +C l

[0114]

[0115]

[0116]

[0117]

[0118]

[0119] Among them, C e Indicates the annual revenue of the charging station operator; C o M represents the annual cost of a charging station operator; h represents the set of origin-destination trip pairs; R represents a single origin-destination trip pair. h R represents the set of alternative routes for electric vehicle owners on the origin-destination journey pair h; r represents the set of alternative routes on R. h Alternative paths in; p represents the average daily total energy consumption of an electric vehicle traveling along path r with the origin-destination journey pair h as the starting point and destination; s This indicates the price at which a charging station sells a unit of electricity; For a 0-1 decision variable, if the electric vehicle owner chooses r as the driving route for the origin-destination trip h, then... Select 1 if the value is 1, otherwise select 0; C n C represents the annual depreciation value of the construction cost of a charging station; p C represents the annual depreciation value of the cost of purchasing charging piles for charging stations; t This indicates the annual operating cost of the charging station; C l The value represents the cost of obtaining electricity for the charging station; N represents the set of candidate charging stations, i represents the candidate charging station, r0 represents the annual depreciation factor of the charging pile, and c represents the annual depreciation factor of the charging pile. z x represents the construction cost of a charging station. i Let x represent a 0-1 decision variable, where x is the decision variable if a charging station is built at candidate site i. i Select 1 otherwise select 0, y e Indicates the service life of the charging station; c r This indicates the purchase cost of the charging station; y s β represents the usable lifespan of the charging station; p represents the annual operating cost of the charging station. e This represents the price per unit of electricity sold by the charging station; r0 represents the annual depreciation factor; and g represents a positive positive number.

[0120] Meanwhile, based on practical considerations, when determining the upper-level objective function, charging station operators should include the following decision variables and constraint functions:

[0121] a. Path constraints:

[0122] This constraint means that electric vehicle owners must choose only one route from all available options as their driving route.

[0123] b. Decision variables:

[0124] This constraint indicates 0-1 decision variables;

[0125] This constraint represents x i Represents 0-1 variables;

[0126] p s The constraint ≥0 indicates that p s It is a non-negative variable.

[0127] The objective function of the lower layer is:

[0128]

[0129] in, f represents the distance of the shortest path s among the alternative paths. h Let γ represent the flow rate over the origin-destination journey h, and let p represent the electricity consumption of the electric vehicle per unit distance traveled. a This indicates the electricity price charged to electric vehicle owners before they changed their routes. p represents the distance of path r among the candidate paths. b This represents the electricity price after the electric vehicle owner changes their route. In this embodiment, it is assumed that all electric vehicles travel at a constant speed. θ represents the conversion coefficient between distance and time, and ε represents the cost per unit time.

[0130] The decision variables and constraint functions of the lower-level objective function are as follows:

[0131] a. Electricity price constraints:

[0132] p a ≥p b This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity.

[0133] b. Path constraints:

[0134] This constraint means that electric vehicle owners must choose only one route from all available options as their driving route.

[0135] This constraint means that in the origin-destination trip pair h, if the deviation distance of the deviation path exceeds twice that of the shortest alternative path, the electric vehicle owner will not choose it.

[0136] c. Decision variables:

[0137] This constraint indicates 0-1 decision variables;

[0138] This constraint represents x i Represents 0-1 variables;

[0139] p a ,p b The constraint ≥0 indicates that p a ,p b It is a non-negative variable.

[0140] S2. Construct a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solve the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm.

[0141] The system collects traffic network information, charging station data, and relevant data on electric vehicles, and inputs this data into a charging station site selection optimization model constructed from upper-level and lower-level objective functions for solution. The relevant data includes: location information of traffic network nodes, traffic flow for origin-destination travel pairs, electric vehicle power and average speed, annual depreciation coefficient of charging piles, construction cost of charging stations, service life of charging stations, purchase cost of charging piles, service life of charging piles, annual operating cost of charging piles, price per unit of electricity obtained by charging stations, and unit time cost for electric vehicle owners.

[0142] To solve the constructed charging station site selection optimization model, this embodiment employs a heuristic algorithm. This algorithm comprises the following components: a k-shortest path algorithm, an iterative greedy algorithm, and an adaptive large neighborhood search. This embodiment first uses the k-shortest path algorithm to generate initial paths; then, based on the iterative greedy algorithm, it provides initial site locations. Next, the adaptive large neighborhood search algorithm optimizes the charging station site selection from the perspective of the charging station operator; finally, the iterative greedy algorithm is used to further optimize the electric vehicle route selection from the perspective of the electric vehicle owner. See also... Figure 2 The specific process is as follows:

[0143] S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm.

[0144] In the initialization phase of electric vehicle paths, the k-shortest path algorithm is used, ignoring the limitation of electric vehicle battery range, to create a candidate path set R. h First, determine the shortest path for each origin-destination travel pair, and default this shortest path as the current travel route for the vehicle in that origin-destination travel pair. Then, based on the path constraints mentioned above ( This constraint states that in a origin-destination trip pair h, if the deviation distance of the deviation path exceeds twice that of the shortest alternative path, it will not be selected by the electric vehicle owner. It limits the length of the alternative paths and generates alternative paths for each origin-destination trip pair one by one. Because the electric vehicle battery capacity limitation is ignored during path initialization, the initialized electric vehicle path may violate the electric vehicle mileage constraint.

[0145] S22. Under the condition of satisfying the electric vehicle battery capacity limit, the initial position of the electric vehicle charging station is selected on the electric vehicle path after initialization based on the iterative greedy algorithm.

[0146] After initializing the electric vehicle paths, to ensure that each selected path meets the battery capacity constraint, it is necessary to initialize the electric vehicle charging station locations on the initialized electric vehicle paths. When initializing the electric vehicle charging station locations, each origin-destination travel pair uses its shortest path as the current driving path of the electric vehicle. To improve the feasibility of the solution, this embodiment uses an iterative greedy algorithm to solve the electric vehicle charging station location initialization problem.

[0147] First, identify breakpoints along the current path of the electric vehicle and represent the feasibility state of each node. Breakpoints are defined as nodes where the maximum distance the electric vehicle can travel with remaining battery power upon reaching the node is less than 0. The feasibility state of a node is the smaller of the maximum distance the vehicle can travel with remaining battery power upon reaching the node and 0. The lowest feasibility state of nodes along the current path of a trip pair is the worst-case feasibility state. Based on the feasibility state, the worst-case feasibility state, and the cost of building a charging station, construct the allocation cost for placing a charging station at each node. Specifically, the feasibility state of node k is defined as a... kr The worst-case feasibility state in path r is represented as Allocation Costs Used to evaluate the magnitude of the improvement to the solution after placing the charging station at position k.

[0148]

[0149]

[0150]

[0151]

[0152] eta1+eta2+eta3=1,eta1,eta2,eta3≥0

[0153] The placement gain is defined as follows: This indicates the degree of improvement in the feasibility of the route after building a charging station at location k; the placement cost is defined as... This represents the construction cost of the charging station at node k; and This indicates the penalty for an unfeasible path. Additionally, a kr and These represent the feasible state of node k before the charging station is placed and the worst-case feasible state of path r, respectively. Similarly, and This represents the feasibility state of node k after placing a charging station at node k and the worst-case feasibility state of path r. η1, η2, and η3 are weighting coefficients. M is a large number.

[0154] Check if each trip is feasible for the current travel path; if not, add it to the infeasible path set. For paths in the infeasible path set, calculate the node allocation cost c based on the above solution. r k The location of the charging station is selected by roulette, and then the feasibility of the path is checked until all paths are feasible.

[0155] S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme.

[0156] Under the condition of satisfying the battery capacity limit of electric vehicles, the initial location of the electric vehicle charging station is selected on the initialized electric vehicle path. Based on this, we optimize the location of the charging station using the Adaptive Large Neighborhood Search (ALNS) algorithm without considering path selection. Specifically, we modify and optimize the charging station location on the current driving path by introducing deletion and insertion operators into the ALNS algorithm.

[0157] The deletion operators include: random deletion: randomly delete a certain number of stations from the current solution; worst-case deletion: delete a certain number of stations with the least benefit from the current solution; single-point removal: randomly select a path and randomly select a position in this path, and delete the station between that position and the original point or destination; two-point removal: randomly select two positions in the route, and delete the station between the two positions.

[0158] Insertion operators include: Greedy insertion: inserting a station at the node with the highest insertion gain; k-regret insertion: inserting a station at the node with the smallest regret value, where the regret value of a node is obtained by the difference between the gain generated by the optimal insertion position and the gain generated by the k-optimal insertion position; and Random insertion: randomly inserting a station among the candidate nodes.

[0159] Since only the interests of the charging station operators were considered when selecting charging station sites, the objective function used by the adaptive large neighborhood search algorithm in this embodiment is the operator's profit, i.e., Z. p =C e -C o .

[0160] In optimizing the location of charging stations using the adaptive large neighborhood search algorithm, we divide the algorithm's iterative process into multiple segments, each containing a certain number of iterations. Specifically:

[0161] The overall iterative process is set to iterate τ times, with intervals of ξ times, and the iterative process is divided into... The process is divided into several segments. In the first iteration, all operator weights are set to the same initial value. At each iteration, a pair of deletion and insertion operators are selected using a roulette wheel mechanism. This is modified for the current solution, and each time an operator is called, its weight changes based on its performance. Operators with larger weights indicate that they performed better in previous processes and therefore have a greater chance of being selected in subsequent iterations. Operator weights are updated at the end of each iteration segment, specifically according to the rule: if χ... ij ≠0, w i(j+1) =(1-μ)w ij +μυ ij / χ ij Otherwise w i(j+1) =w ij , where χ ij and υ ij These represent the number of times operator i is used in the j-th iteration segment and the score of operator i in the j-th iteration segment, respectively; μ is the response factor; w i(j+1) w represents the weight of operator i in segment j+1; ij This represents the weight of operator i in segment j.

[0162] υij The calculation rules are as follows: the score is initially set to 0 and incremented by δ each time it is used. The size of δ depends on the historical performance of operator i in the iteration; if a pair of deletion and insertion operators results in a new global optimal solution, they will receive a score of δ1; if the generated solution is better than the current solution, they will receive a score of δ2; if the solution is merely accepted, they will receive a score of δ3.

[0163] The adaptive large neighborhood search algorithm simulates the iterative process as molecular motion as the temperature decreases. start Let t represent the initial temperature and the current temperature, respectively. At the start of the iteration, the current temperature is set to t = T. start Then, after a given number of iterations, the temperature is updated to t = t-1. The algorithm stops when the temperature drops below 0°C. The global optimal solution s is then checked. best With the initial solution s start If a better solution is found, proceed to the next step; otherwise, terminate the global algorithm and output the current charging station location scheme, which is the optimal charging station location scheme.

[0164] S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

[0165] The iterative greedy algorithm is used to select alternative paths. The basic idea is to iteratively select alternative paths to adjust the electricity price at charging stations and reduce costs for electric vehicle owners. The specific process is as follows:

[0166] Check if the current travel path for all origin-destination trip pairs contains a charging station. If it does, add the origin-destination trip pair to the changeable set M. c middle.

[0167] Check the changeable set M c If there are no breakpoints in the alternative paths of a given journey pair, then add that path to the set of feasible alternative paths R. o middle.

[0168] For a mutable set M cThe alternative routes for the origin-destination trip pair are reselected based on path distance using a roulette wheel approach. When a new alternative route is selected, charging stations that no longer provide traffic services are removed. Then, it is checked whether the cost for the charging vehicle owner has decreased. If the cost has decreased, the selection result is accepted; otherwise, it is discarded. It is checked whether the maximum number of iterations has been reached. If it has, the current iteration ends and the next step is performed; otherwise, the alternative routes are selected again.

[0169] Check if the global optimal solution is better than the initial solution. If a better solution is found, return to S23 to optimize the charging station location. If no better solution is found, end the global algorithm iteration and output the current charging station location scheme as the optimal charging station location scheme.

[0170] This completes the entire process of the game-theoretic site selection method for electric vehicle charging stations in this embodiment.

[0171] Example 2:

[0172] Secondly, the present invention also provides a game-theoretic site selection system for electric vehicle charging stations, the system comprising:

[0173] The objective function acquisition module is used to determine the upper-level objective function based on the game strategy of the charging station operator and the lower-level objective function based on the game strategy of the electric vehicle owner.

[0174] The charging station site selection scheme acquisition module is used to construct a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solve the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm.

[0175] Optionally, the upper-level objective function in the objective function acquisition module is:

[0176] Maximize C e -C o

[0177]

[0178] C o =C n +C p +C t +C l

[0179]

[0180]

[0181]

[0182]

[0183]

[0184] Among them, C e Indicates the annual revenue of the charging station operator; C o M represents the annual cost of a charging station operator; h represents the set of origin-destination trip pairs; R represents a single origin-destination trip pair. h R represents the set of alternative routes for electric vehicle owners on the origin-destination journey pair h; r represents the set of alternative routes on R. h Alternative paths in; p represents the average daily total energy consumption of an electric vehicle traveling along path r with the origin-destination journey pair h as the starting point and destination; s This indicates the price at which a charging station sells a unit of electricity; For a 0-1 decision variable, if the electric vehicle owner chooses r as the driving route for the origin-destination trip h, then... Select 1 if the value is 1, otherwise select 0; C n C represents the annual depreciation value of the construction cost of a charging station; p C represents the annual depreciation value of the cost of purchasing charging piles for charging stations; t This indicates the annual operating cost of the charging station; C l The value represents the cost of obtaining electricity for the charging station; N represents the set of candidate charging stations, i represents the candidate charging station, r0 represents the annual depreciation factor of the charging pile, and c represents the annual depreciation factor of the charging pile. z x represents the construction cost of a charging station. i Let x represent a 0-1 decision variable, where x is the decision variable if a charging station is built at candidate site i. i Select 1 otherwise select 0, y e Indicates the service life of the charging station; c r This indicates the purchase cost of the charging station; y s β represents the usable lifespan of the charging station; p represents the annual operating cost of the charging station. e This represents the price per unit of electricity sold by the charging station; r0 represents the annual depreciation factor; and g represents a positive constant.

[0185] The upper-level objective function includes the following decision variables and constraint functions:

[0186] a. Path constraints:

[0187] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0188] b. Decision variables:

[0189] This constraint indicates 0-1 decision variables;

[0190] This constraint represents x i Represents 0-1 variables;

[0191] p s The constraint ≥0 indicates that p s It is a non-negative variable.

[0192] The objective function of the lower layer is:

[0193]

[0194] in, f represents the distance of the shortest path s among the alternative paths. h Let γ represent the flow rate over the origin-destination journey h, and let p represent the electricity consumption of the electric vehicle per unit distance traveled. a This indicates the electricity price charged to electric vehicle owners before they changed their routes. p represents the distance of path r among the candidate paths. b The electricity price after the electric vehicle owner changes the route is represented. In this invention, it is assumed that the electric vehicles travel at a constant speed. θ represents the conversion coefficient between distance and time, and ε represents the unit time cost.

[0195] The decision variables and constraint functions of the lower-level objective function are as follows:

[0196] a. Electricity price constraints:

[0197] p a ≥p b This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity.

[0198] b. Path constraints:

[0199] This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route;

[0200] This constraint means that in the origin-destination trip pair h, if the deviation distance of the deviation path exceeds twice that of the shortest alternative path, it will not be selected by the electric vehicle owner;

[0201] c. Decision variables:

[0202] This constraint indicates 0-1 decision variables;

[0203] This constraint represents x i Represents 0-1 variables;

[0204] p a ,p b The constraint ≥0 indicates that p a ,p b It is a non-negative variable.

[0205] Optionally, the charging station site selection scheme acquisition module constructs a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solves the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm including:

[0206] S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm;

[0207] S22. Under the condition of satisfying the electric vehicle battery capacity limit, select the initial position of the electric vehicle charging station on the initialized electric vehicle path based on the iterative greedy algorithm;

[0208] S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme.

[0209] S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

[0210] Optionally, step S23, further optimizing the initial location of the electric vehicle charging station based on the adaptive large neighborhood search algorithm to obtain an optimized charging station site selection scheme, and outputting the current charging station site selection scheme as the optimal charging station site selection scheme when the adaptive large neighborhood search algorithm stops iterating and generates a better solution, includes:

[0211] The initial location of electric vehicle charging stations is modified and optimized by introducing deletion and insertion operators into the adaptive large neighborhood search algorithm.

[0212] Optionally, the deletion operator includes: random deletion, worst-case deletion, single-point removal, and two-point removal operator; the insertion operator includes: greedy insertion, k-regret value insertion, and random insertion operator.

[0213] It is understood that the game-theoretic electric vehicle charging station site selection system provided in this embodiment corresponds to the game-theoretic electric vehicle charging station site selection method described above. The explanations, examples, and beneficial effects of the relevant content can be referred to the corresponding content in the game-theoretic electric vehicle charging station site selection method, and will not be repeated here.

[0214] In summary, compared with existing technologies, it has the following beneficial effects:

[0215] This invention determines the upper-level objective function based on the game strategy of charging station operators and the lower-level objective function based on the game strategy of electric vehicle owners. Then, a charging station site selection optimization model is constructed based on these upper and lower-level objective functions. The model is solved using a pre-defined heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme. This invention considers the information interaction among various stakeholders in the electric vehicle charging station site selection process. Based on a realistic game between charging station operators and electric vehicle owners, the obtained charging station site selection results are more accurate, thereby maximizing the interests of both charging station operators and electric vehicle owners, while simultaneously improving the revenue and resource utilization rate of charging station operators.

[0216] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0217] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A game-theoretic method for selecting the location of electric vehicle charging stations, characterized in that, The method includes: The upper-level objective function is determined based on the game strategy of charging station operators, and the lower-level objective function is determined based on the game strategy of electric vehicle owners. A charging station site selection optimization model is constructed based on the upper-level objective function and the lower-level objective function, and the charging station site selection optimization model is solved based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm; The upper-level objective function is: in, This indicates the annual revenue of the charging station operator; This represents the annual costs for charging station operators; This represents the set of origin-destination travel pairs; For a single origin-destination trip pair; This indicates the trip pair between the origin and the destination. A set of alternative routes for electric vehicle owners; Indicates in Alternative paths in; Indicates a trip pair based on origin and destination. Electric vehicles traveling along the route from the starting point to the destination Average daily total energy consumption; This indicates the price at which a charging station sells a unit of electricity; For 0-1 decision variables, if the origin-destination trip is paired... Electric vehicle owners choose For the driving route Select 1 if the value is 1, otherwise select 0. This represents the annual depreciation value of the construction cost of a charging station; This represents the annual depreciation value of the cost of purchasing charging piles for a charging station; This indicates the annual operating cost of the charging station; This indicates the cost of obtaining electrical energy for the charging station; This represents the set of alternative charging stations. Indicates alternative charging station sites. This indicates the annual depreciation factor for the charging station. This indicates the construction cost of the charging station. This represents a 0-1 decision variable, if at the alternative site Building charging stations Select 1 if the value is 1, otherwise select 0. Indicates the service life of the charging station; This indicates the purchase cost of the charging station; Indicates the usable lifespan of the charging station; This indicates the annual operating cost of the charging station; This indicates the price a charging station pays for each unit of electricity it receives. Indicates the annual depreciation factor; Represent a positive integer; The upper-level objective function includes the following decision variables and constraint functions: a. Path constraints: This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route; b. Decision variables: This constraint indicates 0-1 decision variables; This constraint indicates 0-1 decision variables; This constraint indicates It is a non-negative variable; The objective function of the lower layer is: in, Represents the shortest path among the alternative paths. distance, Indicates origin-destination travel pair Traffic on the internet This indicates the electricity consumption per unit distance traveled by an electric vehicle. This indicates the electricity price charged to electric vehicle owners before they changed their routes. Indicates the path among the alternative paths distance, This indicates the electricity price after the electric vehicle owner changes their route. The conversion factor between distance and time. Indicates cost per unit of time; The decision variables and constraint functions of the lower-level objective function are as follows: a. Electricity price constraints: This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity. b. Path constraints: This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route; This constraint indicates that in the origin-destination travel pair If the deviation distance of the deviation path exceeds twice that of the shortest alternative path, it will not be selected by electric vehicle owners. c. Decision variables: This constraint indicates 0-1 decision variables; This constraint indicates 0-1 variables; This constraint indicates It is a non-negative variable.

2. The method as described in claim 1, characterized in that, The charging station site selection optimization model is constructed based on the upper-level objective function and the lower-level objective function, and the charging station site selection optimization model is solved based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme. The preset heuristic algorithms include: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm, including: S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm; S22. Under the condition of satisfying the electric vehicle battery capacity limit, select the initial position of the electric vehicle charging station on the initialized electric vehicle path based on the iterative greedy algorithm; S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme. S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

3. The method as described in claim 1, characterized in that, S23, further optimizing the initial location of the electric vehicle charging station based on the adaptive large neighborhood search algorithm to obtain an optimized charging station site selection scheme, and outputting the current charging station site selection scheme as the optimal charging station site selection scheme when the adaptive large neighborhood search algorithm stops iterating and generates a better solution, includes: The initial location of electric vehicle charging stations is modified and optimized by introducing deletion and insertion operators into the adaptive large neighborhood search algorithm.

4. The method as described in claim 3, characterized in that, The deletion operators include: random deletion, worst-case deletion, single-point removal, and two-point removal operators; the insertion operators include: greedy insertion, k-regret value insertion, and random insertion operators.

5. A game-theoretic site selection system for electric vehicle charging stations, characterized in that, The system includes: The objective function acquisition module is used to determine the upper-level objective function based on the game strategy of the charging station operator and the lower-level objective function based on the game strategy of the electric vehicle owner. The charging station site selection scheme acquisition module is used to construct a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solve the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme; the preset heuristic algorithm includes: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm; The upper-level objective function in the objective function acquisition module is: in, This indicates the annual revenue of the charging station operator; This represents the annual costs for charging station operators; This represents the set of origin-destination travel pairs; For a single origin-destination trip pair; This indicates the trip pair between the origin and the destination. A set of alternative routes for electric vehicle owners; Indicates in Alternative paths in; Indicates a trip pair based on origin and destination. Electric vehicles traveling along the route from the starting point to the destination Average daily total energy consumption; This indicates the price at which a charging station sells a unit of electricity; For 0-1 decision variables, if the origin-destination trip is paired... Electric vehicle owners choose For the driving route Select 1 if the value is 1, otherwise select 0. This represents the annual depreciation value of the construction cost of a charging station; This represents the annual depreciation value of the cost of purchasing charging piles for a charging station; This indicates the annual operating cost of the charging station; This indicates the cost of obtaining electrical energy for the charging station; This represents the set of alternative charging stations. Indicates alternative charging station sites. This indicates the annual depreciation factor for the charging station. This indicates the construction cost of the charging station. This represents a 0-1 decision variable, if at the alternative site Building charging stations Select 1 if the value is 1, otherwise select 0. Indicates the service life of the charging station; This indicates the purchase cost of the charging station; Indicates the usable lifespan of the charging station; This indicates the annual operating cost of the charging station; This indicates the price per unit of electricity sold by the charging station; Indicates the annual depreciation factor; Represent a positive integer; The upper-level objective function includes the following decision variables and constraint functions: a. Path constraints: This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route; b. Decision variables: This constraint indicates 0-1 decision variables; This constraint indicates Represents 0-1 variables; This constraint indicates It is a non-negative variable; The objective function of the lower layer is: in, Represents the shortest path among the alternative paths. distance, Indicates origin-destination travel pair Traffic on the internet This indicates the electricity consumption per unit distance traveled by an electric vehicle. This indicates the electricity price charged to electric vehicle owners before they changed their routes. Indicates the path among the alternative paths distance, This indicates the electricity price after the electric vehicle owner changes their route. The conversion factor between distance and time. Indicates cost per unit of time; The decision variables and constraint functions of the lower-level objective function are as follows: a. Electricity price constraints: This constraint means that electric vehicle owners will only choose to change their current route if changing the current route would reduce the cost of obtaining electricity. b. Path constraints: This constraint means that electric vehicle owners must choose only one route from all alternative routes as their driving route; This constraint indicates that in the origin-destination travel pair If the deviation distance of the deviation path exceeds twice that of the shortest path among the alternative paths, it will not be selected by the electric vehicle owner. c. Decision variables: This constraint indicates 0-1 decision variables; This constraint indicates Represents 0-1 variables; This constraint indicates It is a non-negative variable.

6. The system as described in claim 5, characterized in that, The charging station site selection scheme acquisition module constructs a charging station site selection optimization model based on the upper-level objective function and the lower-level objective function, and solves the charging station site selection optimization model based on a preset heuristic algorithm to obtain the optimal electric vehicle charging station site selection scheme. The preset heuristic algorithms include: k-shortest path algorithm, iterative greedy algorithm, and adaptive large neighborhood search algorithm, including: S21. Without considering the battery capacity limitation of electric vehicles, initialize the electric vehicle path based on the k-shortest path algorithm; S22. Under the condition of satisfying the electric vehicle battery capacity limit, select the initial position of the electric vehicle charging station on the initialized electric vehicle path based on the iterative greedy algorithm; S23. Based on the adaptive large neighborhood search algorithm, further optimize the initial location of the electric vehicle charging station to obtain the optimized charging station location scheme. When the adaptive large neighborhood search algorithm stops iterating and generates a better solution, output the current charging station location scheme as the optimal charging station location scheme. S24. Further optimize the optimized charging station location scheme based on the iterative greedy algorithm, and determine whether a better solution is generated when the iterative greedy algorithm stops iterating. If a better solution is generated, return to execute S23; otherwise, output the current charging station location scheme as the optimal charging station location scheme.

7. The system as described in claim 5, characterized in that, S23, further optimizing the initial location of the electric vehicle charging station based on the adaptive large neighborhood search algorithm to obtain an optimized charging station site selection scheme, and outputting the current charging station site selection scheme as the optimal charging station site selection scheme when the adaptive large neighborhood search algorithm stops iterating and generates a better solution, includes: The initial location of electric vehicle charging stations is modified and optimized by introducing deletion and insertion operators into the adaptive large neighborhood search algorithm.

8. The system as described in claim 7, characterized in that, The deletion operators include: random deletion, worst-case deletion, single-point removal, and two-point removal operators; the insertion operators include: greedy insertion, k-regret value insertion, and random insertion operators.

Images