An electric vehicle charging station planning method and system

Abstract

The application provides an electric vehicle charging station planning method and system, acquires road network data and power distribution network topology information of a planning area, and constructs a time and space distribution model of electric vehicle distributed load; based on electric vehicle historical travel behavior and path probability distribution, an electric vehicle behavior prediction model is constructed in combination with a graph neural network to determine a candidate charging station service radius and a clustering area; a dynamic optimization model coupling energy flow and carbon flow is constructed according to distributed photovoltaic access, a dynamic carbon emission factor and vehicle charging and discharging behavior; an upper charging station site selection model is constructed in the candidate charging station node, and a multi-objective optimization algorithm based on a self-evolution Transformer structure of attention is used to determine an optimal site selection result; a lower dynamic mixed integer programming configuration model is established based on the site selection result; by jointly solving the double-layer optimization model, an optimal deployment scheme considering energy efficiency, green low carbon and supply and demand balance is output.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle charging station planning technology, and particularly relates to an electric vehicle charging station planning method and system. Background Technology

[0002] With the rapid popularization of new energy vehicles, the demand for optimized charging station layout in cities is increasing. Traditional charging station site selection methods often neglect factors such as the spatiotemporal distribution characteristics of user behavior, the absorption capacity of renewable energy, and low-carbon target control, resulting in low energy utilization efficiency, high carbon emissions, and heavy pressure on the power grid, making it difficult to meet the needs of future transportation-energy coordinated development. Therefore, there is an urgent need to propose an intelligent, multi-objective charging station planning method that integrates transportation, power grid, and green energy elements. Summary of the Invention

[0003] This invention provides a method and system for planning electric vehicle charging stations, which solves the technical problem of mismatch between electric vehicle demand and charging station layout and low economic efficiency caused by unreasonable planning and configuration.

[0004] In a first aspect, the present invention provides a method for planning electric vehicle charging stations, comprising:

[0005] Obtain road network data and power distribution network topology information for the planning area, and construct a time and space distribution model of electric vehicle distributed loads;

[0006] Based on the historical travel behavior and path probability distribution of electric vehicles, a prediction model for electric vehicle behavior is constructed by combining graph neural networks to generate the service radius and clustering area of ​​candidate charging stations.

[0007] Based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging and discharging behavior, a dynamic optimization model coupling energy flow and carbon flow is constructed. The dynamic optimization model takes electrical energy loss, carbon emission intensity, and local photovoltaic absorption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism.

[0008] An upper-level charging station location model is constructed from the candidate charging station node set, and the optimal location result is determined by the MOTformer multi-objective optimization algorithm based on the attention-based self-evolving Transformer structure.

[0009] Based on the site selection results, a lower-level dynamic mixed integer programming configuration model is established to determine the configuration quantity of different types of charging piles in each charging station, satisfying user load and network constraints.

[0010] The upper-level location model and the lower-level configuration model are jointly solved to output the optimal deployment scheme for electric vehicle charging stations.

[0011] Furthermore, the acquisition of road network data and power distribution network topology information of the planning area, and the construction of a time and space distribution model of electric vehicle distributed loads, includes the following steps:

[0012] Extract road nodes, edge information, traffic density, entrance and exit locations, and traffic flow direction within the planning area to construct an urban road map G=(V,E), where V represents the set of road intersection nodes and E represents the set of road connecting edges, with length and traffic weight attributes.

[0013] Extract basic data from the distribution network, including bus nodes, branch impedances, voltage levels, load capacity, and existing transformer capacity, and construct the network topology structure G. grid =(N,L), where N represents the set of power grid nodes and L represents the set of power grid lines;

[0014] Based on geographic information, access capacity, and the degree of electricity consumption correlation of transportation nodes, a three-layer structure of "road node - power grid node - carbon emission node" is constructed;

[0015] Based on the real-time power grid functional structure, the carbon emission factor per unit of electricity is corrected by distributed photovoltaic power output to generate a dynamic carbon emission distribution matrix.

[0016] By integrating user behavior patterns and road heatmaps, the charging demand density of different charging station nodes at time t is statistically analyzed, resulting in a load demand function based on charging station node i and time t, namely:

[0017] ,

[0018] in, Indicates will Mapped to The mapping function, Let i be the charging load of charging station node i at time t. Let i be the traffic flow at charging station node i at time t; The percentage of electric vehicles that arrive at charging station node i per unit time; This represents the average parking time. This represents the user's power demand level at the corresponding time.

[0019] Furthermore, based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging and discharging behavior, a dynamic optimization model coupling energy flow and carbon flow is constructed. The objective function of the dynamic optimization model includes:

[0020] Objective function for minimizing grid energy loss:

[0021] ,

[0022] in, This indicates the resistance of line IJ; , These are active and reactive power flows, respectively. The voltage amplitude at grid node I; A set of time periods;

[0023] Objective function for minimizing dynamic carbon emissions:

[0024] ,

[0025] in, Let i be the charging load of charging station node i at time t; This is the dynamic carbon emission correction factor corresponding to time t; A set of charging station nodes;

[0026] Objective function for maximizing local photovoltaic grid integration rate:

[0027] ,

[0028] ,

[0029] in, Let i be the charging load of charging station node i at time t; Let be the photovoltaic power generation of charging station node i at time t; Let be the photovoltaic utilization of charging station node i at time t;

[0030] Objective function for energy-carbon interaction coupling:

[0031] ,

[0032] in, An adaptive coupling weight based on the energy-carbon correlation of charging station nodes; The network loss of branch IJ at time t; Let be the carbon emission intensity of charging station node i at time t;

[0033] The final objective function is:

[0034] ,

[0035] in, and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; It is a nonlinear combination function for carbon feedback.

[0036] Furthermore, the step of constructing an upper-level charging station location model from the candidate charging station node set and using the attention-based self-evolving Transformer multi-objective optimization algorithm MOTformer to determine the optimal location result includes:

[0037] The spatial location information, data parameters, and service range of candidate charging station nodes are encoded into vector sequences and input into the Transformer encoder. An energy-carbon interaction matrix is ​​introduced into the Transformer encoder, with weights based on the energy transmission coupling degree and carbon emission intensity differences between charging station nodes. An attention mechanism is used to model the spatiotemporal energy-carbon dependence between charging station nodes, generating energy-carbon perception features. The reduction rate of grid loss, the reduction of carbon emissions, and the increase in photovoltaic utilization rate are used as feedback signals, and the model parameters and weights are updated through a reinforcement learning module. Combining the Pareto boundary solution set and the reinforcement feedback mechanism, the generated solutions are evaluated, ranked, and selected. A Pareto solution set that meets the requirements of multiple objectives is output for decision-makers to screen, and updated solutions are generated in real time under weight changes.

[0038] Furthermore, the lower-level dynamic mixed-integer programming configuration model established based on the site selection results determines the configuration quantity of different types of charging piles in each charging station, satisfying user load and network constraints, including:

[0039] Bidirectional energy flow constraints during charging and discharging at charging stations and carbon intensity limitations at charging station nodes;

[0040] The charging pile configuration includes two or more types of charging piles, each with different power levels, construction costs, and land area. The specific configuration scheme is obtained through dynamic feedback solution of a dynamic mixed integer linear programming configuration model, forming a dynamic and personalized internal structure of the site that matches the needs of the region.

[0041] In a second aspect, the present invention provides an electric vehicle charging station layout planning system, including an acquisition module, a charging station service radius and clustering region generation module, an optimization model construction module, an upper-level site selection model construction module, a lower-level configuration model construction module, and a solution module;

[0042] The acquisition module is used to acquire road network data and power distribution network topology information of the planning area, and to construct a time and space distribution model of electric vehicle distributed loads.

[0043] The charging station service radius and clustering region generation module is used to generate candidate charging station service radii and clustering regions by combining historical electric vehicle travel behavior and path probability distribution with graph neural networks to build an electric vehicle behavior prediction model.

[0044] The optimization model construction module is used to construct a dynamic optimization model that couples energy flow and carbon flow based on distributed photovoltaic access information, dynamic carbon emission factor calculation data and vehicle charging and discharging behavior. The dynamic optimization model takes electrical energy loss, carbon emission intensity and photovoltaic local consumption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism.

[0045] The upper-level location model construction module is used to construct an upper-level charging station location model from the candidate charging station node set. The optimal location result is determined by the MOTformer multi-objective optimization algorithm based on the attention-based self-evolving Transformer structure.

[0046] The lower-level configuration model construction module is used to build a lower-level dynamic mixed integer programming configuration model based on the site selection results, determine the configuration quantity of different types of charging piles in each charging station, and meet the user load and network constraints.

[0047] The solution module is used to jointly solve the upper-level site selection model and the lower-level configuration model, and output the optimal deployment scheme of electric vehicle charging stations that takes into account energy efficiency, green and low-carbon features and supply and demand matching.

[0048] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the electric vehicle charging station planning method described above.

[0049] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described electric vehicle charging station planning method.

[0050] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0051] This application discloses an electric vehicle charging station planning method and system. It acquires road network data and power distribution network topology information of the planning area to construct a temporal and spatial distribution model of electric vehicle distributed loads. Based on historical travel behavior and path probability distribution of electric vehicles, it combines graph neural networks to construct an electric vehicle behavior prediction model, generating the service radius and clustering area of ​​candidate charging stations. According to distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging behavior, it constructs a dynamic optimization model coupling energy flow and carbon flow. An upper-level charging station site selection model is built from the candidate charging station node set. The optimal site selection result is determined using the attention-based self-evolving Transformer multi-objective optimization algorithm MOTformer. Based on the site selection result, a lower-level dynamic mixed-integer programming configuration model is established. By jointly solving the two-level programming configuration model, a charging station planning configuration scheme is obtained, improving the solution accuracy and speed, increasing photovoltaic absorption rate, and enhancing the utilization rate of charging facilities, thereby achieving carbon reduction and efficiency improvement. This invention can effectively improve grid operating efficiency and photovoltaic absorption rate, reduce carbon emissions, and achieve synergistic optimization of the transportation-energy system. It has high adaptability, high reliability, and high energy efficiency characteristics and can be widely applied in urban smart transportation and green energy infrastructure layout scenarios. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0053] Figure 1 A flowchart of an electric vehicle charging station planning method provided in an embodiment of the present invention;

[0054] Figure 2 A flowchart of a multi-objective optimization algorithm based on the Transformer structure provided in an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0057] Please see Figure 1 As shown, an embodiment of the present invention provides a method for planning electric vehicle charging stations, specifically including the following steps:

[0058] Step S101: Obtain road network data and power distribution network topology information of the planning area, and construct a time and space distribution model of electric vehicle distributed loads;

[0059] By using traffic management platforms, GIS map data, or high-precision route planning systems, extract elements such as road nodes, edge information, traffic density, entrance and exit locations, and traffic flow direction within the planning area, and construct an urban road map G=(V,E), where V represents the set of road intersection nodes and E represents the set of road connecting edges, with attributes such as length and traffic weight.

[0060] Extract basic distribution network data from the power grid dispatching system or distribution automation system, including bus nodes, branch impedance, voltage level, load capacity, and existing transformer capacity, to construct the power grid topology structure G. grid =(N,L), where N represents the set of power grid nodes, representing the connection points and load points in the power grid, and L represents the set of power grid lines, representing the power lines connecting these nodes;

[0061] Based on geographic information, access capacity, and the degree of electricity consumption correlation of transportation nodes, a three-layer structure of "road node - power grid node - carbon emission node" is constructed;

[0062] Based on the real-time power grid functional structure, the carbon emission factor per unit of electricity is corrected by distributed photovoltaic power output to generate a dynamic carbon emission distribution matrix.

[0063] By integrating user behavior patterns and road heatmaps, the charging demand density of different charging station nodes at time t is statistically analyzed, resulting in a load demand function based on charging station node i and time t, namely:

[0064] ,

[0065] in, Indicates will Mapped to The mapping function, Let i be the charging load of charging station node i at time t. Let i be the traffic flow at charging station node i at time t; The percentage of electric vehicles that arrive at charging station node i per unit time; This represents the average parking time. This corresponds to the user's power demand level at that moment;

[0066] Output the spatiotemporal load distribution of each charging station node in the form of a heat map or matrix, which serves as the basic input data for the subsequent charging station site selection-capacity model.

[0067] Step S102: Based on the historical travel behavior and path probability distribution of electric vehicles, a prediction model for electric vehicle behavior is constructed by combining graph neural networks to generate the service radius and clustering area of ​​candidate charging stations.

[0068] Historical travel data of electric vehicles in the target area is collected, including historical charging demand, dwell time, and regional functional type (such as commercial area, residential area), and mapped to road network nodes. Feature analysis is performed using road topology and traffic flow to calculate the probability distribution of electric vehicle path transitions. A spatiotemporal graph neural network model is constructed to aggregate road node information in spatiotemporal dimensions, learn the dynamic distribution of travel demand and path selection probability, and predict the charging demand density of each road node at different future time periods. The model simulates the travel chains of a large number of electric vehicles and their charging decision-making behavior under low battery conditions, generating a high-precision spatial heat map of charging demand. Finally, a spatial clustering algorithm is used to identify candidate areas with high demand, and an acceptable service radius threshold is derived based on user behavior data. The final output is a cluster of candidate charging stations and their service range that combines spatial coverage and demand concentration, providing a quantitative basis for subsequent site selection and capacity optimization.

[0069] Step S103: Based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging behavior, a dynamic optimization model coupling energy flow and carbon flow is constructed. This dynamic optimization model takes energy loss, carbon emission intensity, and local photovoltaic absorption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism. The objective function of the dynamic optimization model includes:

[0070] Objective function for minimizing grid energy loss:

[0071] ,

[0072] in, This indicates the resistance of line IJ; , These are active and reactive power flows, respectively. The voltage amplitude at grid node I; A set of time periods;

[0073] Objective function for minimizing dynamic carbon emissions:

[0074] ,

[0075] in, Let i be the charging load of charging station node i at time t; This is the dynamic carbon emission correction factor corresponding to time t; A set of charging station nodes;

[0076] Objective function for maximizing local photovoltaic grid integration rate:

[0077] ,

[0078] ,

[0079] in, Let i be the charging load of charging station node i at time t; Let be the photovoltaic power generation of charging station node i at time t; Let be the photovoltaic utilization of charging station node i at time t;

[0080] Objective function for energy-carbon interaction coupling:

[0081] ,

[0082] in, An adaptive coupling weight based on the energy-carbon correlation of charging station nodes; The network loss of branch IJ at time t; Let represent the carbon emission intensity of charging station node i at time t.

[0083] Normalizing the above objective function yields the final objective function as follows:

[0084] ,

[0085] in, and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; The energy-carbon feedback nonlinear combination function is updated adaptively through reinforcement learning to achieve a dynamic balance between energy flow and carbon flow in the spatiotemporal dimensions.

[0086] Step S104: Construct an upper-level charging station location model from the candidate charging station node set, and use the attention-based self-evolving Transformer structure multi-objective optimization algorithm MOTformer to determine the optimal location result.

[0087] Please see Figure 2 A multi-objective optimization algorithm based on the Transformer structure includes the following steps;

[0088] The spatial location information, data parameters, and service range of candidate charging station nodes are encoded into vector sequences and input into the Transformer encoder. An energy-carbon interaction matrix is ​​introduced into the Transformer encoder, with weights based on the energy transmission coupling degree and carbon emission intensity differences between charging station nodes. An attention mechanism is used to model the spatiotemporal energy-carbon dependence between charging station nodes, generating energy-carbon perception features. The reduction rate of grid loss, the reduction in carbon emissions, and the increase in photovoltaic utilization rate are used as feedback signals, and the model parameters and weights are updated self-evolvingly through a reinforcement learning module. Combining the Pareto boundary solution set and the reinforcement feedback mechanism, the generated solutions are evaluated, ranked, and selected. A Pareto solution set that meets the multi-objective requirements is output for decision-makers to filter, and updated solutions can be generated in real time under weight changes.

[0089] Step S105: Based on the site selection results, establish a lower-level dynamic mixed integer programming configuration model to determine the configuration quantity of different types of charging piles in each charging station, satisfying user load and network constraints.

[0090] Bidirectional energy flow constraints during charging and discharging at charging stations and carbon intensity limitations at charging station nodes;

[0091] The charging pile configuration includes two or more types of charging piles, each with different power levels, construction costs and land area. The specific configuration scheme is obtained through dynamic feedback solution of dynamic mixed integer linear programming configuration model, forming a dynamic and personalized internal structure of the site that matches the needs of the region.

[0092] Step S106: Jointly solve the upper-level site selection model and the lower-level configuration model to output the optimal deployment scheme for electric vehicle charging stations that takes into account energy efficiency, green and low-carbon features, and supply and demand balance.

[0093] In summary, the method of this application acquires road network data and power distribution network topology information of the planning area to construct a temporal and spatial distribution model of electric vehicle distributed loads; based on the historical travel behavior and path probability distribution of electric vehicles, a graph neural network is used to construct an electric vehicle behavior prediction model to generate the service radius and clustering area of ​​candidate charging stations; based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging behavior, a dynamic optimization model coupling energy flow and carbon flow is constructed; an upper-level charging station site selection model is constructed in the candidate charging station node set, and the optimal site selection result is determined using the attention-based self-evolving Transformer multi-objective optimization algorithm MOTformer; based on the site selection result, a lower-level dynamic mixed integer programming configuration model is established; by jointly solving the two-level programming configuration model, a charging station planning and configuration scheme is obtained, which improves the solution accuracy and speed, can improve the photovoltaic absorption rate, and improve the utilization rate of charging facilities, thereby achieving carbon reduction and efficiency improvement.

[0094] Another embodiment of the present invention provides an electric vehicle charging station planning system, including an acquisition module, a charging station service radius and clustering region generation module, an optimization model construction module, an upper-level site selection model construction module, a lower-level configuration model construction module, and a solution module;

[0095] The acquisition module is used to acquire road network data and power distribution network topology information of the planning area, and to construct a time and space distribution model of electric vehicle distributed loads.

[0096] The charging station service radius and clustering region generation module is used to generate candidate charging station service radii and clustering regions by combining historical electric vehicle travel behavior and path probability distribution with graph neural networks to build an electric vehicle behavior prediction model.

[0097] The optimization model construction module is used to construct a dynamic optimization model that couples energy flow and carbon flow based on distributed photovoltaic access information, dynamic carbon emission factor calculation data and vehicle charging and discharging behavior. The dynamic optimization model takes electrical energy loss, carbon emission intensity and photovoltaic local consumption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism.

[0098] The upper-level location model construction module is used to construct an upper-level charging station location model from the candidate charging station node set. The optimal location result is determined by the MOTformer multi-objective optimization algorithm based on the attention-based self-evolving Transformer structure.

[0099] The lower-level configuration model construction module is used to build a lower-level dynamic mixed integer programming configuration model based on the site selection results, determine the configuration quantity of different types of charging piles in each charging station, and meet the user load and network constraints.

[0100] The solution module is used to jointly solve the upper-level location model and the lower-level configuration model, and output the optimal deployment scheme for electric vehicle charging stations.

[0101] It should be understood that the modules recorded in the electric vehicle charging station planning system correspond to the various steps in the above method embodiments. Therefore, the operations, features, and corresponding technical effects described above for the method are also applicable to the modules in the electric vehicle charging station planning system, and will not be repeated here.

[0102] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, causes the processor to perform the electric vehicle charging station planning method in the above method embodiments.

[0103] In one implementation, the computer-readable storage medium may include a stored program area and a stored data area, wherein the stored program area may store an operating system and an application program required for at least one function; the stored data area may store data created based on the use of the electric vehicle charging station planning system, etc. Furthermore, the computer-readable storage medium may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include memory remotely located relative to the processor, and this remote memory may be connected to the electric vehicle charging station planning system via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0104] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present invention, such as... Figure 3 As shown, the device includes a processor 310 and a memory 320. The electronic device may also include an input device 330 and an output device 340. The processor 310, memory 320, input device 330, and output device 340 can be connected via a bus or other means. Figure 3 Taking a bus connection as an example, the memory 320 is the computer-readable storage medium described above. The processor 310 executes various server functions and data processing by running non-volatile software programs, instructions, and modules stored in the memory 320, thereby implementing the electric vehicle charging station planning method described in the above embodiment. The input device 330 can receive input digital or character information and generate key signal inputs related to user settings and function control of the electric vehicle charging station planning system. The output device 340 may include a display screen or other display device.

[0105] The aforementioned electronic device can execute the method provided in the embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the method. Technical details not described in detail in this embodiment can be found in the method provided in the embodiments of the present invention.

[0106] In one embodiment, the above-described electronic device is applied to an electric vehicle charging station planning system for a client, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute the instructions stored in a computer-readable storage medium.

[0107] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0108] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; 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; and these 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 method for planning electric vehicle charging stations, characterized in that, include: Obtain road network data and power distribution network topology information for the planning area, and construct a time and space distribution model of electric vehicle distributed loads; Based on the historical travel behavior and path probability distribution of electric vehicles, a prediction model for electric vehicle behavior is constructed by combining graph neural networks to generate the service radius and clustering area of ​​candidate charging stations. Based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging and discharging behavior, a dynamic optimization model coupling energy flow and carbon flow is constructed. The dynamic optimization model takes electrical energy loss, carbon emission intensity, and local photovoltaic absorption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism. An upper-level charging station location model is constructed from the candidate charging station node set, and the optimal location result is determined by the MOTformer multi-objective optimization algorithm based on the attention-based self-evolving Transformer structure. Based on the site selection results, a lower-level dynamic mixed integer programming configuration model is established to determine the configuration quantity of different types of charging piles in each charging station, satisfying user load and network constraints. The upper-level site selection model and the lower-level configuration model are jointly solved to output the optimal deployment scheme for electric vehicle charging stations. The method involves constructing a dynamic optimization model that couples energy flow and carbon flow based on distributed photovoltaic access information, dynamic carbon emission factor calculation data, and vehicle charging and discharging behavior. The objective function of the dynamic optimization model includes: Objective function for minimizing grid energy loss: ， in, This indicates the resistance of line IJ; , These are active and reactive power flows, respectively. The voltage amplitude at grid node I; A set of time periods; Objective function for minimizing dynamic carbon emissions: ， in, Let i be the charging load of charging station node i at time t; This is the dynamic carbon emission correction factor corresponding to time t; A set of charging station nodes; Objective function for maximizing local photovoltaic grid integration rate: ， ， in, Let i be the charging load of charging station node i at time t; Let be the photovoltaic power generation of charging station node i at time t; Let be the photovoltaic utilization of charging station node i at time t; Objective function for energy-carbon interaction coupling: , in, An adaptive coupling weight based on the energy-carbon correlation of charging station nodes; The network loss of branch IJ at time t; Let be the carbon emission intensity of charging station node i at time t; The final objective function is: ， in, and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; and They are respectively The minimum and maximum values; It is a nonlinear combination function for carbon feedback.

2. The electric vehicle charging station planning method according to claim 1, characterized in that, The process of acquiring road network data and power distribution network topology information for the planning area, and constructing a temporal and spatial distribution model of electric vehicle distributed loads, includes the following steps: Extract road nodes, edge information, traffic density, entrance and exit locations, and traffic flow direction within the planning area to construct an urban road map G=(V,E), where V represents the set of road intersection nodes and E represents the set of road connecting edges, with length and traffic weight attributes. Extract basic data of the distribution network, including bus nodes, branch impedance, voltage level, load capacity, and existing transformer capacity, and construct a power grid topology structure Ggrid=(N,L), where N represents the set of power grid nodes and L represents the set of power grid lines. Based on geographic information, access capacity, and the degree of electricity consumption correlation of transportation nodes, a three-layer structure of "road node - power grid node - carbon emission node" is constructed; Based on the real-time power grid functional structure, the carbon emission factor per unit of electricity is corrected by distributed photovoltaic power output to generate a dynamic carbon emission distribution matrix. By integrating user behavior patterns and road heatmaps, the charging demand density of different charging station nodes at time t is statistically analyzed, resulting in a load demand function based on charging station node i and time t, namely: ， in, Indicates will Mapped to The mapping function, Let i be the charging load of charging station node i at time t. Let i be the traffic flow at charging station node i at time t; The percentage of electric vehicles that arrive at charging station node i per unit time; This represents the average parking time. This represents the user's power demand level at the corresponding time.

3. The electric vehicle charging station planning method according to claim 1, characterized in that, The step of constructing an upper-level charging station location model from the candidate charging station node set and using the attention-based self-evolving Transformer multi-objective optimization algorithm MOTformer to determine the optimal location result includes: The spatial location information, data parameters, and service range of candidate charging station nodes are encoded into vector sequences and input into the Transformer encoder. An energy-carbon interaction matrix is ​​introduced into the Transformer encoder, with weights based on the energy transmission coupling degree and carbon emission intensity differences between charging station nodes. An attention mechanism is used to model the spatiotemporal energy-carbon dependence between charging station nodes, generating energy-carbon perception features. The reduction rate of grid loss, the reduction of carbon emissions, and the increase in photovoltaic utilization rate are used as feedback signals, and the model parameters and weights are updated through a reinforcement learning module. Combining the Pareto boundary solution set and the reinforcement feedback mechanism, the generated solutions are evaluated, ranked, and selected. A Pareto solution set that meets the requirements of multiple objectives is output for decision-makers to screen, and updated solutions are generated in real time under weight changes.

4. The electric vehicle charging station planning method according to claim 3, characterized in that, The lower-level dynamic mixed-integer programming configuration model established based on the site selection results determines the configuration quantity of different types of charging piles in each charging station, satisfying user load and network constraints, including: Bidirectional energy flow constraints during charging and discharging at charging stations and carbon intensity limitations at charging station nodes; The charging pile configuration includes two or more types of charging piles, each with different power levels, construction costs, and land area. The specific configuration scheme is obtained through dynamic feedback solution of a dynamic mixed integer linear programming configuration model, forming a dynamic and personalized internal structure of the site that matches the needs of the region.

5. An electric vehicle charging station planning system, used to implement the planning method according to any one of claims 1-4, characterized in that, It includes an acquisition module, a charging station service radius and clustering region generation module, an optimization model construction module, an upper-layer site selection model construction module, a lower-layer configuration model construction module, and a solution module; The acquisition module is used to acquire road network data and power distribution network topology information of the planning area, and to construct a time and space distribution model of electric vehicle distributed loads. The charging station service radius and clustering region generation module is used to generate candidate charging station service radii and clustering regions by combining historical electric vehicle travel behavior and path probability distribution with graph neural networks to build an electric vehicle behavior prediction model. The optimization model construction module is used to construct a dynamic optimization model that couples energy flow and carbon flow based on distributed photovoltaic access information, dynamic carbon emission factor calculation data and vehicle charging and discharging behavior. The dynamic optimization model takes electrical energy loss, carbon emission intensity and photovoltaic local consumption rate as collaborative objectives, and achieves spatiotemporal dynamic optimization through an energy-carbon weight adaptive mechanism. The upper-level location model construction module is used to construct an upper-level charging station location model from the candidate charging station node set. The optimal location result is determined by the MOTformer multi-objective optimization algorithm based on the attention-based self-evolving Transformer structure. The lower-level configuration model construction module is used to build a lower-level dynamic mixed integer programming configuration model based on the site selection results, determine the configuration quantity of different types of charging piles in each charging station, and meet the user load and network constraints. The solution module is used to jointly solve the upper-level location model and the lower-level configuration model, and output the optimal deployment scheme for electric vehicle charging stations.

6. An electronic device, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the planning method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the planning method according to any one of claims 1 to 4.

Images