Dynamic charging time planning method, system and device considering first arrival first service constraint, and storage medium

Abstract

The invention discloses a dynamic charging time planning method, system and device considering a first arrival first service constraint, and a storage medium, belongs to the crossing field of an intelligent traffic system and an energy internet, and solves the problem of space-time dynamic distribution of charging resources in a large-scale road network. The method comprises the following steps: setting a regional road network range and extracting a topological structure thereof; a state planning space including time, position and electric quantity is constructed, and vehicles in a road network are divided into three types: vehicles without charging requirements, non-reservation user new energy vehicles and reservation user new energy vehicles; the proportion of the reserved charging piles set in the service area is matched with the proportion of the reserved charging vehicles in the total number of the new energy vehicles; a coupled HJB-FPK equation system is established; the reservation density is listed into an FPK equation, and it is determined that the constraint condition of the FCFS service is established in each period; a rolling optimization mechanism is adopted to realize the minimum total cost of the vehicle arriving at the destination under the FCFS constraint, and the charging problem of the new energy vehicle in the road network travel peak period is effectively solved.

Description

Technical Field

[0001] This invention relates to the intersection of intelligent transportation systems and the energy internet, and more specifically, to a dynamic charging time planning method, system, device, and storage medium that takes into account first-come, first-served constraints. Background Technology

[0002] With the rapid increase in the popularity of new energy vehicles, the charging demand on highway networks has exploded during peak holiday travel periods. During long holidays such as Spring Festival and National Day, it's common to see long queues at charging stations in popular service areas, often resulting in "one hour of charging followed by three hours of waiting." This not only severely impacts the travel experience but also becomes an obstacle for the public to choose new energy vehicles for long-distance travel. Unlike urban charging scenarios, highway charging demand exhibits tidal characteristics, making it difficult to resolve the supply-demand imbalance: daily usage is too low, resulting in low returns on investment for simply increasing charging stations; without any measures, charging resources are significantly scarce during peak holiday periods. This scarcity of charging resources exacerbates the disorderly state of highway services, with frequent instances of "scrambling for charging stations" across various regions. This not only reduces user satisfaction but also decreases road network efficiency and wastes energy resources. Of particular note is the "charging anxiety" experienced by users when they are stranded for extended periods to charge, which, in the long run, will become a significant factor hindering the further popularization of new energy vehicles. As the national transportation artery, the orderly and efficient operation of charging services on highways has become a crucial element in improving the public's sense of gain and happiness during travel.

[0003] However, building a scientifically sound and reasonable charging reservation service system for new energy vehicles on the go is no easy task. Allocating charging opportunities to vehicles in motion is essentially a highly complex dynamic traffic flow allocation task involving the dynamic coupling of multiple dimensions, including vehicle route selection, energy consumption, and charging demand. Traditional methods either rely on historical data statistical models, which cannot capture the interaction between real-time traffic flow and charging behavior; or they employ micro-level traffic simulations, which, while able to finely depict individual behavior, are computationally complex and difficult to apply to large-scale road networks. A deeper challenge lies in the need to balance multiple objectives in charging service planning: meeting users' core demand for rapid passage while rationally utilizing limited charging resources; considering current traffic conditions while predicting future changes in charging demand. Especially during peak travel periods like holidays, the high randomness of user destinations and the diversity of travel routes make it difficult for static planning models based on historical data to accurately predict demand, significantly reducing the reliability of reservation services. How to build an efficient and reliable charging reservation system in a complex and dynamic environment has become a pressing technical challenge for the industry. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the above-mentioned shortcomings of the prior art. The purpose of the present invention is to provide a dynamic charging time planning method that takes into account first-come, first-served constraints.

[0005] The second objective of this invention is to provide a dynamic charging time planning system that takes into account first-come, first-served constraints.

[0006] The third objective of this invention is to provide a computer device.

[0007] The fourth objective of this invention is to provide a computer storage medium.

[0008] To achieve the first objective mentioned above, this invention provides a dynamic charging time planning method that considers first-come, first-served constraints, comprising the following steps: Step S1: Define the area road network scope and extract its topology; Step S2: Build a time-included structure ,Location Battery The system plans the state of the space and divides vehicles into three categories: the first category is non-charging vehicles, the second category is new energy vehicles with no demand data or that do not accept reservation services, and the third category is new energy vehicles of users with clear demand and reservations. The data on the proportion of the three categories of vehicles is obtained through pre-survey, entrance ramp statistics or travel reservation system. Each service area is equipped with a certain number of charging piles as service reserved charging piles. The proportion of service reserved charging piles is matched with the proportion of the third category of vehicles in the total number of the second and third categories of vehicles. Step S3: Establish a coupled HJB-FPK equation system, where the HJB equation describes the optimal charging strategy for the three types of vehicles, and the charging strategy for the third type of vehicle is "charge until full and go", and the FPK equation describes the density evolution of the three types of vehicles. Step S4: Confirm the reservation density As a source term in the FPK equation, it satisfies the FPK equation to effectively describe the process of density evolution for the three types of vehicles, i.e. ; Step S5: Confirm equivalent constraints This ensures that new optimization decisions do not affect confirmed reservations, thereby ensuring that the business constraints of FCFS are met in every cycle; Step S6: Employ a rolling optimization mechanism and set initial conditions. While maintaining the computational efficiency advantage of MFG, it minimizes the total cost for vehicles to reach their destination under FCFS constraints.

[0009] As a further improvement, the contributions of the first type of vehicles, the second type of vehicles, and the third type of vehicles are expressed as follows:

[0010] In the formula, : is a time variable; For spatial location variables, For the set of road network nodes; For battery status variables, It is the first Type of vehicle in time ,Location Battery The density.

[0011] Furthermore, the process by which the HJB equation describes the optimal charging strategy for each type of vehicle is as follows:

[0012] In the formula, For road section The travel time Traffic flow on the road segment The flow-time sensitivity coefficient is the linearized result of the road segment impedance (such as BPR) model. For the first The dynamic function of the battery level of vehicles, i.e.

[0013] in For path decision, For charging volume decisions, Power consumption rate For charging efficiency; For the first Vehicles in location The waiting time, among which, At that time, gasoline vehicles have no need for charging, and the waiting time at charging stations is negligible; only the time lost passing through service areas needs to be considered. ; When considering the waiting time of non-reserved new energy vehicles, that is...

[0014] The aforementioned waiting time can also be replaced by a queuing theory model, which more accurately reflects the actual waiting time of the second type of vehicle; When making an appointment, the waiting time of the user should be taken into account.

[0015] In the formula It is the current position Is it a service area type? The judgment is made such that if the condition is met, then 1 is selected; otherwise, 0 is selected. The time lost by vehicles entering and leaving the service area; The length of the queue for reserved vehicles. Queue length for vehicles without reservations; To reserve the number of charging stations, This refers to the number of charging piles not reserved for future use. Service area service rate.

[0016] Furthermore, the FPK equation describes the spatiotemporal evolution of the density of the three types of vehicles as follows:

[0017] In the formula, For the first The density function of the confirmed reserved vehicles. The time-varying rate of change in the density of confirmed reservation vehicles.

[0018] Furthermore, based on the vehicle density field, the traffic flow and travel time of road segments within the aforementioned regional road network are calculated to establish a correlation between microscopic vehicle behavior and macroscopic road network status, thereby reflecting the dynamic characteristics of traffic congestion and realizing bidirectional feedback between the microscopic and macroscopic levels of the charging dynamic model. The calculations are as follows:

[0019]

[0020]

[0021] In the formula, For position Total vehicle density; For road section Traffic, For road section Free-flow travel time, This is the flow-time sensitivity coefficient.

[0022] Furthermore, charging demand information is extracted from the vehicle density field. This provides a basis for the planning and scheduling of charging facilities, namely

[0023]

[0024]

[0025] In the formula, The safe power threshold; For the first Battery safety threshold for similar vehicle types; satisfying the relationship Since the user's vehicle charging strategy is "charge until full and then leave," it generally meets the requirements. .

[0026] Furthermore, in step S6, the optimization window length of the rolling optimization mechanism is set to... =15-30 minutes, optimize frequency set to =5-10 minutes; at the start of each optimization round, the initial conditions of the road network are set as follows:

[0027]

[0028] In the formula, For the density distribution of newly entering vehicles, For the first The density function for confirmed reservation vehicles; For Dirac delta function; For vehicles In time Location, For vehicles In time The amount of electricity; Indicates the current time A collection of previously confirmed appointment information; To ensure the FCFS constraint is met, the following condition is verified after each round of optimization.

[0029] In the formula For the first The value function for a vehicle class represents the value of a vehicle from state 1. The minimum total cost required to travel from departure to destination.

[0030] To achieve the second objective mentioned above, the present invention provides a dynamic charging time planning system that considers first-come, first-served constraints, comprising: The acquisition module is used to define the regional road network range and extract its topology. Build modules are used to build time-based systems. ,Location Battery The system plans the state of the space and divides vehicles into three categories: the first category is non-charging vehicles, the second category is new energy vehicles with no demand data or that do not accept reservation services, and the third category is new energy vehicles of users with clear demand and reservations. The data on the proportion of the three categories of vehicles is obtained through pre-survey, entrance ramp statistics or travel reservation system. Each service area is equipped with a certain number of charging piles as service reserved charging piles. The proportion of service reserved charging piles is matched with the proportion of the third category of vehicles in the total number of the second and third categories of vehicles. The computation module is used to establish a coupled HJB-FPK equation system, where the HJB equations describe the optimal charging strategy for the three types of vehicles, and the FPK equations describe the density evolution of the three types of vehicles; and to process the confirmed reservation density. As a source term of the FPK equation, it satisfies ;confirm This ensures that new optimization decisions do not affect confirmed reservations, thereby ensuring that the business constraints of FCFS are met in every cycle; a rolling optimization mechanism is adopted, and initial conditions are set. While maintaining the computational efficiency advantage of MFG, it minimizes the total cost for vehicles to reach their destination under FCFS constraints.

[0031] To achieve the above-mentioned objective three, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned dynamic charging time planning method that considers first-come-first-served constraints.

[0032] To achieve the fourth objective mentioned above, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the above-mentioned dynamic charging time planning method that takes into account first-come-first-served constraints.

[0033] Compared with the prior art, the advantages of this invention are as follows: This invention innovatively restructures the target setting of charging services, taking "charge and go" rather than "fully charge as quickly as possible" as the core service concept. This fundamentally aligns with users' real needs—quickly leaving the local road network, with charging within the service area merely a means to ensure departure. This shift in concept not only avoids resource waste but also significantly improves the turnover efficiency of charging piles, enabling limited charging resources to serve more users. By scientifically classifying user behavior into three categories—vehicles with no charging needs, new energy vehicles with unclear or non-reserved needs, and new energy vehicles with reserved needs—this invention maintains the simplicity of the model while accurately capturing the behavioral characteristics of different user groups, providing a theoretical basis for accurately predicting charging demand.

[0034] By leveraging the Mean Field Game (MFG) framework, this invention transforms the complex, high-dimensional multi-group decision-making problem into a system of efficiently solvable partial differential equations, achieving a breakthrough from theoretical infeasibility to practical applicability. The MFG model significantly reduces computational complexity by aggregating the behaviors of individual vehicles into density functions, enabling the system to achieve sub-second response times on ordinary computing devices, providing technical support for real-time optimization of charging reservation services on large-scale highway networks. Crucially, this invention, through rigorous mathematical design, embeds the core business constraint of "established charging arrangements cannot be changed" within the MFG framework, fundamentally guaranteeing the first-come, first-served principle. This innovation resolves the fundamental conflict between MFG global optimization and the FCFS business constraint, providing users with a stable and reliable reservation experience and significantly enhancing user trust and satisfaction with the charging reservation service.

[0035] In summary, this invention, through innovative concept of objectives, scientific classification of user behavior, ingenious application of the MFG model, and mathematical guarantee of FCFS constraints, constructs a charging reservation system that not only meets real user needs but also possesses engineering feasibility. It provides a systematic solution to the charging problem during peak travel periods on holidays and is of great significance for promoting the popularization of new energy vehicles and improving the quality of highway charging services. Attached Figure Description

[0036] Figure 1 This is a flowchart of the present invention; Figure 2 This is a system architecture diagram of the present invention. Detailed Implementation

[0037] Terminology Explanation FCFS (First-Come-First-Serve) is the core business constraint of the road network charging reservation system in this invention. Once a user with needs enters the road network, the reservation will be responded to and arrangements will be made. It is clear that "the reservation arrangements that have been confirmed cannot be changed", that is, the reservation arrangements of the next batch of users cannot affect the confirmed trips of the previous batch of users.

[0038] Mean-Field Game (MFG) is a mathematical framework for handling large-scale agent interactions and plays a central role in this invention. It approximates and simplifies traditional multi-agent games into a game between individual and group average behavior (density field). By solving the Hamilton-Jacobi-Bellman (HJB) and Fokker-Planck-Kolmogorov (FPK) coupled equation system, it reduces the complex high-dimensional nonlinear mixed-integer programming problem to an efficiently solvable partial differential equation problem.

[0039] The HJB (Hamilton-Jacobi-Bellman) equation is the core partial differential equation in the MFG framework describing the optimal control strategy of an individual vehicle. In the road network charging reservation system, it represents the optimal decision-making process of a vehicle. Its key innovation lies in distinguishing three types of vehicles and introducing a confirmed reservation density. As input, ensure the wait time function The system accurately reflects the reservation status, enabling it to generate the optimal charging strategy within milliseconds while strictly satisfying the FCFS constraint, thus providing a theoretical guarantee and computational feasibility for large-scale road network charging reservations.

[0040] The FPK (Fokker-Planck-Kolmogorov) equation describes the dynamic process of population density evolution over time and characterizes the vehicle density field in a road network charging reservation system. The equation, along with the HJB equation, forms a closed-loop coupled system that can accurately capture the density evolution of the three types of vehicles and the road network state feedback. This innovation in the FPK equation resolves the fundamental mathematical conflict in the operational constraints of FCFS.

[0041] Basic Implementation The present invention will be further described below with reference to specific embodiments shown in the accompanying drawings.

[0042] See Figures 1-2 A dynamic charging time planning method that considers first-come, first-served constraints includes the following steps: Step S1: Define the area road network scope and extract its topology; Step S2: Build a time-included structure ,Location Battery The system plans the state of the space and divides vehicles into three categories: the first category is non-charging vehicles, the second category is new energy vehicles with no demand data or that do not accept reservation services, and the third category is new energy vehicles of users with clear demand and reservations. The data on the proportion of the three categories of vehicles is obtained through pre-survey, entrance ramp statistics or travel reservation system. Each service area is equipped with a certain number of charging piles as service reserved charging piles. The proportion of service reserved charging piles is matched with the proportion of the third category of vehicles in the total number of the second and third categories of vehicles. Step S3: Establish a coupled HJB-FPK equation system, where the HJB equation describes the optimal charging strategy for the three types of vehicles, and the charging strategy for the third type of vehicle is "charge until full and go", and the FPK equation describes the density evolution of the three types of vehicles. Step S4: Confirm the reservation density As a source term in the FPK equation, it satisfies the FPK equation to effectively describe the process of density evolution for the three types of vehicles, i.e. ; Step S5: Confirm equivalent constraints This ensures that new optimization decisions do not affect confirmed reservations, thereby ensuring that the business constraints of FCFS are met in every cycle; Step S6: Employ a rolling optimization mechanism and set initial conditions. While maintaining the computational efficiency advantage of MFG, it minimizes the total cost for vehicles to reach their destination under FCFS constraints.

[0043] In step S2, for the road network, the contributions of Class I vehicles, Class II vehicles, and Class III vehicles are expressed as follows:

[0044] In the formula, : is a time variable; For spatial location variables, For the set of road network nodes; This is a power state variable (normalized, 1 represents full charge). It is the first Type of vehicle in time ,Location Battery The density is specifically defined as: k=1: Fuel-powered vehicles do not have the possibility of occupying any charging stations; k=2: New energy vehicles with no demand data or that do not accept reservation services; k=3: New energy vehicles of users with clear real-time needs and who accept pre-booking services.

[0045] This classification accurately captures the behavioral characteristics of different groups and forms the basis for solving the problem of randomness in holiday travel. In its long-term evolution, there has been a relationship of mutual transformation between the first and second categories of vehicles, and between the second and third categories.

[0046] In step S3, following the approach of the MFG model, this invention describes the optimal charging strategy for each type of vehicle based on the HJB equation. In particular, it needs to accurately reflect the impact of confirmed reservations on waiting time. The process of describing the optimal charging strategy for each type of vehicle using the HJB equation is as follows:

[0047] In the formula, For road section The travel time Traffic flow on the road segment The flow-time sensitivity coefficient is the linearized result of the road segment impedance (such as BPR) model. For the first The dynamic function of the battery level of vehicles, i.e.

[0048] in For path decision, For charging volume decisions, Power consumption rate For charging efficiency; For the first Vehicles in location The waiting time, among which, At that time, gasoline vehicles have no need for charging, and the waiting time at charging stations is negligible; only the time lost passing through service areas needs to be considered. ; When considering the waiting time of non-reserved new energy vehicles, that is...

[0049] The aforementioned waiting time can also be replaced by a queuing theory model, which more accurately reflects the actual waiting time of the second type of vehicle; When making an appointment, the waiting time of the user should be taken into account.

[0050] In the formula It is the current position Is it a service area type? The judgment is made such that if the condition is met, then 1 is selected; otherwise, 0 is selected. The time lost by vehicles entering and leaving the service area; The length of the queue for reserved vehicles. Queue length for vehicles without reservations; To reserve the number of charging stations, This refers to the number of charging piles not reserved for future use. Service area service rate.

[0051] Through the reverse engineering process, the HJB equations naturally take into account the future road network conditions. Simultaneously, the confirmed reservation density is incorporated into the waiting time function in the above equation. As a key parameter, the waiting time calculation for reserved vehicles takes into account confirmed reservations, which is the mathematical basis for ensuring the FCFS constraint.

[0052] This invention describes the spatiotemporal evolution of three types of vehicle density using the FPK equation, while ensuring that confirmed reservations are not affected by subsequent optimizations, fundamentally resolving the conflict between the MFG model and FCFS constraints. Specifically, the FPK equation describes the spatiotemporal evolution of the three types of vehicle density as follows:

[0053] In the formula, For the first The density function of the confirmed reserved vehicles. The time-varying rate of change in the density of confirmed reservation vehicles.

[0054] The above formula will confirm the appointment. It serves as a source term in the FPK equation rather than a decision variable. This design ensures that new optimization decisions do not affect confirmed reservations, thus satisfying the business constraints of FCFS.

[0055] Based on the vehicle density field, the traffic flow and travel time of road segments within the aforementioned regional road network are calculated. A correlation is established between microscopic vehicle behavior and macroscopic road network status to reflect the dynamic characteristics of traffic congestion, achieving bidirectional feedback between the micro and macroscopic levels of the charging dynamic model. The calculations are as follows:

[0056]

[0057]

[0058] In the formula, For position Total vehicle density; For road section Traffic, For road section Free-flow travel time, This is the flow-time sensitivity coefficient. It is calibrated in advance through data observation. This method of calculating road segment impedance (travel time) mainly comes from the BPR macroscopic model in the traffic engineering handbook. Engineers can adjust the impedance model as needed without affecting the solution.

[0059] Furthermore, charging demand information is extracted from the vehicle density field. This provides a basis for the planning and scheduling of charging facilities, namely

[0060]

[0061]

[0062] In the formula, Electric vehicles with a safe battery level threshold (e.g., 20%) must be charged if their battery level falls below this threshold. For the first The battery safety threshold for this type of vehicle (e.g., 30%) is set; new energy vehicles with a battery level below this threshold may need to be charged. (This is in accordance with the relationship...) Since the user's vehicle charging strategy is "charge until full and then leave," it generally meets the requirements. .

[0063] In step S6, the optimization window length of the rolling optimization mechanism is set to... =15-30 minutes, optimize frequency set to =5-10 minutes; at the start of each optimization round, the initial conditions of the road network are set as follows:

[0064]

[0065] In the formula, For the density distribution of newly entering vehicles, For the first The density function for confirmed reservation vehicles; For Dirac delta function; For vehicles In time Location, For vehicles In time The amount of electricity; Indicates the current time A collection of previously confirmed appointment information.

[0066] To ensure the FCFS constraint is met, the following condition is verified after each round of optimization.

[0067] In the formula For the first The value function for a vehicle class represents the value of a vehicle from state 1. (i.e., time) ,Location Battery The minimum total cost required to travel from the origin to the destination.

[0068] A dynamic charging time planning system that considers first-come, first-served constraints includes: The acquisition module is used to define the regional road network range and extract its topology. Build modules are used to build time-based systems. ,Location Battery The system plans the state of the space and divides vehicles into three categories: the first category is non-charging vehicles, the second category is new energy vehicles with no demand data or that do not accept reservation services, and the third category is new energy vehicles of users with clear demand and reservations. The data on the proportion of the three categories of vehicles is obtained through pre-survey, entrance ramp statistics or travel reservation system. Each service area is equipped with a certain number of charging piles as service reserved charging piles. The proportion of service reserved charging piles is matched with the proportion of the third category of vehicles in the total number of the second and third categories of vehicles. The computation module is used to establish a coupled HJB-FPK equation system, where the HJB equations describe the optimal charging strategy for the three types of vehicles, and the FPK equations describe the density evolution of the three types of vehicles; and to process the confirmed reservation density. As a source term of the FPK equation, it satisfies ;confirm This ensures that new optimization decisions do not affect confirmed reservations, thereby ensuring that the business constraints of FCFS are met in every cycle; a rolling optimization mechanism is adopted, and initial conditions are set. While maintaining the computational efficiency advantage of MFG, it realizes the transformation of FCFS constraints from external operational suggestions to endogenous mathematical properties.

[0069] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described dynamic charging time planning method that considers first-come-first-served constraints.

[0070] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described dynamic charging time planning method that takes into account first-come-first-served constraints.

[0071] Specific applications The following section uses the Sioux Falls road network, commonly used in traffic planning research, as an example to illustrate the calculation process of the method proposed in this invention.

[0072] Scenario: Sioux Falls road network (Number of road network nodes: 24, Time window: 8:00-10:00 AM) Input data: Initial vehicle density: .

[0073] Service area distance: approximately 50km, location (Service areas are located at nodes 7, 12, 16, and 20).

[0074] Power constraints: (Power safety threshold).

[0075] Reserve ratio: (Percentage of reserved charging piles).

[0076] Implementation steps: (1) Initialization, setting spatial discretization Node, time step .

[0077] (2) Rolling optimization: Optimization is performed every 10 minutes; window length... .

[0078] Convert confirmed appointments into density functions As initial conditions.

[0079] (3) HJB solution: Solve the value function for the three types of vehicles .

[0080] Generate optimal control .

[0081] (4) Evolution of FPK: Solving the FPK equations to generate the evolution density field .

[0082] Consider the density evolution of confirmed reservations .

[0083] (5) Reservation decision: For newly entered vehicles Calculate the optimal route and charging schedule.

[0084] verify Ensure FCFS constraints are met.

[0085] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention, and these will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. A dynamic charging time planning method considering first-come-first-served constraints, characterized in that, The method comprises the following steps: Step S1: setting the range of the regional road network and extracting the topological structure thereof; Step S2: constructing a state planning space containing time , location , and power , and classifying vehicles into at least three categories: the first category of vehicles is non-charging vehicles, the second category of vehicles is new energy vehicles without demand data or not accepting pre-ordered services, and the third category of vehicles is new energy vehicles with clear demand and pre-ordered users; the data of the proportion of the three categories of vehicles is obtained through prior investigation or entrance ramp statistics or a travel reservation system; a certain number of charging piles are set for service reservation charging piles in each service area, and the proportion of the service reservation charging piles is adapted to the proportion of the third category of vehicles in the total number of the second category of vehicles and the third category of vehicles. Step S3: establishing a coupled HJB-FPK equation system, wherein the HJB equation describes the optimal charging strategy of the three types of vehicles, and the charging strategy of the third type of vehicle is "charge enough and go", and the FPK equation describes the density evolution of the three types of vehicles; Step S4: taking the confirmed reservation density as a source term of the FPK equation to meet the process of effectively describing the density evolution of the three types of vehicles by the FPK equation; Step S5: confirming the equivalent constraint condition to ensure that the new optimized decision does not affect the confirmed reservation, thereby realizing that the FCFS service constraint is established every cycle; Step S6: adopting a rolling optimization mechanism and setting an initial condition to realize the minimum total cost of the vehicles arriving at the destination under the FCFS constraint while maintaining the efficiency advantage of the MFG calculation.

2. The dynamic charging time planning method considering first-come first-served constraint of claim 1, wherein, The contributions of the first type of vehicle, the second type of vehicle and the third type of vehicle are represented as: wherein, : is a time variable; : is a spatial position variable, : is a set of road network nodes; : is an electric quantity state variable, is the density of the class vehicles at time , position , electric quantity .

3. The method of claim 1, wherein, The HJB equation describes the process of the optimal charging strategy of each type of vehicle as: wherein, is the travel time of a road segment is the flow of a road segment, is the impedance model of a road segment, is the flow-time sensitivity coefficient of a road segment after linearization of the impedance model; is the electric quantity dynamic function of the th vehicle, i.e. wherein is a path decision, is a charge amount decision, is an electric power consumption rate, is a charging efficiency; is a first vehicle waiting time at a location , wherein, When the oil car has no charging demand, the waiting time in front of the charging pile can be ignored, and only the time loss of passing through the service area is considered, so ; At this time, the waiting time of the non-reserved new energy vehicle is considered, that is, The above waiting time can be replaced by a queuing theory model to more accurately reflect the actual waiting time of the second type of vehicle; At this time, the waiting time of the reservation user is considered, that is, In the formula It is the current position Is it a service area type? The judgment is made such that if the condition is met, then 1 is selected; otherwise, 0 is selected. The time lost by vehicles entering and leaving the service area; The length of the queue for reserved vehicles. Queue length for vehicles without reservations; To reserve the number of charging stations, This refers to the number of charging piles not reserved for future use. Service area service rate.

4. The dynamic charging time planning method considering first-come first-served constraint of claim 3, wherein, The FPK equation describes the spatiotemporal evolution process of the density of the three types of vehicles as: wherein is the density of confirmed reservations for vehicles of the class, is the rate of change of the density of confirmed reservations for vehicles over time.

5. The dynamic charging time planning method considering first-come first-served constraint according to claim 4, characterized in that, Based on the vehicle density field, the road section flow and travel time in the range of the regional road network are calculated, the association between the micro-vehicle behavior and the macro-road network state is established, the dynamic characteristics of traffic congestion are reflected, the micro-macro bidirectional feedback of the charging dynamic model is realized, and the following calculation is performed: wherein is the total vehicle density at location is the flow on link is the free-flow travel time on link is the flow-time sensitivity coefficient.​​​ 6. The dynamic charging time planning method considering first-come first-served constraint of claim 5, wherein, Extracting charging demand information from vehicle density fields providing a basis for charging infrastructure planning and scheduling, i.e. In the formula, is a safety power threshold value; is a power safety threshold value for the first class vehicle; the relationship is satisfied; and since the charging strategy of the user's vehicle is "charge enough and go", the relationship is generally satisfied.

7. The method of claim 1, wherein, In step S6, the optimization window length of the rolling optimization mechanism is set to = 15-30 minutes, and the optimization frequency is set to = 5-10 minutes; at the beginning of each round of optimization, the initial conditions of the road network are set to: wherein is the density distribution of newly entering vehicles, is the density function of vehicles of the class that have confirmed reservations; is the Dirac delta function; is the amount of electric charge of a vehicle at time at location is the amount of electric charge of a vehicle at time denotes the set of reservation information that has been confirmed before the current time ​​ In order to ensure the FCFS constraint, after each round of optimization calculation is completed, the following condition is verified to be established wherein is the value function for the class of vehicles, representing the minimum total cost required to reach the destination from the state .

8. A dynamic charging time planning system considering first-come-first-serve constraints, characterized in that, Comprise: The acquisition module is configured to set the range of the regional road network and extract the topological structure thereof; a construction module for constructing a state planning space containing time , location , and power , and classifying vehicles into three categories: the first category of vehicles is non-charging vehicles, the second category of vehicles is new energy vehicles without demand data or not accepting reservation services, and the third category of vehicles is new energy vehicles with clear demand and reserved users; data of proportions of the three categories of vehicles is obtained through prior investigation or entrance ramp statistics or a travel reservation system, a certain number of charging piles are reserved for services in each service area, and the proportion of the service reserved charging piles is adapted to the proportion of the third category of vehicles in the total number of the second category of vehicles and the third category of vehicles. A computing module is configured to establish a coupled HJB-FPK equation system, wherein the HJB equation describes the optimal charging strategy of the three types of vehicles, and the FPK equation describes the density evolution of the three types of vehicles; the confirmed reservation density As a source term of the FPK equation, it satisfies ; confirm , ensure that the new optimization decision does not affect the confirmed reservation, and thus the service constraint of FCFS is always true; a rolling optimization mechanism is adopted, and initial conditions , while maintaining the computational efficiency advantage of MFG, the transformation of the FCFS constraint from external operation suggestion to endogenous mathematical property is realized. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the dynamic charging time planning method considering the first-come-first-served constraint according to any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the dynamic charging time planning method considering the first-come-first-served constraint according to any one of claims 1-7.

Images