Method and system for optimizing coordinated scheduling of urban electric vehicle fleet, device, medium

Abstract

The application discloses a kind of urban electric logistics truck team's cooperative scheduling optimization method and system, equipment, medium, it is based on Markov decision model to construct truck team cooperative scheduling model, by charging, distribution, transport capacity scheduling and receiving order are included into the action space of agent, comprehensively consider the influence of path planning, charging, freight demand, transport capacity scheduling and other factors to vehicle scheduling strategy, greatly improve the cooperative scheduling optimization effect;And, by the vehicle scheduling action selection mechanism in each region is converted into 0 / 1 programming problem, greatly improve the flexibility of action real-time selection, can be well adapted to the scheduling scene of delivery demand and charging cost change;Meanwhile, each grid area of city distribution area is taken as object and is scheduled and optimized, it is favorable to alleviate city area power load imbalance phenomenon.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric logistics vehicle fleet scheduling optimization, in particular, relates to a collaborative scheduling optimization method and system for urban electric logistics vehicle fleet, an electronic device and a computer readable storage medium. BACKGROUND

[0002] Under the "double carbon" goal, the process of urban vehicle electrification is constantly advancing, and the logistics transportation link is the focus of urban energy saving and emission reduction. In addition, with the development of intelligentization and networking of logistics systems, user distribution demand is showing a trend of diversification and randomization. Therefore, urban electric logistics vehicle fleet not only needs to meet the growing and changing "last mile" distribution demand, but also needs to have the ability to interact with the power grid through vehicle charging behavior. Because the charging service prices of current urban charging facility operators are inconsistent, vehicle fleet operators can perform collaborative scheduling during vehicle distribution, which can effectively reduce charging costs and improve cargo delivery rates, use less electricity and obtain higher revenue in the same time.

[0003] However, the existing urban electric logistics vehicle fleet scheduling method only considers path planning and charging behavior control, without considering the impact of uncertainty and randomness of freight demand on vehicle scheduling strategy, resulting in unsatisfactory scheduling effect. For example, patent CN108764777A discloses an electric logistics vehicle scheduling method and system with time window, which considers the hard constraints of planning demand and charging in the vehicle scheduling process, and obtains the vehicle charging scheduling result through a mixed integer programming model. However, this method is limited to path planning and performance optimization of electric logistics vehicles, and does not consider the impact of freight demand on vehicle scheduling strategy. SUMMARY

[0004] The present application provides a collaborative scheduling optimization method and system for urban electric logistics vehicle fleet, an electronic device and a computer readable storage medium, which comprehensively considers the impact of path planning, charging, freight demand, transportation capacity scheduling and other factors on vehicle scheduling strategy, greatly improves the collaborative scheduling optimization effect, and can also be well applied to scheduling scenarios with changing delivery demand and charging cost, and is beneficial to alleviate the imbalance of urban regional power load.

[0005] According to one aspect of the present application, a collaborative scheduling optimization method for urban electric logistics vehicle fleet is provided, comprising the following contents:

[0006] Grid modeling is performed on the urban distribution area of the urban electric logistics vehicle fleet;

[0007] The vehicle fleet scheduling platform is an intelligent agent, the urban distribution area of the vehicle fleet is an interactive environment, a vehicle fleet cooperative scheduling model is constructed based on a Markov decision model, and a 0 / 1 programming problem is converted from a vehicle scheduling action selection mechanism in each area in the vehicle fleet cooperative scheduling model, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to distribute, and selecting to schedule transport capacity.

[0008] An optimal scheduling result in each area is obtained based on the vehicle fleet cooperative scheduling model, and each area is cooperatively scheduled according to the optimal scheduling result.

[0009] Further, the process of converting the vehicle scheduling action selection mechanism in each area in the vehicle fleet cooperative scheduling model into a 0 / 1 programming problem includes the following contents:

[0010] All possible actions performed by the vehicle in each area are determined, and a state-action value table corresponding to the vehicle and the action in each area is calculated;

[0011] A 0 / 1 variable is used to represent the selection result of the vehicle corresponding to different actions in each area, wherein 1 represents a successful selection, and 0 represents a failed selection;

[0012] The state-action value sum of each area is calculated based on the state-action value table and the 0 / 1 variable, and maximizing the state-action value sum is used as an optimization objective function of each area.

[0013] Further, the optimization objective function is f=Max∑(a i,j ×Q i,j ), wherein f represents the optimization objective function, a i,j represents the 0 / 1 variable of the vehicle i corresponding to the action j, and Q i,j represents the state-action value of the vehicle i selecting the action j.

[0014] Further, the constraint conditions of the optimization objective function include a charging assumption that the vehicle in the charging area needs to be fully charged before the next action selection after accessing the charging pile; a driving assumption that the time used by the vehicle to move from one area to any adjacent area is consistent; and a selection assumption that a vehicle can only select one action.

[0015] Further, the process of modeling the urban distribution area of the urban electric vehicle fleet in a grid includes the following contents:

[0016] The urban distribution area of the urban electric vehicle fleet is determined, a time step is determined according to a preset time length, the distance between the centers of the grids is determined according to the average speed of the logistics vehicle, the urban distribution area is regionally gridded, and the distribution of the public charging stations in the urban distribution area is counted to mark the grid area with the public charging stations.

[0017] Further, after completing the collaborative scheduling, the following is further included:

[0018] updating the state of the agent and the environment parameters.

[0019] Further, the process of updating the state of the agent and the environment parameters includes the following:

[0020] If the vehicle chooses to accept new goods, the vehicle's cargo flow information is updated; if the vehicle chooses to charge, the number of available charging piles in the region and the vehicle's battery state are updated; if the vehicle chooses to deliver, after the vehicle moves to the next region according to the planned route, the vehicle information and quantity in each region are updated; if the vehicle chooses to dispatch, after the vehicle moves to the selected adjacent region, the vehicle information and quantity in each region are updated.

[0021] In addition, the present application also provides a collaborative scheduling optimization system for an urban electric animal logistics vehicle fleet, comprising:

[0022] an environment modeling module for grid modeling the urban distribution region of the urban electric animal logistics vehicle fleet;

[0023] a model construction module for constructing a fleet collaborative scheduling model based on a Markov decision model, taking the fleet scheduling platform as an agent and the grid urban distribution region as an interactive environment, and converting the vehicle scheduling action selection mechanism in each region in the fleet collaborative scheduling model into a 0 / 1 planning problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to deliver, and selecting to dispatch;

[0024] a collaborative scheduling module for obtaining the optimal scheduling result in each region based on the fleet collaborative scheduling model, and collaboratively scheduling each region according to the optimal scheduling result.

[0025] In addition, the present application also provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program, and the processor, by calling the computer program stored in the memory, is used to execute the steps of the method as described above.

[0026] In addition, the present application also provides a computer-readable storage medium for storing a computer program for collaborative scheduling optimization of an urban electric animal logistics vehicle fleet, wherein the computer program executes the steps of the method as described above when running on a computer.

[0027] The present application has the following beneficial effects:

[0028] The synergistic scheduling optimization method of the urban electric logistics vehicle team of the application constructs a team synergistic scheduling model based on a Markov decision model, comprehensively considers the influence of path planning, charging, freight demand, and transport capacity scheduling on vehicle scheduling strategies by including charging, distribution, transport capacity scheduling, and order receiving into the action space of the agent, greatly improves the synergistic scheduling optimization effect, and greatly improves the flexibility of real-time action selection by converting the vehicle scheduling action selection mechanism in each region into a 0 / 1 planning problem, which can be well applied to scheduling scenarios with changing delivery demand and charging cost. At the same time, scheduling optimization is performed on each grid area of the urban distribution area, which is beneficial to relieving the urban area power load imbalance phenomenon.

[0029] In addition, the synergistic scheduling optimization system of the urban electric logistics vehicle team of the application also has the above advantages.

[0030] In addition to the purposes, features and advantages described above, the application has other purposes, features and advantages. The application will be further described below with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0031] The accompanying drawings, which form a part of this application, are included to provide a further understanding of the application, and are incorporated in and constitute a part of this application. The embodiments of the application illustrated in the drawings, and their description, serve to explain the application without limiting it. In the drawings:

[0032] Figure 1 is a flowchart of the synergistic scheduling optimization method of the urban electric logistics vehicle team of the preferred embodiment of the application;

[0033] Figure 2 is Figure 1 is a subflowchart of step S2 in

[0034] Figure 3 is another flowchart of the synergistic scheduling optimization method of the urban electric logistics vehicle team of the preferred embodiment of the application;

[0035] Figure 4 is a module structure diagram of the synergistic scheduling optimization system of the urban electric logistics vehicle team of another embodiment of the application. DETAILED DESCRIPTION

[0036] It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict. The application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0037] With reference to Figure 1 The preferred embodiment of the application provides a synergistic scheduling optimization method of an urban electric logistics vehicle team, including the following contents:

[0038] Step S1: grid modeling of the urban distribution area of the urban electric logistics vehicle fleet;

[0039] Step S2: taking the fleet scheduling platform as an agent and the grid urban distribution area as an interactive environment, constructing a fleet cooperative scheduling model based on a Markov decision model, and converting the vehicle scheduling action selection mechanism in each region of the fleet cooperative scheduling model into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to distribute, and selecting to schedule transport capacity;

[0040] Step S3: obtaining the optimal scheduling result in each region based on the fleet cooperative scheduling model, and cooperatively scheduling each region according to the optimal scheduling result.

[0041] It can be understood that the cooperative scheduling optimization method of the urban electric logistics vehicle fleet of the embodiment first performs grid modeling on the urban distribution area of the urban electric logistics vehicle fleet, then takes the fleet scheduling platform as an agent and the grid urban distribution area as an interactive environment, constructs a fleet cooperative scheduling model based on a Markov decision model, and converts the vehicle scheduling action selection mechanism in each region of the fleet cooperative scheduling model into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to distribute, and selecting to schedule transport capacity, so that the optimal scheduling result in each region can be obtained based on the fleet cooperative scheduling model, and each region is cooperatively scheduled according to the optimal scheduling result. The cooperative scheduling optimization method of the present application constructs a fleet cooperative scheduling model based on a Markov decision model, integrates charging, distribution, transport capacity scheduling, and order receiving into the action space of the agent, comprehensively considers the influence of path planning, charging, freight demand, transport capacity scheduling, and other factors on the vehicle scheduling strategy, greatly improves the cooperative scheduling optimization effect, and greatly improves the flexibility of real-time action selection, which can be well applied to scheduling scenarios with changing delivery demand and charging cost. At the same time, each grid region of the urban distribution area is taken as an object for scheduling optimization, which is conducive to relieving the unbalanced phenomenon of urban area power load.

[0042] It can be understood that in the step S1, the process of grid modeling of the urban distribution area of the urban electric logistics vehicle fleet includes the following contents:

[0043] The urban distribution area of the urban electric logistics vehicle fleet is determined, a time step is determined as a preset time length, the distance between each grid center is determined according to the average speed of the logistics vehicle, the urban distribution area is regionally gridded, and the distribution of public charging stations in the urban distribution area is counted to mark the grid regions with public charging stations.

[0044] Specifically, first, the urban distribution area of the urban electric logistics vehicle fleet is determined, that is, the pickup area of the vehicle in the city, and then the distance between each grid center is determined according to the average speed of the logistics vehicle, so as to divide the urban distribution area into a plurality of regional grids, that is, to divide the urban distribution area into a plurality of regional grids, so as to facilitate subsequent cooperative scheduling optimization of the vehicles in each region, and to alleviate the urban regional power load imbalance phenomenon. And, the distribution of public charging stations in the urban distribution area is counted, and the grid area with public charging stations is marked. In addition, for the grid area with public charging stations, the number of charging piles, the charging power and the charging price need to be counted.

[0045] It can be understood that in the step S2, the fleet scheduling platform is an intelligent agent, the grid urban distribution area is an interactive environment, and a fleet cooperative scheduling model is constructed based on a Markov decision model, wherein the Markov decision model is composed of a five-tuple, which can be represented as <S, A, R, P, γ>, S and A represent the state set and the action set that can be selected by the intelligent agent, that is, the state space and the action space, R represents the reward function, represents the immediate reward obtained by the intelligent agent after performing the action a in the state s and the state is transferred to s', P represents the probability function, represents the probability of the intelligent agent transferring to the state s' after selecting the action a in the state s, and γ is a discount factor, which represents the degree of attention of the intelligent agent to future rewards. In addition, the state value and the state-action value of the vehicle j at time t can be represented as V(s j ) and Q(s j ,a j ), since the four parts of the reward function are all vehicle benefits, the state value V(s j ) of the intelligent agent can be regarded as the cumulative benefit of the fleet. The intelligent agent will select actions according to the set action selection mechanism, and collect historical information when the intelligent agent and the environment interact and store it in the experience replay pool, and then sample it to train the neural network. In order to maintain a certain exploration rate, the algorithm uses an ε-greedy algorithm during training, for example, the intelligent agent randomly schedules the vehicle with a probability of ε, and selects actions using a cooperative scheduling strategy based on the time difference learning method with a probability of (1-ε). During training, ε is set to a random number between 0 and 1, while in the test phase, ε is set to 0.

[0046] It can be understood that each time the agent and the environment interact, the interaction information (s, a, r, s') is stored in the experience replay pool with a capacity of B, wherein s represents the current state, a represents the action selected by the agent, r represents the immediate reward obtained by the agent selecting the action a, and s' represents the next state observed by the agent after executing the action a. Two neural networks are set up as an estimation network V(s, a; θ) and a target network V'(s, a; θ) for fitting the state value V(s) of the agent in the state s. Among them, when the agent interacts with the environment, the current time cargo distribution information (cargo volume and weight) is extracted in real time, and the starting point and the distribution destination are arranged according to the specified region division label to form a complete action option with the rest of the information. According to the charging area mark, the number of charging piles, the charging power and the charging price are determined. After storing the interaction information in the experience replay pool, the target value TargetQ=(r+γ*V'(s,a)) is calculated according to the target network, and the network parameters of the estimation network are updated according to the loss function L(θ)=E[(TargetQ-Q(s,a;θ)) 2 ] When the number of parameter updates of the estimation network reaches the preset requirement, the network parameters of the target network are updated.

[0047] In addition, in the vehicle fleet cooperative scheduling model, the vehicle scheduling action selection mechanism in each region is converted into a 0 / 1 programming problem, which greatly improves the flexibility of real-time action selection and can be well applied to scheduling scenarios with changing delivery demands and charging costs. As shown in Figure 2 The process of converting the vehicle scheduling action selection mechanism in each region in the vehicle fleet cooperative scheduling model into a 0 / 1 programming problem includes the following contents:

[0048] Step S21: Determine all possible actions of the vehicle in each region, and calculate the state-action value table corresponding to each vehicle and action in each region;

[0049] Step S22: Use 0 / 1 variables to represent the selection results of different actions of the vehicle in each region, wherein 1 represents success and 0 represents failure;

[0050] Step S23: Calculate the state-action value sum of each region based on the state-action value table and the 0 / 1 variable, and maximize the state-action value sum as the optimization objective function of each region.

[0051] Specifically, first determine all possible actions of the vehicle in each region, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to deliver, and selecting to schedule transport capacity, and then calculate the state-action value table corresponding to each vehicle and action in each region, for example, Q i，jThe state-action value of the agent after the vehicle i performs the action j is represented, and the specific calculation process is prior art and will not be described here. Then, a 0 / 1 variable is used to represent the selection result of the vehicle in each region for different actions, where 1 represents a successful selection and 0 represents a failed selection. For example, the action space that a vehicle in a certain region can select includes charging, distribution, and capacity scheduling. If the vehicle selects charging successfully, the 0 / 1 variable corresponding to the vehicle and the charging action is 1, while the 0 / 1 variables corresponding to the vehicle and the distribution action and the capacity scheduling action are 0. Next, the state-action value sum of each region is calculated based on the state-action value table and the 0 / 1 variable, and the optimization objective function of each region is taken as the maximum state-action value sum, where the optimization objective function is: f = Max∑(a i,j ×Q i,j ), where f represents the optimization objective function, a i,j represents the 0 / 1 variable of the vehicle i corresponding to the action j, and Q i,j represents the state-action value of the vehicle i selecting the action j. In addition, the constraint conditions of the optimization objective function include the charging assumption that the vehicle in the charging region must be fully charged before performing the next action selection after connecting to the charging pile; the travel assumption that the time used by the vehicle to move from one region to any adjacent region is consistent; and the selection assumption that a vehicle can only select one action.

[0052] It can be understood that in the step S3, the optimal scheduling result in each region is obtained based on the fleet cooperative scheduling model. The specific process is as follows: the agent calculates the reward r that can be obtained by the vehicle in each region after selecting an action, obtains the state-action value q(s, a) of selecting the action a in the state s, integrates the state-action value table Q(s, a) of the vehicle in each region for all actions, combines the 0 / 1 variable results of all vehicle action selections to construct the optimization objective function, solves the optimization objective function, obtains the optimal 0 / 1 variable result in each region, i.e., obtains the optimal action strategy, and then cooperatively schedules each region according to the optimal action strategy. The specific solving process belongs to prior art and will not be described here. For example, the existing dichotomy or Dinkelbach algorithm can be used to solve the 0 / 1 programming problem.

[0053] In addition, as shown in Figure 3 , the cooperative scheduling optimization method of the urban electric vehicle fleet further includes the following contents after completing the cooperative scheduling:

[0054] Step S4: updating the state of the agent and the environmental parameters.

[0055] Specifically, if the vehicle selects to accept new goods, the vehicle's cargo flow information is updated; if the vehicle selects to charge, the number of available charging piles in the region and the battery status of the vehicle are updated; if the vehicle selects to deliver, after the vehicle moves to the next region according to the planned route, the vehicle information and the number in each region are updated; if the vehicle selects to dispatch, after the vehicle moves to the selected adjacent region, the vehicle information and the number in each region are updated. After the vehicle in each region performs the action, the state of the agent and the environmental parameters are automatically updated, and the interaction between the agent and the environment is performed again at the next time to facilitate the next scheduling optimization.

[0056] In addition, as shown in Figure 4 Another embodiment of the present application also provides a collaborative scheduling optimization system for an urban electric logistics vehicle fleet, which preferably adopts the urban electric logistics vehicle fleet collaborative scheduling optimization method as described above, and comprises:

[0057] An environment modeling module is configured to model the urban distribution region of the urban electric logistics vehicle fleet in a grid manner;

[0058] A model construction module is configured to take the fleet scheduling platform as an agent, take the grid urban distribution region as an interactive environment, construct a fleet collaborative scheduling model based on a Markov decision model, and convert the vehicle scheduling action selection mechanism in each region in the fleet collaborative scheduling model into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to deliver, and selecting to dispatch;

[0059] A collaborative scheduling module is configured to obtain the optimal scheduling result in each region based on the fleet collaborative scheduling model, and perform collaborative scheduling on each region according to the optimal scheduling result.

[0060] It can be understood that the cooperative scheduling optimization system of the urban electric animal logistics vehicle team in the embodiment first models the urban distribution area of the urban electric animal logistics vehicle team in a grid, then takes the vehicle team scheduling platform as an agent, takes the grid urban distribution area as an interactive environment, constructs a team cooperative scheduling model based on a Markov decision model, and converts the vehicle scheduling action selection mechanism in each region in the team cooperative scheduling model into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to distribute, and selecting to schedule transport capacity, so that the optimal scheduling result in each region can be obtained based on the team cooperative scheduling model, and each region is cooperatively scheduled according to the optimal scheduling result. The cooperative scheduling optimization system of the application constructs a team cooperative scheduling model based on a Markov decision model, integrates charging, distribution, transport capacity scheduling, and order receiving into the action space of the agent, comprehensively considers the influence of path planning, charging, freight demand, transport capacity scheduling, and other factors on the vehicle scheduling strategy, greatly improves the cooperative scheduling optimization effect, and greatly improves the flexibility of real-time action selection, which can be well applied to scheduling scenarios with changing delivery demand and charging cost. At the same time, each grid area of the urban distribution area is taken as an object for scheduling optimization, which is conducive to relieving the urban area power load imbalance phenomenon.

[0061] In addition, the cooperative scheduling optimization system of the urban electric animal logistics vehicle team further comprises:

[0062] The data updating module is configured to update the state of the agent and the environmental parameters.

[0063] In addition, another embodiment of the application further provides an electronic device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the steps of the method described above by invoking the computer program stored in the memory.

[0064] In addition, another embodiment of the application further provides a computer-readable storage medium for storing a computer program for cooperatively scheduling and optimizing an urban electric animal logistics vehicle team, wherein the computer program executes the steps of the method described above when running on a computer.

[0065] Generally, computer readable storage media includes any media that can be accessed by a computer. By way of example, and not limitation, computer readable storage media can comprise storage media such as RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other storage medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, or twisted pair, then the coaxial cable, fiber optic cable, or twisted pair are included in the definition of medium. Disk and disc, as used herein, includes compact disc (CD), laser disc, optical disc, Blu-ray® disc, and floppy disk used to store software.

[0066] Those skilled in the art will appreciate that embodiments of the present application can be further implemented in a computer program product tangibly embodied in a machine-readable storage medium (e.g., memory elements) including instructions configured to cause a machine to perform the

[0067] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks. Figure 1 The flowchart and / or block diagram in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart and / or block diagrams can represent a module, segment, or portion of instructions, which comprises one or more Figure 1 The means can also be implemented in software and / or firmware for execution by various components such as the processor components. As an example, the means for performing an operation is hardware (e.g., coprocessor), is a processor of a device (e.g., a central processing unit of a computer system) programmed to perform the operation, or is a combination of hardware and

[0068] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0069] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0070] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0071] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

[0072] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing the coordinated scheduling of an urban electric vehicle fleet, characterized in that, The application comprises the following contents: Grid modeling is performed on the urban distribution area of the urban electric logistics vehicle team; A team coordination scheduling model is constructed based on the Markov decision model, taking the team scheduling platform as an intelligent agent and the grid urban distribution area as an interactive environment, and the vehicle scheduling action selection mechanism in each area of the team coordination scheduling model is converted into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept new goods, selecting to charge, selecting to distribute, and selecting to schedule the transport capacity; Optimal scheduling results are obtained in each area based on the team coordination scheduling model, and each area is coordinated and scheduled according to the optimal scheduling results; The process of converting the vehicle scheduling action selection mechanism in each area of the team coordination scheduling model into a 0 / 1 programming problem comprises the following contents: All possible actions performed by the vehicle in each area are determined, and a state-action value table corresponding to the vehicle and the action in each area is calculated; 0 / 1 variables are used to represent the selection results of the vehicle corresponding to different actions in each area, wherein 1 represents a successful selection and 0 represents a failed selection; The total state-action value for each region is calculated based on the state-action value table and 0 / 1 variables. Maximizing this total state-action value is used as the optimization objective function for each region. The optimization objective function is: ,in, f This represents the objective function to be optimized. Indicates the action corresponding to vehicle i j 0 / 1 variables, Indicates the action selected by vehicle i. j The value of state actions; The process of obtaining optimal scheduling results in each area based on the team coordination scheduling model comprises the following contents: the intelligent agent calculates the reward r that can be obtained by the vehicle in each area after selecting an action, obtains the state-action value q(s, a) of selecting the action a in the state s, integrates the state-action value table Q(s, a) of the vehicle in each area for all actions, and then constructs an optimization objective function in combination with the 0 / 1 variable results of the action selected by all vehicles, solves the optimization objective function, and obtains the optimal 0 / 1 variable results in each area, i.e., the optimal action strategy; The constraint conditions of the optimization objective function include a charging assumption that the vehicle in the charging area needs to be fully charged before selecting the next action after connecting to the charging pile; a driving assumption that the time used by the vehicle to move from one area to any adjacent area is consistent; and a selection assumption that a vehicle can only select one action.

2. The method for coordinated scheduling optimization of an urban electric vehicle fleet according to claim 1, wherein, The process of grid modeling on the urban distribution area of the urban electric logistics vehicle team comprises the following contents: The urban distribution area of the urban electric logistics vehicle team is determined, a time step is determined as a preset time length, the distance between the centers of the grids is determined according to the average driving speed of the logistics vehicle, the urban distribution area is regionally gridded, and the distribution of public charging stations in the urban distribution area is counted to mark the grid area with the public charging stations.

3. The method of claim 1, wherein, After the coordination scheduling is completed, the following contents are further included: The state and environmental parameters of the intelligent agent are updated.

4. The method of claim 3, wherein, The process of updating the state and environmental parameters of the intelligent agent comprises the following contents: If the vehicle selects to accept new goods, the vehicle's cargo information is updated; If the vehicle selects to charge, the number of available charging piles in the area and the battery state of the vehicle are updated; If the vehicle selects to distribute, the vehicle information and quantity in each area are updated after the vehicle moves to the next area according to the planned route; If the vehicle selects to schedule the transport capacity, the vehicle information and quantity in each area are updated after the vehicle moves to the selected adjacent area.

5. A coordinated scheduling optimization system for an urban electric vehicle fleet, employing the coordinated scheduling optimization method for an urban electric vehicle fleet according to any one of claims 1 to 4, characterized in that, The application comprises the following contents: An environment modeling module is configured to model a city distribution area of the city electric logistics vehicle fleet by gridding; A model construction module is configured to construct a fleet cooperative scheduling model based on a Markov decision model, taking the fleet scheduling platform as an agent and the gridded city distribution area as an interactive environment, and convert a vehicle scheduling action selection mechanism in the fleet cooperative scheduling model into a 0 / 1 programming problem, wherein the vehicle scheduling action includes selecting to accept a new cargo, selecting to charge, selecting to distribute, and selecting to schedule a transport capacity. A cooperative scheduling module is configured to obtain an optimal scheduling result in each region based on the fleet cooperative scheduling model, and cooperatively schedule each region according to the optimal scheduling result.

6. An electronic device, comprising: A computer program product includes a processor and a memory having stored therein a computer program, the processor being configured to execute the steps of the method of any one of claims 1-4 by invoking the computer program stored in the memory.

7. A computer readable storage medium for storing a computer program for optimizing coordinated scheduling of urban electric vehicle fleets, characterized in that, The computer program product is configured to execute the steps of the method of any one of claims 1-4 when running on a computer.

Images