Cooperative scheduling optimization method and system for urban electric logistics fleet, equipment and medium

Abstract

The invention discloses a cooperative scheduling optimization method and system for an urban electric logistics fleet, equipment and a medium, and the method comprises the steps: building a fleet cooperative scheduling model based on a Markov decision model, and bringing charging, distribution, transport capacity scheduling and order receiving into an action space of an intelligent agent, the influence of factors such as path planning, charging, freight demand, transport capacity scheduling and the like on a vehicle scheduling strategy is comprehensively considered, and the collaborative scheduling optimization effect is greatly improved; moreover, by converting a vehicle scheduling action selection mechanism in each area into a 0 / 1 planning problem, the flexibility of real-time action selection is greatly improved, and the method can be well suitable for a scheduling scene with a delivery demand and a charging cost change. And meanwhile, scheduling optimization is carried out by taking each grid region of the urban distribution region as an object, so that the phenomenon of unbalanced power load of the urban region can be relieved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric logistics fleet scheduling optimization, and in particular, to a collaborative scheduling optimization method and system for urban electric logistics fleets, electronic equipment, and a computer-readable storage medium. Background Art

[0002] Under the "dual carbon" goal, the electrification process of urban vehicles is constantly advancing, and the logistics and transportation link is the focus of urban energy conservation and emission reduction. In addition, with the development of intelligent and networked logistics systems, user delivery needs are becoming more diversified and random. Therefore, urban electric logistics fleets not only need to meet the growing and changing "last mile" delivery needs, but also have the ability to use the fleet charging behavior to achieve a benign interaction with the power grid. Due to the current inconsistent charging service electricity prices of urban charging facility operators, fleet operators can coordinate and dispatch vehicles during the distribution process, which can effectively reduce charging costs and increase cargo delivery rates, using less electricity in the same time and obtaining higher returns.

[0003] However, the existing urban electric logistics fleet scheduling method only considers path planning and charging behavior control, but does not consider the impact of the uncertainty and randomness of freight demand on vehicle scheduling strategies, resulting in unsatisfactory scheduling results. For example, patent CN108764777A discloses a scheduling method and system for electric logistics vehicles with a time window, which considers planning requirements and charging hard constraints in the vehicle scheduling process, and obtains vehicle charging scheduling results through a mixed integer programming model. However, this method is limited to the path planning and performance optimization problems of electric logistics vehicles, and does not consider the impact of freight demand on vehicle scheduling strategies. Summary of the invention

[0004] The present invention provides a collaborative scheduling optimization method and system for an urban electric logistics fleet, an electronic device, and a computer-readable storage medium. The method comprehensively considers the impact of factors such as path planning, charging, freight demand, and capacity scheduling on vehicle scheduling strategies, greatly improving the collaborative scheduling optimization effect. The method can also be well applied to scheduling scenarios with changes in delivery demand and charging costs, and is conducive to alleviating the imbalance of power load in urban areas.

[0005] According to one aspect of the present invention, a coordinated dispatch optimization method for an urban electric logistics fleet is provided, comprising the following contents:

[0006] Grid modeling of urban distribution areas for urban electric logistics fleets;

[0007] Taking the fleet dispatching platform as the intelligent body and the gridded urban distribution area as the interactive environment, a fleet collaborative dispatching model is constructed based on the Markov decision model, and the vehicle dispatching action selection mechanism in each area in the fleet collaborative dispatching model is converted into a 0 / 1 planning problem, where the vehicle dispatching actions include selecting to accept new goods, selecting charging, selecting delivery, and selecting capacity dispatching;

[0008] Based on the fleet collaborative scheduling model, the optimal scheduling result in each area is obtained, and each area is collaboratively scheduled according to the optimal scheduling result.

[0009] Furthermore, the process of converting the vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model into a 0 / 1 planning problem includes the following:

[0010] Determine all possible actions that vehicles can perform in each area, and calculate the state-action value table corresponding to each vehicle and action in each area;

[0011] A 0 / 1 variable is used to represent the selection results of different actions of vehicles in each area, where 1 represents a successful selection and 0 represents a failed selection.

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

[0013] Furthermore, 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 corresponding to action j of vehicle i, Q i,j Represents the state-action value of vehicle i choosing action j.

[0014] Furthermore, the constraints of the optimization objective function include the charging assumption: after a vehicle in the charging area is connected to a charging pile, the vehicle must be fully charged before the next action can be selected; the driving assumption: the time taken by a vehicle to move from one area to any adjacent area is the same; the selection assumption: a vehicle can only choose one action.

[0015] Furthermore, the process of gridding modeling the urban distribution area of ​​the urban electric logistics fleet includes the following contents:

[0016] Determine the urban distribution area of ​​the urban electric logistics fleet, take the preset time as a time step, determine the distance between each grid center according to the average speed of the logistics vehicles, grid the urban distribution area, and count the distribution of public charging stations in the urban distribution area, and mark the grid areas with public charging stations.

[0017] Furthermore, after the coordinated scheduling is completed, the following contents are also included:

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

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

[0020] If the vehicle chooses to accept new cargo, the vehicle's cargo logistics information will be updated; if the vehicle chooses to charge, the number of available charging piles in the area and the battery status of the vehicle will be updated; if the vehicle chooses to deliver, the vehicle information and quantity in each area will be updated after the vehicle moves to the next area according to the planned route; if the vehicle chooses capacity scheduling, the vehicle information and quantity in each area will be updated after the vehicle moves to the selected adjacent area.

[0021] In addition, the present invention also provides a coordinated dispatch optimization system for an urban electric logistics fleet, comprising:

[0022] Environmental modeling module, used for grid modeling of urban distribution areas for urban electric logistics fleets;

[0023] The model building module is used to build a fleet collaborative scheduling model based on the Markov decision model, with the fleet scheduling platform as the intelligent agent and the gridded urban distribution area as the interactive environment. The vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model is converted into a 0 / 1 planning problem, where the vehicle scheduling actions include selecting to accept new goods, selecting charging, selecting distribution, and selecting capacity scheduling;

[0024] The collaborative scheduling module is used to obtain the optimal scheduling result in each area based on the fleet collaborative scheduling model, and to perform collaborative scheduling for each area based on the optimal scheduling result.

[0025] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0026] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for collaborative scheduling optimization of urban electric logistics fleets, and the computer program executes the steps of the method described above when running on a computer.

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

[0028] The collaborative scheduling optimization method for the urban electric logistics fleet of the present invention constructs a fleet collaborative scheduling model based on the Markov decision model. By incorporating charging, distribution, capacity scheduling and order receiving into the action space of the intelligent body, it comprehensively considers the influence of factors such as path planning, charging, freight demand, capacity scheduling, etc. on the vehicle scheduling strategy, and greatly improves the collaborative scheduling optimization effect; and, by converting the vehicle scheduling action selection mechanism in each area into a 0 / 1 planning problem, the flexibility of real-time action selection is greatly improved, which can be well applied to scheduling scenarios with changes in delivery demand and charging cost; at the same time, each grid area in the urban distribution area is used as an object for scheduling optimization, which is conducive to alleviating the imbalance of power load in urban areas.

[0029] In addition, the collaborative scheduling optimization system for the urban electric logistics fleet of the present invention also has the above advantages.

[0030] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0032] Figure 1 It is a flow chart of a collaborative dispatch optimization method for an urban electric logistics fleet according to a preferred embodiment of the present application;

[0033] Figure 2 yes Figure 1 Schematic diagram of the sub-process of step S2;

[0034] Figure 3 This is another flow chart of the collaborative scheduling optimization method of the urban electric logistics fleet in the preferred embodiment of the present application;

[0035] Figure 4 It is a schematic diagram of the module structure of a collaborative scheduling optimization system for an urban electric logistics fleet according to another embodiment of the present application. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] Reference Figure 1 The preferred embodiment of the present application provides a collaborative scheduling optimization method for an urban electric logistics fleet, including the following contents:

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

[0039] Step S2: Taking the fleet dispatching platform as the intelligent agent and the gridded urban distribution area as the interactive environment, a fleet collaborative dispatching model is constructed based on the Markov decision model, and the vehicle dispatching action selection mechanism in each area in the fleet collaborative dispatching model is converted into a 0 / 1 planning problem, where the vehicle dispatching actions include selecting to accept new goods, selecting to charge, selecting to deliver, and selecting to dispatch capacity;

[0040] Step S3: Obtain the optimal dispatch result in each area based on the fleet collaborative dispatch model, and perform collaborative dispatch for each area according to the optimal dispatch result.

[0041] It can be understood that the collaborative scheduling optimization method for the urban electric logistics fleet of this embodiment first grids the urban distribution area of ​​the urban electric logistics fleet, and then uses the fleet scheduling platform as the intelligent body and the gridded urban distribution area as the interactive environment to build a fleet collaborative scheduling model based on the Markov decision model, and converts the vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model into a 0 / 1 planning problem, wherein the vehicle scheduling action includes choosing to accept new goods, choosing charging, choosing distribution, and choosing capacity scheduling, so that the optimal scheduling result in each area can be obtained based on the fleet collaborative scheduling model, and each area can be collaboratively scheduled according to the optimal scheduling result. The collaborative scheduling optimization method of the present invention constructs a fleet collaborative scheduling model based on a Markov decision model. By incorporating charging, distribution, capacity scheduling and order receiving into the action space of the intelligent body, the influence of factors such as path planning, charging, freight demand, capacity scheduling and the like on the vehicle scheduling strategy is comprehensively considered, thereby greatly improving the collaborative scheduling optimization effect. Moreover, by converting the vehicle scheduling action selection mechanism in each area into a 0 / 1 planning problem, the flexibility of real-time action selection is greatly improved, which can be well applied to scheduling scenarios with changes in delivery demand and charging cost. At the same time, scheduling optimization is performed on each grid area in the urban distribution area, which is conducive to alleviating the imbalance of power load in urban areas.

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

[0043] Determine the urban distribution area of ​​the urban electric logistics fleet, take the preset time as a time step, determine the distance between each grid center according to the average speed of the logistics vehicles, grid the urban distribution area, and count the distribution of public charging stations in the urban distribution area, and mark the grid areas with public charging stations.

[0044] Specifically, first determine the urban distribution area of ​​the urban electric logistics fleet, that is, the vehicle retrieval area in the city, and then use 15 minutes as a time step to determine the distance between each grid center according to the average speed of the logistics vehicle, so as to regionally grid the urban distribution area, that is, divide the urban distribution area into multiple regional grids, so as to facilitate the subsequent coordinated dispatch optimization of vehicles in each area, which is conducive to alleviating the imbalance of power load in the urban area. In addition, count the distribution of public charging stations in the urban distribution area, and mark the grid areas with public charging stations. In addition, for grid areas with public charging stations, it is necessary to count the number of charging piles, charging power and charging electricity prices.

[0045] It can be understood that in step S2, the fleet dispatching platform is used as the intelligent agent and the gridded urban distribution area is used as the interactive environment. The fleet collaborative dispatching model is constructed based on the Markov decision model, wherein the Markov decision model consists of a five-tuple, which can be expressed as<S,A,R,P,γ> , S and A represent the set of states that the agent can observe and the set of actions that can be selected, that is, the state space and the action space, respectively. R represents the reward function. It represents the immediate reward obtained by the agent after executing action a in state s and transferring the state to s'. P represents the probability function. represents the probability that the agent will transfer to state s' after selecting action a in state s, and γ is the discount factor, which indicates the importance that the agent attaches to future rewards. In addition, the state value and state action value of vehicle j at time t can be expressed 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 ) can be regarded as the cumulative revenue of the fleet. The agent will select actions according to the set action selection mechanism, and collect historical information when interacting with the environment and store it in the experience revisit pool, and then sample from it to train the neural network. In order to maintain a certain exploration rate, the algorithm uses the ε-greedy algorithm during training. For example, the agent randomly dispatches vehicles with a probability of ε, and selects actions with a probability of (1-ε) using a coordinated scheduling strategy based on the temporal difference learning method. During training, ε is set to a random number between [0, 1], and in the test phase, ε is set to 0.

[0046] It can be understood that every time the agent interacts with the environment, the interaction information (s, a, r, s') will be stored in the experience revisit pool with a capacity of B, where s represents the current state, a represents the action selected by the agent, r represents the immediate reward obtained by the agent for selecting action a, and s' represents the next state observed after the agent performs action a. Two neural networks are set up as the estimation network V(s, a; θ) and the target network V'(s, a; θ) to fit the state value V(s) of the agent in state s. Among them, when the agent interacts with the environment, it will extract the current cargo delivery information (cargo volume and weight) in real time, divide and organize the starting point and the delivery destination according to the specified area, and form a complete action option with the rest of the information. It will also determine the number of charging piles, charging power and charging electricity price according to the charging area mark. After storing the interaction information in the experience revisit pool, the target value is calculated according to the target network: TargetQ = (r + γ * V' (s, a)), and then according to the loss function L (θ) = E [(TargetQ - Q (s, a; θ)) 2 ]Update the network parameters of the estimated network. When the number of parameter updates of the estimated network reaches the preset requirements, update the network parameters of the target network.

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

[0048] Step S21: determine all possible actions that vehicles can perform in each area, and calculate a state-action value table that corresponds one-to-one between vehicles and actions in each area;

[0049] Step S22: using 0 / 1 variables to represent the selection results of different actions corresponding to the vehicles in each area, where 1 represents a successful selection and 0 represents a failed selection;

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

[0051] Specifically, we first determine all possible actions that vehicles can perform in each area, where vehicle scheduling actions include choosing to accept new goods, choosing to charge, choosing to deliver, and choosing to dispatch transportation capacity. Then we calculate the state action value table that corresponds one vehicle to one action in each area, for example, Q i，jRepresents the state-action value of the intelligent agent after vehicle i performs action j. The specific calculation process belongs to the existing technology and will not be repeated here. Then, a 0 / 1 variable is used to represent the selection results of different actions corresponding to vehicles in each area, where 1 represents a successful selection and 0 represents a failed selection. For example, the action space that vehicles can choose in a certain area includes charging, distribution, and transportation scheduling. If the vehicle chooses to charge successfully, the 0 / 1 variable corresponding to the vehicle and the charging action is 1, while the 0 / 1 variable corresponding to the vehicle and the distribution action and the transportation scheduling action is 0. Next, based on the state-action value table and the 0 / 1 variable, the sum of the state-action values ​​of each area is calculated, and the maximization of the sum of the state-action values ​​is used as the optimization objective function of each area, 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 corresponding to the action j of vehicle i, Q i,j Indicates the state action value of vehicle i choosing action j. In addition, the constraints of the optimization objective function include the charging assumption: after the vehicle in the charging area is connected to the charging pile, the vehicle must be fully charged before the next action can be selected; the driving assumption: the time taken by the vehicle to move from one area to any adjacent area is the same; the selection assumption: a vehicle can only choose one action.

[0052] It can be understood that in step S3, the optimal scheduling result in each area is obtained based on the fleet collaborative scheduling model. The specific process is: the intelligent agent calculates the reward r that can be obtained after the vehicle in each area selects an action, obtains the state action value q(s,a) of selecting action a under state s, integrates the state action value table Q(s,a) of all actions of vehicles in each area, and then combines the 0 / 1 variable results of all vehicle selection actions to construct an optimization objective function, solve the optimization objective function, obtain the optimal 0 / 1 variable result in each area, that is, obtain the optimal action strategy, and then coordinate each area according to the optimal action strategy. Among them, the specific solution process belongs to the existing technology and will not be repeated here. For example, the existing dichotomy or Dinkelbach algorithm can be used to solve the 0 / 1 planning problem.

[0053] In addition, if Figure 3 As shown, the collaborative scheduling optimization method of the urban electric logistics fleet further includes the following contents after the collaborative scheduling is completed:

[0054] Step S4: Update the state and environment parameters of the agent.

[0055] Specifically, if the vehicle chooses to accept new cargo, the vehicle's cargo logistics information is updated; if the vehicle chooses to charge, the number of available charging piles in the area and the battery status of the vehicle are updated; if the vehicle chooses to deliver, 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 chooses to dispatch, the vehicle information and quantity in each area are updated after the vehicle moves to the selected adjacent area. After the vehicle in each area performs the action, the state and environmental parameters of the agent are automatically updated, and the interaction between the agent and the environment is performed at the next moment to facilitate the next scheduling optimization.

[0056] In addition, if Figure 4 As shown, another embodiment of the present invention further provides a coordinated dispatching optimization system for a city electric logistics fleet, preferably using the coordinated dispatching optimization method for a city electric logistics fleet as described above, including:

[0057] Environmental modeling module, used for grid modeling of urban distribution areas for urban electric logistics fleets;

[0058] The model building module is used to build a fleet collaborative scheduling model based on the Markov decision model, with the fleet scheduling platform as the intelligent agent and the gridded urban distribution area as the interactive environment. The vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model is converted into a 0 / 1 planning problem, where the vehicle scheduling actions include selecting to accept new goods, selecting charging, selecting distribution, and selecting capacity scheduling;

[0059] The collaborative scheduling module is used to obtain the optimal scheduling result in each area based on the fleet collaborative scheduling model, and to perform collaborative scheduling for each area based on the optimal scheduling result.

[0060] It can be understood that the collaborative scheduling optimization system of the urban electric logistics fleet of this embodiment first grids the urban distribution area of ​​the urban electric logistics fleet, and then uses the fleet scheduling platform as the intelligent body and the gridded urban distribution area as the interactive environment to build a fleet collaborative scheduling model based on the Markov decision model, and converts the vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model into a 0 / 1 planning problem, wherein the vehicle scheduling action includes choosing to accept new goods, choosing charging, choosing distribution, and choosing capacity scheduling, so that the optimal scheduling result in each area can be obtained based on the fleet collaborative scheduling model, and each area can be collaboratively scheduled according to the optimal scheduling result. The collaborative scheduling optimization system of the present invention constructs a fleet collaborative scheduling model based on the Markov decision model. By incorporating charging, distribution, capacity scheduling and order receiving into the action space of the intelligent body, it comprehensively considers the impact of factors such as path planning, charging, freight demand, capacity scheduling, etc. on the vehicle scheduling strategy, and greatly improves the collaborative scheduling optimization effect; and, by converting the vehicle scheduling action selection mechanism in each area into a 0 / 1 planning problem, it greatly improves the flexibility of real-time action selection, which can be well applied to scheduling scenarios with changes in delivery demand and charging costs; at the same time, taking each grid area in the urban distribution area as an object for scheduling optimization is conducive to alleviating the imbalance of power load in urban areas.

[0061] In addition, the collaborative dispatching optimization system of the urban electric logistics fleet also includes:

[0062] The data update module is used to update the state and environment parameters of the intelligent agent.

[0063] In addition, another embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0064] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for collaborative scheduling optimization of urban electric logistics fleets, wherein the computer program executes the steps of the method described above when running on a computer.

[0065] The general form of computer readable storage media includes: floppy disk, flexible disk, hard disk, magnetic tape, any other magnetic medium, CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash erasable programmable read-only memory (FLASH-EPROM), any other memory chip or cartridge, or any other medium that can be read by a computer. The instructions can further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium that can be used to store, encode or carry instructions for execution by a machine, and includes digital or analog communication signals or other intangible media that facilitate the communication of the above instructions. Transmission media include coaxial cables, copper wires and optical fibers, which include the wires of a bus used to transmit a computer data signal.

[0066] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0067] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0068] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0069] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0070] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0072] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A collaborative scheduling optimization method for an urban electric logistics fleet, characterized in that: Includes the following: Grid modeling of urban distribution areas for urban electric logistics fleets; Taking the fleet dispatching platform as the intelligent body and the gridded urban distribution area as the interactive environment, a fleet collaborative dispatching model is constructed based on the Markov decision model, and the vehicle dispatching action selection mechanism in each area in the fleet collaborative dispatching model is converted into a 0 / 1 planning problem, where the vehicle dispatching actions include selecting to accept new goods, selecting charging, selecting delivery, and selecting capacity dispatching; Based on the fleet collaborative scheduling model, the optimal scheduling result in each area is obtained, and each area is collaboratively scheduled according to the optimal scheduling result.

2. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 1, characterized in that: The process of converting the vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model into a 0 / 1 planning problem includes the following: Determine all possible actions that vehicles can perform in each area, and calculate the state-action value table corresponding to each vehicle and action in each area; A 0 / 1 variable is used to represent the selection results of different actions of vehicles in each area, where 1 represents a successful selection and 0 represents a failed selection. The sum of the state-action value of each region is calculated based on the state-action value table and the 0 / 1 variable, and the maximization of the sum of the state-action value is used as the optimization objective function of each region.

3. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 2, characterized in that: 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 corresponding to the action j of vehicle i, Q i,j Represents the state-action value of vehicle i choosing action j.

4. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 3 is characterized in that: The constraints of the optimization objective function include the charging assumption: after a vehicle in the charging area is connected to a charging pile, the vehicle must be fully charged before the next action can be selected; the driving assumption: the time taken by a vehicle to move from one area to any adjacent area is the same; the selection assumption: a vehicle can only choose one action.

5. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 1, characterized in that: The process of gridding modeling the urban distribution area of ​​the urban electric logistics fleet includes the following: Determine the urban distribution area of ​​the urban electric logistics fleet, take the preset time as a time step, determine the distance between each grid center according to the average speed of the logistics vehicles, grid the urban distribution area, and count the distribution of public charging stations in the urban distribution area, and mark the grid areas with public charging stations.

6. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 1, characterized in that: After the coordinated scheduling is completed, the following contents are also included: Update the state of the agent and the environment parameters.

7. The coordinated dispatch optimization method for urban electric logistics fleet according to claim 6, characterized in that: The process of updating the state and environment parameters of the agent includes the following: If the vehicle chooses to accept new cargo, the vehicle's cargo logistics information is updated; If the vehicle chooses to charge, the number of available charging stations in the area and the battery status of the vehicle are updated; If the vehicle chooses delivery, the vehicle information and quantity in each area will be updated after the vehicle moves to the next area according to the planned route; If the vehicle selects capacity scheduling, the vehicle information and quantity in each area will be updated after the vehicle moves to the selected adjacent area.

8. A coordinated dispatch optimization system for urban electric logistics fleets, characterized in that: include: Environmental modeling module, used for grid modeling of urban distribution areas for urban electric logistics fleets; The model building module is used to build a fleet collaborative scheduling model based on the Markov decision model, with the fleet scheduling platform as the intelligent agent and the gridded urban distribution area as the interactive environment. The vehicle scheduling action selection mechanism in each area in the fleet collaborative scheduling model is converted into a 0 / 1 planning problem, where the vehicle scheduling actions include selecting to accept new goods, selecting charging, selecting distribution, and selecting capacity scheduling; The collaborative scheduling module is used to obtain the optimal scheduling result in each area based on the fleet collaborative scheduling model, and to perform collaborative scheduling for each area based on the optimal scheduling result.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for collaborative scheduling optimization of an urban electric logistics fleet, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images