An electric heavy truck charging and discharging scheduling cooperative optimization method and system

Abstract

An electric heavy truck charging and discharging scheduling coordination optimization method and system, collects vehicle price sensitivity, queuing history and real-time location data to generate charging and discharging strategy data, and then through iterative game optimization and branch cutting processing, dynamically generates constraint information in each round of update to guide the convergence of the strategy. The core is to combine centralized scheduling with distributed decision-making, use the branch cutting algorithm to eliminate strategy conflicts in real time and quantify resource margin thresholds, both preserving vehicle autonomous selection space and ensuring load balancing of the battery swap station. This method effectively solves the problem of response lag and lack of adaptability of static scheduling schemes, and realizes efficient matching of vehicle demand and system resources in a dynamic scenario, improving the feasibility and execution efficiency of the scheduling scheme.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle charging and discharging scheduling optimization technology, and particularly relates to a collaborative optimization method and system for charging and discharging scheduling of electric heavy trucks. Background Technology

[0002] With the increasing use of electric heavy-duty trucks in logistics transportation, highway battery swapping stations face the challenge of scheduling charging resources due to competition among multiple vehicles. Because vehicle charging routes, price sensitivity, and the service capacity of battery swapping stations are dynamically interdependent, a method is needed to collaboratively optimize charging and discharging scheduling to balance user demand and station load, avoid localized congestion, and improve overall efficiency.

[0003] Currently, existing solutions employ a charging and discharging scheduling method based on a centralized optimization model. This method calculates the charging demand of each electric heavy-duty truck through a cloud-based central system and combines it with real-time capacity data from battery swapping stations to generate a static scheduling plan with the goal of minimizing global energy consumption. For example, patent documents with publication numbers CN119313096A and CN120046935A demonstrate that these methods can optimize the charging process of electric heavy-duty trucks through a comprehensive energy system and charging scheduling model. However, these solutions rely on linear programming to solve for optimal allocation and use fixed time windows to constrain the charging periods of vehicles.

[0004] Existing technologies have poor adaptability in dynamic scenarios. Because they do not introduce a vehicle autonomous decision-making mechanism, when some vehicles adjust their charging needs due to unforeseen circumstances, the global scheduling scheme must be recalculated, leading to increased computational latency. At the same time, static allocation strategies are difficult to cope with fluctuations in the load of battery swapping stations, which can easily result in idle resources or excessive congestion during certain periods, affecting scheduling efficiency. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a collaborative optimization method and system for charging and discharging scheduling of electric heavy-duty trucks. This method dynamically coordinates vehicle autonomous strategies and system constraints through a parallel computing framework. First, it collects vehicle price sensitivity, queuing history, and real-time location data to generate charging and discharging strategy data. Then, through iterative game theory optimization and branch-cutting processing, it dynamically generates constraint information in each update to guide strategy convergence. Its core lies in combining centralized scheduling with distributed decision-making, using a branch-cutting algorithm to eliminate strategy conflicts in real time and quantify resource reserve thresholds, thus preserving vehicle autonomous selection space while ensuring load balancing at battery swapping stations. This method effectively solves the problems of sluggish response and insufficient adaptability in static scheduling schemes, achieving efficient matching of vehicle demand and system resources in dynamic scenarios, and improving the feasibility and execution efficiency of scheduling schemes.

[0006] The present invention adopts the following technical solution.

[0007] This invention proposes a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks, comprising: Acquire battery swapping price-sensitive data, historical queuing data, and real-time location data for electric heavy-duty trucks; the battery swapping price-sensitive data includes the maximum acceptable price set by the vehicle owner. Price factor, waiting factor, and distance factor are calculated based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data, respectively, and then normalized. Dynamic weights are assigned to the three factors according to the current traffic conditions, and the weighted sum is used to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate charging and discharging strategy data. The charging and discharging strategy data includes the target battery swapping station and the charging time period. Based on the charging and discharging strategy data of all electric heavy trucks after each iteration, the scheduling strategy of electric heavy trucks is updated accordingly. Based on the updated scheduling strategy, the charging and discharging strategy data of each electric heavy truck is iteratively updated. If, after each iteration, the charging and discharging strategy data of an electric heavy truck does not meet the preset convergence condition, then the charging and discharging strategy data of all electric heavy trucks in the current iteration round is branched and processed to generate constraint information. Based on the constraint information, the update process of the next iteration round is limited until the charging and discharging strategy data of all electric heavy trucks meet the preset convergence condition. The charging and discharging strategy data of all electric heavy-duty trucks at the convergence point are combined into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy-duty trucks.

[0008] More preferably, the method for generating charge / discharge strategy data is as follows: The price factor is the ratio of the highest acceptable price in the battery swapping price-sensitive data to the price quoted by the battery swapping station; The waiting factor is the ratio of the tolerable waiting time of car owners in the historical queuing data to the actual estimated queuing time; the tolerable waiting time of car owners is the historical queuing waiting time of car owners. The distance factor is the reciprocal of the distance from the vehicle to the battery swapping station; The price factor, the waiting factor, and the distance factor are normalized, dynamic weights are assigned according to the current traffic conditions, and finally the weighted sum is obtained to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate strategy data. The method for setting dynamic weights is as follows: Based on the traffic condition data, the traffic congestion index is set to θ, ranging from 0 to 1; the waiting factor weight = waiting base weight + θ × adjustment range, the distance factor weight = distance base weight + θ × adjustment range, and the price factor weight = 1 - (waiting factor weight + distance factor weight); where the waiting base weight, distance base weight, and adjustment range are set according to the actual situation, and all range from 0 to 1.

[0009] More preferably, the charging and discharging strategy data of all electric heavy-duty trucks are placed in a shared storage area, and a scheduling strategy for the electric heavy-duty trucks is generated based on the strategy data in the shared storage area through a parallel computing framework; the parallel computing framework contains multiple computing units; each unit is responsible for optimizing the charging and discharging strategy data of one electric heavy-duty truck. Each calculation unit detects whether the target battery swapping station of this electric heavy-duty truck is overloaded or has time conflict due to the charging and discharging strategies of other electric heavy-duty trucks in the current iteration round, based on the current charging and discharging strategy data of other electric heavy-duty trucks in the current iteration round. If there is a conflict, the strategy score is recalculated. All computing units simultaneously read the strategy data from the previous iteration of other computing units from the shared memory area through the parallel computing framework, update it synchronously, and write new charging and discharging strategy data into the shared memory area. After all computing units have completed their calculations, the results are aggregated to form the scheduling strategy for the next iteration round.

[0010] More preferably, the method for implementing the branch cutting is as follows: All charging and discharging strategy data for electric heavy-duty trucks in the current iteration are divided into an executable group and a pending group based on the service capacity of the battery swapping station. The service capacity of the battery swapping station refers to the maximum number of electric heavy-duty trucks that a single battery swapping station can serve per unit time. The executable group refers to the set of vehicle strategies that meet the current service capacity limit of the battery swapping station, and their charging and discharging strategy data can be executed directly. The pending group refers to the set of charging and discharging strategy data that cannot be executed temporarily due to insufficient service capacity of the battery swapping station or time period conflicts. Remove charging / discharging strategy data that conflicts with the executable group from the pending group to form an updated pending group; generate constraint information based on the removal results, and apply the constraint information to the strategy update process of all vehicles in the next iteration: The prohibited selection strategy rule refers to the constraint condition that explicitly prohibits electric heavy trucks from selecting certain combinations of battery swapping stations and time periods; The remaining service capacity threshold refers to the maximum number of additional vehicles that a battery swapping station can still accept in each time period.

[0011] More preferably, each discharge strategy data in the pending group is traversed, and it is checked whether the battery swapping station and time period requested by the charge / discharge strategy data overlap with the strategy in the executable group; if there is an overlap and the superposition exceeds the service capacity of the battery swapping station, the charge / discharge strategy data is marked as a conflicting strategy and removed from the pending group; after removal, an updated pending group is formed.

[0012] More preferably, the method for generating the constraint information includes: All conflicting charging and discharging strategy data in the removal results are categorized according to the applied battery swapping station number and charging period in their charging and discharging strategy data; multiple conflicting strategies occurring at the same battery swapping station during the same period are statistically analyzed, and different conflict type labels are marked according to the frequency and degree of overlap to form a conflict type statistics table; the type labels include high-frequency complete conflict type, time-extended conflict type, capacity over-limit conflict type, and low-frequency occasional conflict type. Based on the charging and discharging strategy data of electric heavy trucks in the executable group, calculate the service reserve of each battery swapping station in different time periods, subtract the service reserve of each time period from the maximum service capacity of the battery swapping station, and generate a reserve table of battery swapping station-time period dimension. Configure the corresponding prohibition selection policy rule according to the conflict type flag, and convert the service margin value into the remaining service margin threshold; combine the prohibition selection policy rule and the remaining service margin threshold into constraint information.

[0013] More preferably, the method for configuring the corresponding prohibition selection policy rule according to the conflict type flag is as follows: Based on the conflict type marker, the corresponding basic policy prohibition condition is extracted from the preset dynamic rule base; the preset dynamic rule base refers to the pre-established set of correspondences between conflict types and basic restriction conditions, which stores the initial processing rules corresponding to various conflict modes; the basic policy prohibition condition refers to the basic restriction requirements pre-set for a specific conflict type. Based on real-time traffic flow data, the prohibition conditions of the basic policy are adjusted, and a prohibition rule framework is constructed according to the adjusted basic policy prohibition conditions; the real-time traffic flow data includes vehicle traffic density and speed data collected by road sensors; the prohibition rule framework refers to the constraint condition structure after real-time data correction. Within the aforementioned prohibition rule framework, when a conflict is detected between adjacent time periods, the range of prohibited time periods is expanded; the adjacent time periods refer to the time intervals directly preceding and following the current prohibited time period. By integrating the aforementioned prohibition rule framework and the expanded prohibited time period range, prohibition selection strategy rules are generated.

[0014] More preferably, the method for adjusting the prohibition conditions of the basic policy based on real-time traffic flow data is as follows: Based on the real-time traffic flow data, the traffic congestion index for the current time period is obtained, and the constraint degree of the basic prohibition conditions is adjusted according to the index size by a preset ratio. The constraint strength = basic strength × congestion index. The method to expand the range of the disabled time period is as follows: Analyze the prohibited time periods in the prohibited rule framework and check whether there are historical conflict records between the periods before and after them; if so, expand the original prohibited time period range according to the expansion ratio to form an expanded interval that includes the buffer time. The revised prohibition rule framework is integrated with the extended time period intervals to generate detailed prohibition entries; each rule clearly indicates the applicable battery swapping station, time period range, and constraint strength.

[0015] More preferably, the method for converting the service surplus value into a remaining service surplus threshold is as follows: Calculate the difference between the service margin and the preset buffer margin, and multiply the difference by the dynamic attenuation coefficient to generate the remaining service margin threshold; The dynamic attenuation coefficient is a correction factor calculated based on the real-time operating status data of the battery swapping station equipment. The calculation formula is: Dynamic attenuation coefficient = Basic attenuation coefficient - Equipment aging rate × Service life + Recent maintenance gain × Number of maintenance.

[0016] More preferably, the preset convergence condition is defined as the maximum change in the charging and discharging strategy data of all electric heavy trucks between adjacent iterations being less than a set threshold, wherein the change is calculated by the absolute value of the strategy score difference.

[0017] This invention also proposes a collaborative optimization system for charging and discharging scheduling of electric heavy-duty trucks using the above method, comprising an acquisition module, a first generation module, a second generation module, an update module, a cutting module, and a combination module: The acquisition module is used to acquire battery swapping price-sensitive data, historical queuing data, and real-time location data of electric heavy trucks.

[0018] The first generation module is used to generate charging and discharging strategy data for each electric heavy truck based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data.

[0019] The second generation module is used to generate a scheduling strategy based on the charging and discharging strategy data of all electric heavy trucks through a parallel computing framework.

[0020] The update module is used to iteratively update the charging and discharging strategy data of each electric heavy truck based on the scheduling strategy.

[0021] The cutting module is used to perform branch cutting processing on the charging and discharging strategy data of all electric heavy-duty trucks in the current iteration round when the updated charging and discharging strategy data of all electric heavy-duty trucks does not meet the preset convergence conditions, generate constraint information, and limit the update process of the next iteration round based on the constraint information until the charging and discharging strategy data of all electric heavy-duty trucks meets the preset convergence conditions.

[0022] The combination module is used to combine the charging and discharging strategy data of all electric heavy trucks at the convergence point into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy trucks.

[0023] The present invention also proposes a computing device having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0024] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention establishes a multi-dimensional decision-making foundation to ensure that the scheduling scheme takes into account economy, timeliness, and spatial distribution characteristics. It forms personalized charging path planning, allowing each vehicle's strategy to reflect its operational characteristics and preferences. It enables rapid collaborative computation of large-scale vehicle strategies, improving response speed in complex scenarios. Through a dynamic feedback mechanism, it continuously optimizes strategies, gradually approaching the global optimum. It intelligently identifies and eliminates strategy conflicts, providing precise optimization direction constraints for the next iteration. It outputs an executable balanced scheduling scheme, achieving a balance between vehicle autonomous selection and overall system optimization.

[0025] 2. This invention employs a branching process. First, it divides the strategy data into a schedulable group and a pending group based on the battery swapping station capacity. Strategies in the pending group that conflict with the schedulable group are removed. Based on the removal results, constraint information containing prohibition rules and margin thresholds is generated. This processing mechanism effectively identifies and isolates strategy conflicts, guides the optimization direction through dynamic constraints, and achieves a reasonable allocation of vehicle charging demand while ensuring the service capacity of battery swapping stations, thereby improving the feasibility and execution efficiency of the scheduling scheme. Attached Figure Description

[0026] Figure 1 This is a flowchart illustrating the collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks provided by the present invention. Figure 2 This is a flowchart of a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electric heavy-duty truck charging and discharging scheduling collaborative optimization system provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a computing device provided in an embodiment of the present invention. Detailed Implementation

[0027] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0028] Existing electric heavy-duty truck charging and discharging scheduling schemes mostly adopt centralized optimization models, relying on static linear programming to generate global scheduling plans, without considering the game relationship between vehicle autonomous decision-making and the dynamic load of battery swapping stations. This fixed-time-window scheduling method requires frequent recalculation of the global plan when there are sudden changes in vehicle demand or fluctuations in battery swapping station capacity, resulting in increased response delays. Furthermore, it is difficult to balance individual demand with overall system efficiency, and local resource idleness or congestion is prone to occur, thus restricting the adaptability and real-time performance of the scheduling system.

[0029] To address the aforementioned issues, this application proposes a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks. This method dynamically coordinates vehicle autonomous strategies and system constraints through a parallel computing framework. First, it collects vehicle price sensitivity, queuing history, and real-time location data to generate charging and discharging strategy data. Then, through iterative game theory optimization and branch-cutting processing, it dynamically generates constraint information in each update round to guide strategy convergence. Its core lies in combining centralized scheduling with distributed decision-making, using a branch-cutting algorithm to eliminate strategy conflicts in real time and quantify resource reserve thresholds, thus preserving vehicle autonomous selection space while ensuring load balancing at battery swapping stations. This method effectively solves the problems of sluggish response and insufficient adaptability of static scheduling schemes, achieving efficient matching of vehicle demand and system resources in dynamic scenarios, and improving the feasibility and execution efficiency of scheduling schemes.

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0031] The present invention proposes the following technical solution: This invention proposes a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks, such as... Figure 1 As shown: Acquire battery swapping price-sensitive data, historical queuing data, and real-time location data for electric heavy-duty trucks; the battery swapping price-sensitive data includes the maximum acceptable price set by the vehicle owner. Price factor, waiting factor, and distance factor are calculated based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data, respectively, and then normalized. Dynamic weights are assigned to the three factors according to the current traffic conditions, and the weighted sum is used to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate charging and discharging strategy data. The charging and discharging strategy data includes the target battery swapping station and the charging time period. The method for generating charge / discharge strategy data is as follows: The price factor is the ratio of the highest acceptable price in the battery swapping price-sensitive data to the price quoted by the battery swapping station; The waiting factor is the ratio of the tolerable waiting time of car owners in the historical queuing data to the actual estimated queuing time; the tolerable waiting time of car owners is the historical queuing waiting time of car owners. The distance factor is the reciprocal of the distance from the vehicle to the battery swapping station; The price factor, the waiting factor, and the distance factor are normalized, dynamic weights are assigned according to the current traffic conditions, and finally the weighted sum is obtained to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate strategy data. The method for setting dynamic weights is as follows: Based on the traffic condition data, the traffic congestion index is set to θ, ranging from 0 to 1; the waiting factor weight = waiting base weight + θ × adjustment range, the distance factor weight = distance base weight + θ × adjustment range, and the price factor weight = 1 - (waiting factor weight + distance factor weight); where the waiting base weight, distance base weight, and adjustment range are set according to the actual situation, and all range from 0 to 1.

[0032] Based on the charging and discharging strategy data of all electric heavy trucks after each iteration, the scheduling strategy of electric heavy trucks is updated accordingly. All charging and discharging strategy data of electric heavy-duty trucks are placed in a shared storage area. Based on the strategy data in the shared storage area, a scheduling strategy for electric heavy-duty trucks is generated through a parallel computing framework. The parallel computing framework contains multiple computing units. Each unit is responsible for optimizing the charging and discharging strategy data of one electric heavy-duty truck. Each calculation unit detects whether the target battery swapping station of this electric heavy-duty truck is overloaded or has time conflict due to the charging and discharging strategies of other electric heavy-duty trucks in the current iteration round, based on the current charging and discharging strategy data of other electric heavy-duty trucks in the current iteration round. If there is a conflict, the strategy score is recalculated. All computing units simultaneously read the strategy data from the previous iteration of other computing units from the shared memory area through the parallel computing framework, update it synchronously, and write new charging and discharging strategy data into the shared memory area. After all computing units have completed their calculations, the results are aggregated to form the scheduling strategy for the next iteration round.

[0033] Based on the updated scheduling strategy, the charging and discharging strategy data of each electric heavy truck is iteratively updated. If, after each iteration, the charging and discharging strategy data of an electric heavy truck does not meet the preset convergence condition, then the charging and discharging strategy data of all electric heavy trucks in the current iteration round is branched and processed to generate constraint information. Based on the constraint information, the update process of the next iteration round is limited until the charging and discharging strategy data of all electric heavy trucks meet the preset convergence condition. The method for implementing the branch cutting is as follows: All charging and discharging strategy data for electric heavy-duty trucks in the current iteration are divided into an executable group and a pending group based on the service capacity of the battery swapping station. The service capacity of the battery swapping station refers to the maximum number of electric heavy-duty trucks that a single battery swapping station can serve per unit time. The executable group refers to the set of vehicle strategies that meet the current service capacity limit of the battery swapping station, and their charging and discharging strategy data can be executed directly. The pending group refers to the set of charging and discharging strategy data that cannot be executed temporarily due to insufficient service capacity of the battery swapping station or time period conflicts. Remove charging / discharging strategy data that conflicts with the executable group from the pending group to form an updated pending group; generate constraint information based on the removal results, and apply the constraint information to the strategy update process of all vehicles in the next iteration: The prohibited selection strategy rule refers to the constraint condition that explicitly prohibits electric heavy trucks from selecting certain combinations of battery swapping stations and time periods; The remaining service capacity threshold refers to the maximum number of additional vehicles that a battery swapping station can still accept in each time period.

[0034] Iterate through each discharge strategy data in the pending group and check whether the battery swapping station and time period requested by the charge / discharge strategy data overlap with the strategy in the executable group; if there is an overlap and the superposition exceeds the service capacity of the battery swapping station, then mark the charge / discharge strategy data as a conflicting strategy and remove it from the pending group; after removal, an updated pending group is formed.

[0035] The method for generating the constraint information includes: All conflicting charging and discharging strategy data in the removal results are categorized according to the applied battery swapping station number and charging period in their charging and discharging strategy data; multiple conflicting strategies occurring at the same battery swapping station during the same period are statistically analyzed, and different conflict type labels are marked according to the frequency and degree of overlap to form a conflict type statistics table; the type labels include high-frequency complete conflict type, time-extended conflict type, capacity over-limit conflict type, and low-frequency occasional conflict type. Based on the charging and discharging strategy data of electric heavy trucks in the executable group, calculate the service reserve of each battery swapping station in different time periods, subtract the service reserve of each time period from the maximum service capacity of the battery swapping station, and generate a reserve table of battery swapping station-time period dimension. Configure the corresponding prohibition selection policy rule according to the conflict type flag, and convert the service margin value into the remaining service margin threshold; combine the prohibition selection policy rule and the remaining service margin threshold into constraint information.

[0036] The method for configuring the corresponding prohibition selection policy rule based on the conflict type flag is as follows: Based on the conflict type marker, the corresponding basic policy prohibition condition is extracted from the preset dynamic rule base; the preset dynamic rule base refers to the pre-established set of correspondences between conflict types and basic restriction conditions, which stores the initial processing rules corresponding to various conflict modes; the basic policy prohibition condition refers to the basic restriction requirements pre-set for a specific conflict type. Based on real-time traffic flow data, the prohibition conditions of the basic policy are adjusted, and a prohibition rule framework is constructed according to the adjusted basic policy prohibition conditions; the real-time traffic flow data includes vehicle traffic density and speed data collected by road sensors; the prohibition rule framework refers to the constraint condition structure after real-time data correction. Within the aforementioned prohibition rule framework, when a conflict is detected between adjacent time periods, the range of prohibited time periods is expanded; the adjacent time periods refer to the time intervals directly preceding and following the current prohibited time period. By integrating the aforementioned prohibition rule framework and the expanded prohibited time period range, prohibition selection strategy rules are generated.

[0037] The method for adjusting the prohibition conditions of the basic policy based on real-time traffic flow data is as follows: Based on the real-time traffic flow data, the traffic congestion index for the current time period is obtained, and the constraint degree of the basic prohibition conditions is adjusted according to the index size by a preset ratio. The constraint strength = basic strength × congestion index. The method to expand the range of the disabled time period is as follows: Analyze the prohibited time periods in the prohibited rule framework and check whether there are historical conflict records between the periods before and after them; if so, expand the original prohibited time period range according to the expansion ratio to form an expanded interval that includes the buffer time. The revised prohibition rule framework is integrated with the extended time period intervals to generate detailed prohibition entries; each rule clearly indicates the applicable battery swapping station, time period range, and constraint strength.

[0038] The method for converting the service reserve value into a remaining service reserve threshold is as follows: Calculate the difference between the service margin and the preset buffer margin, and multiply the difference by the dynamic attenuation coefficient to generate the remaining service margin threshold; The dynamic attenuation coefficient is a correction factor calculated based on the real-time operating status data of the battery swapping station equipment. The calculation formula is: Dynamic attenuation coefficient = Basic attenuation coefficient - Equipment aging rate × Service life + Recent maintenance gain × Number of maintenance.

[0039] The charging and discharging strategy data of all electric heavy-duty trucks at the convergence point are combined into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy-duty trucks.

[0040] The preset convergence condition is defined as the maximum change in the charging and discharging strategy data of all electric heavy trucks between adjacent iterations being less than a set threshold, wherein the change is calculated by the absolute value of the strategy score difference.

[0041] This invention also proposes a collaborative optimization system for charging and discharging scheduling of electric heavy-duty trucks using the above method, comprising an acquisition module, a first generation module, a second generation module, an update module, a cutting module, and a combination module: The acquisition module is used to acquire battery swapping price-sensitive data, historical queuing data, and real-time location data of electric heavy trucks.

[0042] The first generation module is used to generate charging and discharging strategy data for each electric heavy truck based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data.

[0043] The second generation module is used to generate a scheduling strategy based on the charging and discharging strategy data of all electric heavy trucks through a parallel computing framework.

[0044] The update module is used to iteratively update the charging and discharging strategy data of each electric heavy truck based on the scheduling strategy.

[0045] The cutting module is used to perform branch cutting processing on the charging and discharging strategy data of all electric heavy-duty trucks in the current iteration round when the updated charging and discharging strategy data of all electric heavy-duty trucks does not meet the preset convergence conditions, generate constraint information, and limit the update process of the next iteration round based on the constraint information until the charging and discharging strategy data of all electric heavy-duty trucks meets the preset convergence conditions.

[0046] The combination module is used to combine the charging and discharging strategy data of all electric heavy trucks at the convergence point into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy trucks.

[0047] The present invention also proposes a computing device having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0048] Example 1 This invention proposes a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks, such as... Figure 2 As shown, the specific steps are as follows.

[0049] Step 101: Obtain battery swapping price-sensitive data, historical queuing data, and real-time location data for electric heavy-duty trucks.

[0050] In step 101, the battery swapping price-sensitive data represents data reflecting the electric heavy-duty truck's acceptance of battery swapping service prices, including the maximum acceptable price set by the vehicle owner and the frequency of historical selection of low-priced battery swapping stations. Historical queuing data represents the historical record of the electric heavy-duty truck's queuing time at each battery swapping station. Real-time location data represents the electric heavy-duty truck's current geographical location information, used to calculate the distance and time to reach each battery swapping station.

[0051] In this embodiment, the real-time location coordinates of the electric heavy-duty truck are collected via an on-board terminal, and historical battery swapping records of the vehicle are retrieved from the operation platform database to extract price-sensitive features and queue waiting time data. These three types of data are associated and stored according to vehicle identification, forming a complete vehicle status dataset. Location data is updated in real-time via a satellite positioning module, while price and queue waiting data are obtained on demand from a cloud database.

[0052] For example, electric heavy truck A is currently traveling on a highway, and its onboard GPS obtains its latitude and longitude coordinates as real-time location data. The platform retrieves A's past 10 battery swap records, showing that 80% of the time it chose battery swap stations with prices below 6 yuan, generating price-sensitive data. Simultaneously, the average queuing time for A at various battery swap stations is extracted as 40 minutes as historical queuing data. These three types of data are stored in association with the vehicle ID.

[0053] Step 102: Based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data, generate charging and discharging strategy data for each electric heavy truck.

[0054] In step 102, the charging and discharging strategy data represents a personalized vehicle plan that includes decision-making factors such as the target battery swapping station and the charging period.

[0055] In this embodiment, based on the three types of data obtained in step 101, the following are calculated: first, the price factor is the ratio of the vehicle's acceptable price to the battery swapping station's quoted price; the waiting factor is the ratio of the vehicle owner's tolerable waiting time to the actual estimated queue length; and the distance factor is the reciprocal of the distance from the vehicle to the battery swapping station. Then, the three factors are normalized, dynamically weighted according to the current traffic conditions, and finally, a weighted sum is calculated to obtain the strategy score for each battery swapping station. The battery swapping station with the highest score and the corresponding time period are selected to generate strategy data.

[0056] Step 103: Based on the charging and discharging strategy data of all electric heavy trucks, generate a scheduling strategy through a parallel computing framework.

[0057] In step 103, the parallel computing framework represents a distributed system architecture that simultaneously processes multi-vehicle strategy calculations. The scheduling strategy refers to a global allocation scheme generated after coordinating the charging and discharging strategy data of all electric heavy-duty trucks through the parallel computing framework. Essentially, it is the set of optimal charging choices for each vehicle in the current game round, including the target battery swapping station identifier, charging time range, and corresponding strategy score for each vehicle, used to guide subsequent iterative updates. This strategy ensures initial coordination between individual decisions and overall system constraints through real-time data synchronization and conflict detection among parallel computing units.

[0058] In this embodiment, the strategy data for each vehicle is distributed to different computing units, with each unit responsible for optimizing the strategy for one vehicle. Units synchronize the latest strategies through a shared storage area, and each unit recalculates its optimal strategy based on the current strategies of other vehicles. After all units have completed their calculations, the results are aggregated to form a new generation scheduling strategy.

[0059] Specifically, the computing unit detects whether the target battery swapping station for this vehicle is experiencing capacity overruns or time slot conflicts due to other vehicles' strategies. If a conflict exists, the score is recalculated. In a preferred embodiment of this invention, the recalculation employs the optimal response strategy from game theory, i.e., while keeping other vehicles' strategies constant, it seeks the charging scheme that maximizes the vehicle's utility. During the calculation, factors such as battery swapping station capacity constraints and time slot overlaps are considered, and conflicts are avoided by adjusting charging times or changing battery swapping stations. All computing units simultaneously read the strategy data from the previous iteration of other computing units from the shared memory area through a parallel computing framework, perform synchronous updates, and write the new strategy into the shared memory area.

[0060] For example, the strategy data for electric heavy trucks A, B, and C are distributed to three computing units. After reading the strategies for B and C from the shared area, Unit 1 recalculates the score for A charging at station X from 14:00 to 15:00, which is reduced to 0.6, and adjusts it to a score of 0.8 for charging at station X from 15:00 to 16:00. After the three units complete the calculations synchronously, a new generation of scheduling strategy is generated.

[0061] Step 104: Based on the updated scheduling strategy, iteratively update the charging and discharging strategy data for each electric heavy truck.

[0062] In step 104, the iteration round indicates the number of loops for policy updates.

[0063] In this embodiment, the new strategy generated in step 103 is used as input, and the strategy calculation and update process is repeated. After each iteration, the change magnitude of all vehicle strategies is checked. If the maximum change is less than a threshold, convergence is determined; otherwise, the next iteration begins.

[0064] For example, the strategy for electric heavy truck A is adjusted from 14:00-15:00 to 15:00-16:00 at station X, a change of 0.2; the strategy for truck B remains unchanged; and the strategy for truck C is adjusted from 15:00-16:00 to 16:00-17:00 at station Y, a change of 0.15. The maximum change of 0.2 is greater than the threshold of 0.1, requiring further iteration.

[0065] Step 105: If, after each iteration update, the charging and discharging strategy data of electric heavy-duty trucks does not meet the preset convergence conditions, then perform branch cutting processing on the charging and discharging strategy data of all electric heavy-duty trucks in the current iteration round to generate constraint information. Based on the constraint information, limit the update process of the next iteration round until the charging and discharging strategy data of all electric heavy-duty trucks meet the preset convergence conditions.

[0066] In step 105, branch cutting represents a method of partitioning the policy space and removing infeasible solutions. Constraint information represents rules and thresholds that limit the policy search space. The preset convergence condition refers to the threshold standard used to determine whether the iterative optimization process has terminated. Specifically, it is defined as the maximum change in the charging and discharging policy data of all electric heavy trucks between adjacent iterations being less than a set threshold, where the change is calculated using the absolute value of the policy score difference. When the adjustment range of all vehicle policies is below this threshold, the system is considered to have reached a Nash equilibrium state, at which point each vehicle's policy has no unilateral motivation to adjust given the policies of other vehicles.

[0067] In this embodiment, the core purpose of branching is to accelerate overall convergence and avoid invalid searches, rather than to reduce the size of the optimization problem. Specifically, the charging and discharging strategy data of all electric heavy-duty trucks in the current iteration round are divided into executable and pending groups based on the capacity of the battery swapping stations. Strategies in the pending group that conflict with those in the executable group are removed, and two types of constraints are generated based on the removal results: prohibiting combinations of specific battery swapping station time periods and setting the maximum acceptable number of vehicles for each time period. These two types of constraints will be applied to the strategy update process for all vehicles in the next round.

[0068] For example, if the demand for electric heavy trucks A and B at station X from 15:00 to 16:00 has reached the limit of 2 vehicles, the strategy for vehicle C in the pending group during the same time period is removed, and the following constraint is generated: No new vehicles are allowed at station X from 15:00 to 16:00, and only 1 vehicle is allowed from 16:00 to 17:00.

[0069] Step 106: Combine the charging and discharging strategy data of all electric heavy trucks at the convergence point into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy trucks.

[0070] In step 106, the balanced scheduling result represents the optimal strategy combination that satisfies all constraints. The charging and discharging scheduling instruction set represents the specific execution plan including battery swapping stations, charging piles, and time periods.

[0071] In this embodiment of the application, the converged vehicle strategies are sorted by time period, and specific charging piles are allocated according to the status of the battery swapping station, generating a complete instruction set containing vehicle ID, battery swapping station, pile number, and time period.

[0072] For example, it was finally determined that electric heavy truck A would charge at station X, pile 1 from 15:00 to 16:00, truck B would charge at station X, pile 2 from 15:30 to 16:30, and truck C would charge at station Y, pile 1 from 16:00 to 17:00, thus forming an executable scheduling instruction.

[0073] This method generates an initial strategy by dynamically collecting personalized vehicle data, accelerates the game process using parallel computing, and eliminates conflicts in real time by combining branch-based processing, ultimately outputting a balanced scheduling scheme. It achieves a balance between autonomous selection by electric heavy-duty trucks and overall system optimization, improving the resource utilization efficiency of battery swapping stations and the feasibility of the scheduling scheme.

[0074] To address the strategy conflict issue during the charging and discharging scheduling of electric heavy-duty trucks and further improve the feasibility of the scheduling scheme, in some embodiments, step 105 involves branching the charging and discharging strategy data of all electric heavy-duty trucks in the current iteration round to generate constraint information, including: Step 201: Divide the charging and discharging strategy data of all electric heavy trucks in the current iteration according to the service capacity of the battery swapping station, and form an executable group and a pending group.

[0075] In step 201, the service capacity of a battery swapping station refers to the maximum number of electric heavy-duty trucks that a single battery swapping station can serve per unit of time. This data is calculated based on the hardware configuration (number / power of charging piles) and operational strategy (battery swapping duration standard) of the battery swapping station. The executable group refers to the set of vehicle strategies that meet the current service capacity limitations of the battery swapping station, and their strategy data can be executed directly. The pending group refers to the set of strategies that cannot be executed temporarily due to insufficient battery swapping station capacity or time slot conflicts, and require further screening and processing.

[0076] In this embodiment, the strategy data of all electric heavy-duty trucks in the current iteration is first read and grouped according to the time-period capacity limit of each battery swapping station. For each time period of each battery swapping station, the number of vehicles requesting charging is counted, and the strategies for vehicles that do not exceed the capacity limit are assigned to the executable group, while those exceeding the limit are assigned to the pending group.

[0077] Step 202: Remove the charging and discharging strategy data of electric heavy trucks that conflict with the executable group from the pending group.

[0078] In step 202, the removal operation is the process of removing these strategies from the pending group.

[0079] In this embodiment, each policy in the pending group is traversed, and it is checked whether its requested battery swapping station and time period overlap with policies in the executable group. If there is an overlap and the combined effect exceeds the battery swapping station capacity, the policy is marked as a conflicting policy and removed from the pending group. After removal, an updated pending group is formed.

[0080] Step 203: Based on the removal results, generate constraint information, which includes rules for prohibiting selection strategies and a threshold for remaining service capacity.

[0081] In step 203, the prohibition selection strategy rule refers to the constraint condition that explicitly prohibits vehicles from selecting certain combinations of battery swapping station time periods. The remaining service capacity threshold refers to the upper limit of the number of additional vehicles that a battery swapping station can still accept in each time period.

[0082] In this embodiment, the combination of high-frequency conflicting battery swapping station time slots is analyzed based on the conflict removal strategy to generate prohibition selection rules. Simultaneously, the occupancy status of schedulable groups in each battery swapping station time slot is statistically analyzed, and the remaining capacity is obtained by subtracting the allocated number from the total capacity, which is then converted into threshold constraints. Finally, the two types of constraints are combined into complete constraint information.

[0083] Here is a specific example: In the high-speed logistics trunk line scenario, the current iteration includes charging and discharging strategy data for electric heavy trucks A, B, and C. The maximum service capacity of battery swapping station X during the 15:00-16:00 period is 2 vehicles. First, the data is divided according to the battery swapping station capacity: Vehicle A requests charging at station X from 15:00-16:00, and Vehicle B requests charging at station X from 15:30-16:30. The overlapping demand during 15:30-16:00 is 2 vehicles, which does not exceed the capacity limit. Therefore, the strategies for A and B are classified into the executable group. Vehicle C also requests charging at station X from 15:00-16:00. After being combined with the dispatchable group, the total demand during 15:00-16:00 reaches 3 vehicles, exceeding the capacity limit. Therefore, the strategy for C is classified into the pending group. Next, conflicting policies are removed from the pending group: It was detected that the policy of vehicle C overlaps with the policies of A and B in the schedulable group during the 15:00-16:00 period at station X, therefore, policy C is removed from the pending group. Finally, constraint information is generated: Based on the removal results, a new entry "No new charging requests allowed at station X from 15:00-16:00" is added to the prohibited selection policy rules; the remaining service capacity threshold is calculated as follows: 2 vehicles have been allocated at station X from 15:00-16:00, so the remaining service capacity threshold = maximum capacity 2 vehicles - 2 allocated vehicles = 0 vehicles; 1 vehicle B has been allocated during the 16:00-17:00 period, so the remaining threshold = 2 - 1 = 1 vehicle. The formula for calculating the remaining service capacity threshold is: Remaining threshold = Station maximum capacity - Number of allocated vehicles, where the station maximum capacity is determined by the number of charging piles. In this example, station X is equipped with 2 charging piles, so the maximum capacity is 2 vehicles. The final generated constraint information will be used to guide the policy update in the next iteration, ensuring that the scheduling scheme conforms to the actual service capacity of the battery swapping station.

[0084] In this embodiment of the application, the branch cutting process generates precise constraints to guide subsequent optimization by dynamically identifying and eliminating policy conflicts. This avoids overload operation of the battery swapping station while retaining reasonable policy adjustment space, effectively improving the rationality and execution efficiency of the scheduling scheme.

[0085] To further improve the accuracy and usability of constraint information, in some embodiments, step 203: generating constraint information based on the removal result includes: Step 301: Mark all conflict strategy data in the removed pending group according to the battery swapping station identifier and time period range to indicate the conflict type.

[0086] In step 301, the conflict strategy data refers to the electric heavy-duty truck charging and discharging strategy data that conflict with the executable group and has been removed from the pending group after branch cutting. This data includes charging request schemes that are deemed unexecutable due to the service capacity of the battery swapping station exceeding the limit or time period overlap. Specifically, this manifests as strategy records where the requested charging time period overlaps with the dispatched vehicles, and the combined effect exceeds the maximum service capacity of the battery swapping station. The battery swapping station identifier (e.g., station number) in the battery swapping station identifier and time period range refers to the unique code that identifies the battery swapping station, derived from the target battery swapping station information recorded in the electric heavy-duty truck charging and discharging strategy data. The time period range refers to the planned charging time interval in the strategy (e.g., 14:00-15:30), derived from the charging time period field in the charging and discharging strategy data. Conflict type labeling refers to the process of classifying and labeling the removed conflict strategies according to the battery swapping station and time period, used to identify high-frequency conflict patterns.

[0087] In this embodiment, all strategies removed from the pending group are first organized and categorized according to their applied battery swapping station number and charging time period. Multiple conflicting strategies occurring at the same battery swapping station during the same time period are statistically analyzed, and different type labels are assigned based on conflict frequency and overlap, forming a conflict type statistics table. These type labels include high-frequency complete conflict type (high frequency and complete overlap of conflicts at the same battery swapping station during the same time period), time-extended conflict type (partial overlap of time periods triggering cascading conflicts in adjacent time periods), capacity over-limit conflict type (multiple strategies significantly exceeding the battery swapping station's capacity limit after aggregation), and low-frequency occasional conflict type (low conflict frequency and minimal overlap). Each type corresponds to a different processing priority and constraint generation strategy.

[0088] Step 302: Calculate the service capacity already occupied by each battery swapping station at different time periods based on the charging and discharging strategy data of the electric heavy trucks in the executable group.

[0089] In step 302, the charging and discharging strategy data for electric heavy-duty trucks refers to the set of charging and discharging strategies that can be executed immediately after being screened by the service capacity of the battery swapping station. This set is generated by comparing the charging and discharging strategy data of all current electric heavy-duty trucks with the real-time available capacity of the battery swapping station. Strategies not selected are placed in the pending group. Service margin refers to the number of vehicles that the battery swapping station can still serve during a specific time period; its calculation requires statistics on the allocated strategy data. The occupied service margin is equal to the number of confirmed charging requests during that time period. The selection principle for the executable group is to ensure that the combined demand of all strategies within the group does not exceed the real-time available capacity of the battery swapping station for each time period (i.e., the sum of demand capacity ≤ available capacity). During selection, the strategy combination with the fewest conflicts and the highest priority is prioritized.

[0090] In this embodiment, all strategies in the executable group are traversed, and the number of vehicles allocated to each battery swapping station in each time period is counted. The service reserve value for each time period is obtained by subtracting the allocated number from the maximum service capacity of the battery swapping station, and a reserve table of battery swapping station-time period dimension is generated.

[0091] Step 303: Configure the corresponding prohibition selection policy rule according to the conflict type flag, and convert the service surplus into the remaining service surplus threshold.

[0092] In step 303, the prohibition selection policy rule is a restriction condition formulated based on the conflict type, explicitly prohibiting certain combinations of battery swapping station time slots. The remaining service capacity threshold is a service capacity converted into an operable limit value to control new requests.

[0093] In this embodiment, explicit prohibition rule entries are generated based on the high-frequency conflict battery swapping station time periods marked by conflict type. Simultaneously, the calculated service margin is standardized and converted into easily comparable threshold standards, such as "Station X can add a maximum of 0 vehicles between 15:00 and 16:00".

[0094] Step 304: Combine the prohibited selection policy rule and the remaining service margin threshold into constraint information.

[0095] In this embodiment of the application, the generated prohibition rules and margin thresholds are associated and combined according to the dimensions of battery swapping stations and time periods to form a structured constraint information table, ensuring that the two types of constraints can work together when applied.

[0096] Here is a specific example: In the high-speed logistics trunk line scenario, based on the obtained removal results, i.e., after vehicle C's charging strategy at station X from 15:00 to 16:00 is removed due to a conflict with the schedulable group, the specific process of generating constraint information is as follows: First, the conflicting strategy data of the removed vehicles is marked with conflict type. Statistical analysis shows that three similar conflicts have occurred at station X during the 15:00-16:00 period, which is marked as the "peak-hour core conflict" type. Next, based on the strategy data of vehicles A and B in the schedulable group, the occupancy status of station X in each time period is calculated. During the 15:00-16:00 period, 2 vehicles have been occupied, reaching the maximum capacity, so the service reserve = maximum capacity 2 vehicles - 2 allocated vehicles = 0 vehicles. During the 16:00-17:00 period, 1 vehicle B has been allocated, so the service reserve = 2 - 1 = 1 vehicle. The service reserve calculation formula is: Service reserve = station maximum capacity - 2 allocated vehicles. The number of vehicles and the maximum capacity of a station are determined by the number of charging piles. Station X is equipped with 2 charging piles, so its maximum capacity is 2 vehicles. Then, prohibition rules are configured according to the conflict type marking. For the "peak period core conflict" type, a rule is generated that "prohibits new requests from 15:00 to 16:00 to Station X". At the same time, the calculated service reserve is converted into a remaining service reserve threshold, which is set to 0 vehicles from 15:00 to 16:00 and 1 vehicle from 16:00 to 17:00. Finally, the prohibition selection policy rule and the remaining service reserve threshold are combined into complete constraint information, which includes specific items such as "prohibits new requests from 15:00 to 16:00 to Station X with 0 vehicles remaining, and allows 1 new vehicle from 16:00 to 17:00". This constraint information will be used to guide the adjustment of vehicle policies in the next iteration to ensure that the newly generated policies meet the actual service capacity limitations of the battery swapping station.

[0097] In this embodiment of the application, the constraint information generation process forms accurate and feasible scheduling constraints by systematically analyzing conflict patterns and resource reserves. This not only avoids the overload operation of battery swapping stations, but also reserves reasonable space for vehicle strategy adjustments, effectively improving the rationality and execution efficiency of the scheduling scheme.

[0098] To further improve the dynamic adaptability and accuracy of constraint information, in some embodiments, step 303: configuring the corresponding prohibition selection policy rule according to the conflict type label, and converting the service surplus into a remaining service surplus threshold, includes: Step 401: Based on the conflict type marker, extract the corresponding basic policy prohibition conditions from the preset dynamic rule base.

[0099] In step 401, the preset dynamic rule base refers to a pre-established set of correspondences between conflict types and basic constraints, which stores the initial processing rules corresponding to various conflict modes. Basic policy prohibition conditions refer to the basic restrictions pre-set for specific conflict types.

[0100] In this embodiment, the system retrieves a rule base based on the conflict type tag. For example, for the "peak-hour core conflict" type, it extracts the basic prohibition condition "prohibit new requests during the same time period". The retrieval considers the spatiotemporal characteristics and frequency of occurrence of the conflict to match the most suitable basic rule.

[0101] Step 402: Based on real-time traffic flow data, adjust the prohibition conditions of the basic policy, and construct a prohibition rule framework according to the adjusted prohibition conditions of the basic policy.

[0102] In step 402, real-time traffic flow data refers to vehicle density and speed data collected by road sensors from the real-time traffic monitoring system of the traffic management department. The prohibition rule framework refers to the constraint structure after real-time data correction.

[0103] In this embodiment, the traffic congestion index for the current time period is obtained, and the strictness of the basic prohibition conditions is adjusted according to a preset ratio based on the index magnitude. The more congested the traffic, the stronger the constraint is, forming a prohibition rule framework that adapts to real-time traffic conditions. The adjusted constraint strength = basic strength × congestion index.

[0104] Step 403: In the prohibited rule framework, when a conflict is detected between adjacent time periods, the range of prohibited time periods is expanded.

[0105] In step 403, adjacent time periods refer to the time intervals directly preceding and following the current prohibited time period (e.g., when the prohibited time period is 14:00-15:00, the adjacent time periods are 13:30-14:00 and 15:00-15:30). Associated conflicts refer to the extended conflicts that may be caused by the continuous charging demand of vehicles in adjacent time periods.

[0106] In this embodiment, the prohibited time periods within the prohibition rule framework are analyzed, and it is checked whether there are historical conflict records between the prohibited time periods and their adjacent time periods. If so, the original prohibited time period range is expanded proportionally to form an expanded interval that includes a buffer time.

[0107] Specifically, the expansion ratio is set based on historical conflict data. The expansion ratio = basic expansion coefficient × (1 + associated conflict frequency weight), where the basic expansion coefficient is usually set to 0.2 (i.e., 20% of the original time period). The associated conflict frequency weight is dynamically adjusted according to the number of historical conflicts in adjacent time periods (the more conflicts, the higher the weight). For example, if there are 3 recent conflicts in adjacent time periods, the weight can be 0.3, then the total expansion ratio = 0.2 × (1 + 0.3) = 0.26, which means that the original prohibited time period is expanded by 26% in the direction of conflict.

[0108] Step 404: Integrate the aforementioned prohibition rule framework and the expanded prohibited time period range to generate prohibition selection strategy rules.

[0109] In this embodiment, the adjusted prohibition rule framework is integrated with the extended time period interval to generate detailed prohibition entries. Each rule clearly indicates the applicable battery swapping station, time period range, and level of constraint.

[0110] Step 405: Calculate the difference between the service margin and the preset buffer margin, and multiply the difference by the dynamic attenuation coefficient to generate the remaining service margin threshold.

[0111] In step 405, the dynamic attenuation coefficient is a correction factor calculated based on the real-time operating status data of the battery swapping station equipment. It reflects the degree of attenuation of the actual available service capacity of the charging pile. This coefficient is generated by the equipment health scoring system, and the specific calculation formula is: Dynamic attenuation coefficient = Basic attenuation coefficient 1.0 - Equipment aging rate 0.1 × Service life + Recent maintenance gain 0.05 × Maintenance times. The value range is between 0.8 and 1.2, and it is used to dynamically adjust the effective available value of the service margin.

[0112] In this embodiment, after deducting a preset buffer value from the service margin, the remaining capacity is multiplied by an attenuation coefficient reflecting the equipment's health status. The worse the equipment's condition, the smaller the attenuation coefficient, and the stricter the final threshold, ensuring the scheduling scheme is safe and reliable. The specific process is as follows: First, the actual remaining service capacity of the battery swapping station for the current time period is obtained as the service margin. This is then subtracted from the preset buffer margin safety value to obtain the basic difference. This difference is then multiplied by a dynamic attenuation coefficient reflecting the equipment's operating status for correction, ultimately yielding the available service capacity threshold after considering equipment performance degradation. For example: A battery swapping station currently has a service margin of 3 vehicles, a preset buffer margin of 1 vehicle, and an equipment attenuation coefficient of 0.8. Therefore, the remaining service margin threshold = 3 - 1 × 0.8 = 1.6 vehicles, which, after rounding, becomes 1 vehicle.

[0113] Here is a specific example: In the high-speed logistics trunk line scenario, the process of generating refined constraints based on the obtained "peak-hour core conflict" markers and service surplus data is as follows: First, the basic prohibition condition "prohibit new requests during the same time period" corresponding to the "peak-hour core conflict" is extracted from the preset rule base. At this time, the real-time traffic data shows that the current road segment congestion index is 1.3. According to the adjustment formula, the intensity is adjusted as follows: intensity = basic intensity 1.0 × congestion index 1.3 = 1.3, thus constructing the prohibition rule framework with increased intensity. Historical related conflict records are detected in the adjacent time period 14:30-15:00. According to the expansion formula, the extension duration is calculated as follows: duration = basic conflict duration 1 hour × congestion coefficient 0.2 × congestion index 1.3 = 0.26 hours. The disabled period is extended from 15:00-16:00 to 14:44-16:00; after integration, a disabled selection policy rule is generated: "X station 14:44-16:00 disabled new additions, intensity 1.3"; at the same time, the remaining service capacity threshold is calculated. In the 15:00-16:00 period, the service capacity is 0 vehicles, minus the preset buffer capacity of 0.5 vehicles, which is -0.5 vehicles. Multiplying this by the dynamic attenuation coefficient of 0.9, which reflects the equipment status, we get the adjusted threshold = -0.5 × 0.9 = -0.45 vehicles. After rounding, the final remaining service capacity threshold for this period is determined to be 0 vehicles. In the 16:00-17:00 period, the calculated remaining capacity is 1 vehicle, minus the 0.5 vehicle buffer capacity, multiplied by the attenuation coefficient of 0.9, which is 0.45 vehicles. This is rounded to 1 vehicle. The dynamic attenuation coefficient is calculated as follows: Attenuation coefficient = base coefficient 1.0 - equipment aging rate 0.1 × service life 2 years + maintenance gain 0.05 × maintenance times 3 times = 0.85, which is then finely adjusted to 0.9 by adding the real-time performance test results.

[0114] In this embodiment, the scheme dynamically adjusts the constraint strength and intelligently expands the time period range to make the generated prohibition rules more in line with actual road conditions; at the same time, it considers the equipment status and the margin calculation of emergency buffer to ensure the safety and reliability of the threshold. Overall, it achieves the best match between the constraint conditions and the real-time operating environment, effectively improving the adaptability and safety of the scheduling scheme.

[0115] To further improve the accuracy and adaptability of the charging and discharging strategy generation for electric heavy-duty trucks, in some embodiments, step 102: generating charging and discharging strategy data for each electric heavy-duty truck based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data includes: Step 501: Establish a decision factor set including price impact factor, waiting tolerance factor and site distance factor, wherein the price impact factor is associated with the battery swapping price sensitive data, the waiting tolerance factor is associated with the historical queuing data, and the site distance factor is associated with the real-time location data.

[0116] In step 501, the price impact factor is a quantitative indicator reflecting the vehicle's acceptance of battery swapping prices; a higher value indicates greater price sensitivity. The waiting tolerance factor measures the vehicle's tolerance for queuing time; a higher value indicates a less willing vehicle to wait. The site distance factor assesses the ease with which a vehicle can reach a battery swapping station; a higher value indicates greater convenience due to proximity.

[0117] In this embodiment of the application, the acquired raw data is first standardized, and the battery swapping price-sensitive data is converted into a price influence factor, which is calculated as the ratio of the vehicle's acceptable price to the battery swapping station's quoted price; the historical queuing data is converted into a waiting tolerance factor, which is calculated as the ratio of the vehicle owner's acceptable waiting time to the actual estimated queuing time; and the real-time location data is converted into a station distance factor, which is calculated as the ratio of the vehicle's actual distance to the battery swapping station to the maximum acceptable distance.

[0118] Step 502: Based on the traffic status data of the current time period, calculate the dynamic weight coefficient of each factor in the decision factor set.

[0119] In step 502, the traffic status data includes road congestion data and battery swapping station load data. The current time period's traffic status data is obtained by acquiring real-time congestion data collected by road sensors and load data uploaded by battery swapping stations. Dynamic weighting coefficients refer to parameters that dynamically adjust the importance of each decision factor based on real-time traffic conditions; the more congested the traffic, the greater the weight of waiting and distance factors.

[0120] In this embodiment, real-time traffic monitoring system acquires current road condition data, analyzes road congestion levels, and dynamically adjusts the weight ratios of three factors according to a preset weight allocation rule. During congestion, the weights of waiting and distance factors are increased; during smooth traffic, the weight of the price factor is increased. The base weights for waiting and distance, as well as the adjustment range, are set according to actual conditions, all ranging from 0 to 1.

[0121] For example, the dynamic weighting coefficients are quantified using a piecewise function: Let the traffic congestion index be θ (range 0-1), then the waiting factor weight = basic waiting weight + θ × adjustment range, the distance factor weight = basic distance weight + θ × adjustment range, and the price factor weight = 1 - (waiting factor weight + distance factor weight). For example, if the basic waiting weight is 0.3, the basic distance weight is 0.2, and the adjustment range is 0.2, when θ = 0.6, the waiting factor weight = 0.3 + 0.6 × 0.2 = 0.42, the distance factor weight = 0.2 + 0.6 × 0.2 = 0.32, and the price factor weight = 1 - 0.74 = 0.26.

[0122] Step 503: Perform a weighted summation operation on the values ​​of each factor in the decision factor set and the corresponding dynamic weight coefficients to obtain the strategy scoring result, and generate charging and discharging strategy data based on the strategy scoring result.

[0123] In step 503, the values ​​of each factor in the decision factor set are quantitative indicator values ​​generated after standardization and calculation of the original data (battery swapping price-sensitive data, historical queuing data, and real-time location data), rather than directly using the original data itself. Specifically: the price impact factor value is obtained by standardizing the battery swapping price-sensitive data; the waiting tolerance factor value is the result of quantifying and transforming historical queuing data; and the site distance factor value is generated based on real-time location data using a distance calculation model. The strategy score result is a comprehensive evaluation value for each battery swapping station, and the optimal charging and discharging scheme is selected based on the strategy score.

[0124] In this embodiment, the standardized values ​​of the three factors are multiplied by their corresponding dynamic weights and then summed to obtain the final strategy score for each battery swapping station. The battery swapping station and time segment combination with the highest score is selected as the charging and discharging strategy data for that vehicle.

[0125] In this embodiment, the scheme generates a charging and discharging strategy by dynamically weighting multi-dimensional factors, which takes into account both the personalized needs of the vehicle and the real-time traffic conditions, achieving an optimal balance of economy, timeliness and convenience, and effectively improving the scientificity and practicality of the charging and discharging decision of electric heavy trucks.

[0126] To further improve the generation efficiency of electric heavy-duty truck charging and discharging scheduling strategies, in some embodiments, step 103: generating a scheduling strategy based on the charging and discharging strategy data of all electric heavy-duty trucks through a parallel computing framework includes: Step 601: Assign a corresponding parallel computing unit to each electric heavy truck, and synchronously store the charging and discharging strategy data of all electric heavy trucks in the shared storage area. All parallel computing units are used to form a parallel computing framework.

[0127] In step 601, a parallel computing unit refers to a computing module that independently processes the strategy optimization of a single electric heavy-duty truck. Multiple units working simultaneously can improve the overall computing speed. The shared storage area is a public storage space used for centralized storage and synchronization of the latest strategy data for all vehicles, ensuring data consistency across computing units.

[0128] In this embodiment, a dedicated computing unit is first created for each electric heavy-duty truck, and the strategy data for the current iteration of all vehicles is uniformly stored in a shared storage area. Each computing unit corresponds one-to-one with a vehicle, and data synchronization is achieved in real time via a high-speed network connection.

[0129] Step 602: Control each parallel computing unit to obtain the latest charging and discharging strategy data required from the shared memory area, and output a scheduling strategy based on the latest charging and discharging strategy data.

[0130] In step 602, the latest charging and discharging strategy data refers to the latest optimization scheme for all vehicles in the current iteration round.

[0131] In this embodiment, each computing unit obtains the latest strategy data of other vehicles from the shared area, and recalculates the optimal charging scheme based on the charging needs of the vehicles it is responsible for. After the calculation is completed, the result is written back to the shared area, and a new generation scheduling strategy is formed after all units have completed their calculations.

[0132] Example 2 This invention proposes a collaborative optimization system for charging and discharging scheduling of electric heavy-duty trucks, such as... Figure 3 As shown, it includes an acquisition module, an update module, a cutting module, and a combination module: The acquisition module 21 is used to acquire battery swapping price-sensitive data, historical queuing data, and real-time location data of electric heavy trucks.

[0133] The first generation module 22 is used to generate charging and discharging strategy data for each electric heavy truck based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data.

[0134] The second generation module 23 is used to generate a scheduling strategy based on the charging and discharging strategy data of all electric heavy trucks through a parallel computing framework.

[0135] The update module 24 is used to iteratively update the charging and discharging strategy data of each electric heavy truck based on the scheduling strategy.

[0136] The cutting module 25 is used to perform branch cutting processing on the charging and discharging strategy data of all electric heavy trucks in the current iteration round when the updated charging and discharging strategy data of all electric heavy trucks does not meet the preset convergence conditions, generate constraint information, and limit the update process of the next iteration round based on the constraint information until the charging and discharging strategy data of all electric heavy trucks meet the preset convergence conditions.

[0137] The combination module 26 is used to combine the charging and discharging strategy data of all electric heavy trucks at the convergence time into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy trucks.

[0138] Figure 3 The aforementioned electric heavy-duty truck charging and discharging scheduling collaborative optimization system can execute... Figure 2The implementation principle and technical effects of the electric heavy-duty truck charging and discharging scheduling collaborative optimization method described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit performs operations in the electric heavy-duty truck charging and discharging scheduling collaborative optimization system described in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0139] Example 4 In one possible design, Figure 3 The electric heavy-duty truck charging and discharging scheduling collaborative optimization system of the embodiment shown can be implemented as a computing device, such as... Figure 4 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0140] The processing component 32 is used to perform the above. Figure 2 The embodiment describes a collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks.

[0141] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the above-described method.

[0142] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0143] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0144] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0145] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0146] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0147] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 2 The embodiment shown is a collaborative optimization method for charging and discharging scheduling of electric heavy trucks.

[0148] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0149] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0150] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A collaborative optimization method for charging and discharging scheduling of electric heavy-duty trucks, characterized in that, include: Acquire sensitive data on battery swapping prices, historical queuing data, and real-time location data for electric heavy-duty trucks; Battery swapping price-sensitive data includes the maximum acceptable price set by car owners; The price factor, waiting factor, and distance factor are calculated based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data, respectively, and then normalized. The three factors are dynamically weighted according to the current traffic conditions, and the weighted sum is used to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate charging and discharging strategy data. The charging and discharging strategy data includes the target battery swapping station and the charging time period. Based on the charging and discharging strategy data of all electric heavy trucks after each iteration, the scheduling strategy of electric heavy trucks is updated accordingly. Based on the updated scheduling strategy, the charging and discharging strategy data of each electric heavy truck is iteratively updated. If, after each iteration, the charging and discharging strategy data of an electric heavy truck does not meet the preset convergence condition, then the charging and discharging strategy data of all electric heavy trucks in the current iteration round is branched and processed to generate constraint information. Based on the constraint information, the update process of the next iteration round is limited until the charging and discharging strategy data of all electric heavy trucks meet the preset convergence condition. The method for implementing the branch cutting is as follows: All charging and discharging strategy data for electric heavy-duty trucks in the current iteration are divided into an executable group and a pending group based on the service capacity of the battery swapping station. The service capacity of the battery swapping station refers to the maximum number of electric heavy-duty trucks that a single battery swapping station can serve per unit time. The executable group refers to the set of vehicle strategies that meet the current service capacity limit of the battery swapping station, and their charging and discharging strategy data can be executed directly. The pending group refers to the set of charging and discharging strategy data that cannot be executed temporarily due to insufficient service capacity of the battery swapping station or time period conflicts. Remove charge / discharge strategy data that conflicts with executable groups from the pending group to form an updated pending group; Constraint information is generated based on the removal results, and this constraint information is applied to the policy update process of all vehicles in the next iteration: The prohibited selection strategy rule refers to the constraint condition that explicitly prohibits electric heavy trucks from selecting certain combinations of battery swapping stations and time periods; The remaining service capacity threshold refers to the maximum number of additional vehicles that a battery swapping station can still accept in each time period; The method for generating the constraint information includes: All conflicting charging and discharging strategy data in the removal results are categorized according to the applied battery swapping station number and charging period in their charging and discharging strategy data; multiple conflicting strategies occurring at the same battery swapping station during the same period are statistically analyzed, and different conflict type labels are marked according to the frequency and degree of overlap to form a conflict type statistics table; the type labels include high-frequency complete conflict type, time-extended conflict type, capacity over-limit conflict type, and low-frequency occasional conflict type. Based on the charging and discharging strategy data of electric heavy trucks in the executable group, calculate the service reserve of each battery swapping station in different time periods, subtract the service reserve of each time period from the maximum service capacity of the battery swapping station, and generate a reserve table of battery swapping station-time period dimension. Configure the corresponding prohibition selection policy rule according to the conflict type flag, and convert the service margin value into the remaining service margin threshold; combine the prohibition selection policy rule and the remaining service margin threshold into constraint information; The charging and discharging strategy data of all electric heavy-duty trucks at the convergence point are combined into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy-duty trucks.

2. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: The method for generating charge / discharge strategy data is as follows: The price factor is the ratio of the highest acceptable price in the battery swapping price-sensitive data to the price quoted by the battery swapping station; The waiting factor is the ratio of the tolerable waiting time of car owners in the historical queuing data to the actual estimated queuing time. The tolerable waiting time for car owners is based on their historical queuing waiting time. The distance factor is the reciprocal of the distance from the vehicle to the battery swapping station; The price factor, the waiting factor, and the distance factor are normalized, dynamic weights are assigned according to the current traffic conditions, and finally the weighted sum is obtained to obtain the strategy score of each battery swapping station. The battery swapping station and time period with the highest score are selected to generate strategy data. The method for setting dynamic weights is as follows: Based on the traffic data, the traffic congestion index is set to θ, ranging from 0 to 1; the waiting factor weight = waiting base weight + θ × adjustment range, the distance factor weight = distance base weight + θ × adjustment range, and the price factor weight = 1 - (waiting factor weight + distance factor weight).

3. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: All charging and discharging strategy data of electric heavy-duty trucks are placed in a shared storage area. Based on the strategy data in the shared storage area, a scheduling strategy for electric heavy-duty trucks is generated through a parallel computing framework. The parallel computing framework contains multiple computing units. Each unit is responsible for optimizing the charging and discharging strategy data of one electric heavy-duty truck. Each calculation unit detects whether the target battery swapping station of this electric heavy-duty truck is overloaded or has time conflict due to the charging and discharging strategies of other electric heavy-duty trucks in the current iteration round, based on the current charging and discharging strategy data of other electric heavy-duty trucks in the current iteration round. If there is a conflict, the strategy score is recalculated. All computing units simultaneously read the strategy data from the previous iteration of other computing units from the shared memory area through the parallel computing framework, update it synchronously, and write new charging and discharging strategy data into the shared memory area. After all computing units have completed their calculations, the results are aggregated to form the scheduling strategy for the next iteration round.

4. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: Iterate through each discharge strategy data in the pending group and check whether the battery swapping station and time period requested by the charge and discharge strategy data overlap with the strategy in the executable group; If there is overlap and the combined effect exceeds the service capacity of the battery swapping station, the charging and discharging strategy data will be marked as a conflicting strategy and removed from the pending group. After removal, an updated pending group is formed.

5. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: The method for configuring the corresponding prohibition selection policy rule based on the conflict type flag is as follows: Based on the conflict type marker, the corresponding basic policy prohibition condition is extracted from the preset dynamic rule base; the preset dynamic rule base refers to the pre-established set of correspondences between conflict types and basic restriction conditions, which stores the initial processing rules corresponding to various conflict modes; the basic policy prohibition condition refers to the basic restriction requirements pre-set for a specific conflict type. Based on real-time traffic flow data, the prohibition conditions of the basic policy are adjusted, and a prohibition rule framework is constructed according to the adjusted basic policy prohibition conditions; the real-time traffic flow data includes vehicle traffic density and speed data collected by road sensors; the prohibition rule framework refers to the constraint condition structure after real-time data correction. Within the aforementioned prohibition rule framework, when a conflict is detected between adjacent time periods, the range of prohibited time periods is expanded. The adjacent time period refers to the time interval before and after the current disabled time period that is directly continuous with it; By integrating the aforementioned prohibition rule framework and the expanded prohibited time period range, prohibition selection strategy rules are generated.

6. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 5, characterized in that: The method for adjusting the prohibition conditions of the basic policy based on real-time traffic flow data is as follows: Based on the real-time traffic flow data, the traffic congestion index for the current time period is obtained, and the constraint degree of the basic prohibition conditions is adjusted according to the index size by a preset ratio. The constraint strength = basic strength × congestion index. The method to expand the range of the disabled time period is as follows: Analyze the prohibited time periods in the prohibited rule framework and check whether there are historical conflict records between the periods before and after them; if so, expand the original prohibited time period range according to the expansion ratio to form an expanded interval that includes the buffer time. The revised prohibition rule framework is integrated with the extended time period intervals to generate detailed prohibition entries; each rule clearly indicates the applicable battery swapping station, time period range, and constraint strength.

7. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: The method for converting the service reserve value into a remaining service reserve threshold is as follows: Calculate the difference between the service margin and the preset buffer margin, and multiply the difference by the dynamic attenuation coefficient to generate the remaining service margin threshold; The dynamic attenuation coefficient is a correction factor calculated based on the real-time operating status data of the battery swapping station equipment. The calculation formula is: Dynamic attenuation coefficient = Basic attenuation coefficient - Equipment aging rate × Service life + Recent maintenance gain × Number of maintenance.

8. The method for coordinated optimization of charging and discharging scheduling of electric heavy-duty trucks according to claim 1, characterized in that: The preset convergence condition is defined as the maximum change in the charging and discharging strategy data of all electric heavy trucks between adjacent iterations being less than a set threshold, wherein the change is calculated by the absolute value of the strategy score difference.

9. A collaborative optimization system for charging and discharging scheduling of electric heavy-duty trucks using the method of any one of claims 1-8, comprising an acquisition module, a first generation module, a second generation module, an update module, a cutting module, and a combination module, characterized in that: The acquisition module is used to acquire battery swapping price-sensitive data, historical queuing data, and real-time location data of electric heavy trucks; The first generation module is used to generate charging and discharging strategy data for each electric heavy truck based on the battery swapping price-sensitive data, the historical queuing data, and the real-time location data. The second generation module is used to generate a scheduling strategy based on the charging and discharging strategy data of all electric heavy trucks through a parallel computing framework. The update module is used to iteratively update the charging and discharging strategy data of each electric heavy truck based on the scheduling strategy. The cutting module is used to perform branch cutting processing on the charging and discharging strategy data of all electric heavy trucks in the current iteration round when the updated charging and discharging strategy data of all electric heavy trucks does not meet the preset convergence conditions, generate constraint information, and limit the update process of the next iteration round based on the constraint information until the charging and discharging strategy data of all electric heavy trucks meet the preset convergence conditions. The combination module is used to combine the charging and discharging strategy data of all electric heavy trucks at the convergence point into a balanced scheduling result to generate a collaboratively optimized charging and discharging scheduling instruction set, which is used to schedule the charging and discharging behavior of electric heavy trucks.

10. A computing device having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-8.

Images