End-to-end electric vehicle energy transaction pairing method and device and medium thereof

Abstract

The application provides an end-to-end electric vehicle energy transaction pairing method and device and a medium thereof. Real-time vehicle information and transaction request orders of multiple electric vehicles are obtained. Each real-time vehicle information is input into a space-time state prediction model to obtain target vehicle information of each electric vehicle. According to the target vehicle information of each electric vehicle and the transaction request order, multiple target supply vehicles and their expected supply quotes, multiple target demand vehicles and their expected demand quotes are determined. Bidirectional auction bidding is performed on each expected supply quote and each expected demand quote to determine multiple winner supply vehicles and multiple winner demand vehicles. The shortest distance pairing is performed on each winner supply vehicle and each winner demand vehicle to obtain multiple final transaction pairs. Game calculation is performed on the expected supply quote and the expected demand quote of each final transaction pair to generate a clearing transaction price, thereby realizing autonomous and efficient win-win energy transaction.

Description

Technical Field

[0001] This application relates to the field of electric vehicle technology, and in particular to an end-to-end electric vehicle energy trading pairing method, device and medium. Background Technology

[0002] Currently, vehicle-to-vehicle (V2V) energy trading matching primarily relies on centralized solution architectures. Such solutions, such as those based on mixed-integer linear programming (MILP), suffer from the curse of dimensionality when dealing with large-scale concurrent vehicle matching requests, resulting in exponentially increasing solution time and failing to meet the millisecond-level real-time requirements of dynamic transactions. Furthermore, centralized processing forces the collection of full information from all vehicles, posing a serious risk of privacy breaches.

[0003] To reduce solution complexity, some studies have introduced heuristic algorithms such as particle swarm optimization and ant colony optimization. However, these mechanisms have a fatal logical flaw: they rely on a central server to calculate the global comprehensive score and perform greedy forced pairing, treating electric vehicle owners as completely passive and unconscious nodes, ignoring their self-interested nature as independent rational individuals seeking to maximize their own benefits. This command-style scheduling often leads to low user participation. Furthermore, existing models generally focus on improving hardware transmission rates, with overly general scenario settings, lacking deep integration of owner-driven pricing, diverse cost considerations, and real-time incentives, requiring numerous auxiliary parameters for correction in practical deployments. Summary of the Invention

[0004] The purpose of this application is to at least solve one of the technical problems existing in the prior art, and to provide an end-to-end electric vehicle energy trading pairing method, equipment and medium, which realizes autonomous, efficient and win-win energy trading through spatiotemporal prediction, two-way auction and shortest distance pairing.

[0005] To achieve the above objectives, a first aspect of this application proposes an end-to-end electric vehicle energy trading matching method, comprising: Obtain real-time vehicle information and transaction request orders for multiple electric vehicles; The real-time vehicle information is input into the spatiotemporal state prediction model for state reasoning to obtain the target vehicle information of each electric vehicle during the transaction period. Based on the target vehicle information and transaction request orders of each electric vehicle, identify multiple target supply vehicles, the expected supply price of each target supply vehicle, multiple target demand vehicles, and the expected demand price of each target demand vehicle among multiple electric vehicles. A two-way auction is conducted on each expected supply price and each expected demand price to determine multiple winning supply vehicles and multiple winning demand vehicles. The winning supply vehicles and winning demand vehicles are then matched with the shortest distance in spatial location to obtain multiple final trading pairs. For each final trading pair, game theory calculations are performed on the corresponding expected supply and expected demand quotes to generate a settlement price, and a trading execution contract is generated based on the settlement price.

[0006] Furthermore, in some embodiments, real-time vehicle information includes vehicle identification, vehicle current location coordinates, real-time state of charge, and vehicle parking duration; the spatiotemporal state prediction model includes a long short-term memory network layer. Specifically, real-time vehicle information is input into a spatiotemporal state prediction model for state reasoning to obtain target vehicle information for each electric vehicle during the transaction period, including: The vehicle identifier, current location coordinates, real-time state of charge, and parking duration of each electric vehicle are input into the spatiotemporal state prediction model, so that the long short-term memory network layer can perform spatiotemporal feature inference on the activity trajectory of each electric vehicle to obtain the target vehicle information of each electric vehicle during the transaction period.

[0007] Furthermore, in some embodiments, based on the target vehicle information and transaction request orders for each electric vehicle, multiple target supply vehicles, expected supply prices for each target supply vehicle, multiple target demand vehicles, and expected demand prices for each target demand vehicle are determined among the multiple electric vehicles, including: Based on the transaction request orders for each electric vehicle, each electric vehicle is divided into multiple supply vehicles and multiple demand vehicles; Based on the target vehicle information of each supply vehicle and the target vehicle information of each demand vehicle, multiple target supply vehicles are selected from multiple supply vehicles, and multiple target demand vehicles are selected from multiple demand vehicles. Based on the target vehicle information and transaction request orders for each target supply vehicle, determine the expected supply price for each target supply vehicle. Based on the target vehicle information and transaction request orders for each target vehicle, determine the expected demand quote for each target vehicle.

[0008] Furthermore, in some embodiments, based on the target vehicle information of each supply vehicle and the target vehicle information of each demand vehicle, multiple target supply vehicles are selected from multiple supply vehicles, and multiple target demand vehicles are selected from multiple demand vehicles, including: Based on the target vehicle information of each supply vehicle, select multiple candidate supply vehicles that are effective in representing the state of charge from multiple supply vehicles. Based on the target vehicle information of each demand vehicle, multiple candidate demand vehicles that represent the effective state of charge are selected from multiple demand vehicles. Based on the target vehicle information of each candidate supply vehicle and each candidate demand vehicle, a feasibility assessment is conducted on the supply and demand capacity between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple target supply vehicles from multiple candidate demand vehicles and multiple target demand vehicles from multiple candidate demand vehicles.

[0009] Furthermore, in some embodiments, the target vehicle information includes the predicted state of charge, battery capacity, and available time window from entry to exit of the trading parking area; Based on the target vehicle information of each candidate supply vehicle and each candidate demand vehicle, a feasibility assessment is conducted on the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle to determine multiple target supply vehicles from multiple candidate demand vehicles and multiple target demand vehicles from multiple candidate demand vehicles, including: Based on the available time windows of each candidate supply vehicle and each candidate demand vehicle, a time feasibility condition assessment is conducted on the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle to determine multiple time-feasible supply vehicles from multiple candidate demand vehicles and multiple time-feasible demand vehicles from multiple candidate demand vehicles. Based on the predicted state of charge and battery capacity of each candidate supply vehicle and the predicted state of charge and battery capacity of each candidate demand vehicle, an energy feasibility assessment is conducted on the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple target supply vehicles and multiple target demand vehicles from multiple time-feasible supply vehicles.

[0010] Furthermore, in some embodiments, the expected supply price for each target supply vehicle is determined based on the target vehicle information and transaction request order, including: Based on the multi-attribute utility function, the supply distance cost, battery degradation cost, and grid purchase opportunity cost of each target supply vehicle are calculated according to the target vehicle information and transaction request orders of each target supply vehicle. The expected supply price for each target vehicle is determined by summing the supply distance cost, battery degradation cost, and grid purchase opportunity cost.

[0011] Furthermore, in some embodiments, the expected demand price for each target vehicle is determined based on the target vehicle information and transaction request order for each target demand vehicle, including: Based on the multi-attribute utility function, the charging station charging cost and charging time cost of each target supply vehicle are calculated according to the target vehicle information and transaction request order of each target supply vehicle. The difference between the charging cost and charging time cost of each target vehicle is calculated to determine the expected demand price for each target vehicle.

[0012] Furthermore, in some embodiments, a two-way auction is conducted on each expected supply bid and each expected demand bid to determine multiple winning supply vehicles and multiple winning demand vehicles, including: Construct a demand curve vector by arranging multiple expected demand quotes in descending order, and construct a supply curve vector by arranging multiple expected supply quotes in ascending order. Simultaneously traverse and compare the demand curve vector and the supply curve vector to find the largest integer index value that satisfies the requirement that the buyer's bid covers the seller's asking price; Based on the largest integer index value, the multiple target demand vehicles corresponding to the expected demand quotes ranked in the corresponding positions before the demand curve vector are identified as winning demand vehicles, and the multiple target supply vehicles corresponding to the expected supply quotes ranked in the corresponding positions before the supply curve vector are identified as winning supply vehicles.

[0013] To achieve the above objectives, a second aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the end-to-end electric vehicle energy trading matching method of the second aspect embodiment described above.

[0014] To achieve the above objectives, a third aspect of the present application provides a storage medium, which is a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the end-to-end electric vehicle energy trading matching method of the second aspect embodiment described above.

[0015] According to an embodiment of this application, an end-to-end electric vehicle energy trading matching method, device, and medium have at least the following beneficial effects: By reasoning about vehicle information through a spatiotemporal state prediction model, the generation of bids from both supply and demand sides is based on accurate state predictions for future trading periods, improving the reliability of the trading basis. Replacing traditional centralized forced matching with a two-way auction bidding mechanism, each vehicle autonomously submits its desired bid, and the winner set is determined through market game theory between supply and demand, fully respecting the self-interested decision-making rights of vehicle owners and achieving decentralized, efficient resource matching. Further shortest-distance matching is performed within the winner pool, effectively reducing the energy consumption and time loss incurred by vehicles in completing the transaction. Independent game theory calculations are performed for each final trading pair to generate a settlement price, ensuring that the transaction price is both above the seller's floor price and below the buyer's psychological price, allowing both parties to obtain economic surplus from the transaction. Finally, a transaction execution contract is automatically generated, providing a traceable contractual basis for energy delivery and fund settlement.

[0016] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description and the accompanying drawings. Attached Figure Description

[0017] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0018] The present application will be further described below with reference to the accompanying drawings and embodiments; Figure 1 This is an optional flowchart of the end-to-end electric vehicle energy trading and matching method provided in the embodiments of this application; Figure 2 This is provided by the embodiments of this application. Figure 1 An optional flowchart for step S103; Figure 3 This is provided by the embodiments of this application. Figure 2 An optional flowchart in step S202; Figure 4 This is provided by the embodiments of this application. Figure 3 An optional flowchart for step S303; Figure 5 This is an optional flowchart provided in an embodiment of this application for determining the expected supply price for each target supply vehicle; Figure 6 This is an optional flowchart provided in an embodiment of this application for determining the expected demand quote for each target vehicle; Figure 7 This is an optional flowchart provided in the embodiments of this application for conducting a two-way auction bidding for each expected supply price and each expected demand price; Figure 8 This is a schematic diagram of an optional hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0019] This section will describe in detail the specific embodiments of this application. Preferred embodiments of this application are shown in the accompanying drawings. The purpose of the drawings is to supplement the textual description with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of this application, but they should not be construed as limiting the scope of protection of this application.

[0020] In the description of this application, the use of "first" and "second" is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of indicated technical features or the order of the indicated technical features. It should be understood that such use of data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0022] With increasing global emphasis on environmental protection and sustainable development, electric vehicles (EVs) are rapidly becoming the mainstream alternative to traditional internal combustion engine vehicles. EVs do not produce greenhouse gases during operation, thus being considered a key technology for addressing climate change. However, the surge in the number of EVs has also brought significant challenges to existing power infrastructure. These challenges manifest in three core dimensions. First, there is the dramatic increase in grid load. Large numbers of EVs simultaneously charging on the grid in an uncoordinated manner, especially during peak evening electricity consumption periods, generate extremely high instantaneous power demand, exacerbating the duck curve effect of the power grid and easily leading to transformer overload, local voltage instability, and even regional power outages. Second, there is range anxiety among users. Despite continuous improvements in battery energy density and driving range, users' concerns about whether their vehicles can safely reach their destination or whether they can find available charging resources in emergencies remain a core pain point hindering the widespread adoption of EVs. Finally, there is the structural shortage and uneven distribution of charging infrastructure. In rural areas, remote regions, and older residential areas, the coverage density and actual availability of charging stations are severely insufficient, making it difficult to meet the ever-growing and highly spatially and dynamically distributed charging demand.

[0023] To address these issues, vehicle-to-vehicle (V2V) energy trading has emerged, allowing direct energy transfer and sharing between electric vehicles. However, existing V2V energy trading methods generally employ centralized solution architectures, such as matching models based on mixed-integer linear programming. These models face the curse of dimensionality when handling large-scale concurrent vehicle requests, resulting in exponentially increasing solution time and failing to meet the millisecond-level real-time requirements of dynamic trading. Furthermore, centralized processing forces the collection of all vehicle information, posing a serious risk of privacy breaches. Some studies have introduced heuristic algorithms such as particle swarm optimization and ant colony optimization to reduce computational complexity. However, these mechanisms rely on a central server to calculate a global comprehensive score and perform greedy forced pairing, treating electric vehicle owners as completely passive, unconscious nodes. This completely ignores the self-interested nature of owners as independent, rational individuals seeking to maximize their own benefits, leading to low user participation. In addition, existing solutions generally focus on improving hardware transmission rates, with overly general scenario settings, lacking deep integration of owner-driven pricing, diverse cost considerations, and real-time incentives. Although some studies have attempted to introduce penalty factors to constrain vehicles that do not actively match, from a practical experience perspective, positive incentives are far more effective than mandatory penalties, and existing technologies lack effective market-based incentive mechanisms.

[0024] Based on this, embodiments of this application provide an end-to-end electric vehicle energy trading pairing method, device and medium, which realizes autonomous, efficient and win-win energy trading through spatiotemporal prediction, two-way auction and shortest distance pairing.

[0025] Therefore, the embodiments of this application will be further described below with reference to the accompanying drawings.

[0026] Reference Figure 1 As shown, Figure 1 This is an optional flowchart of an end-to-end electric vehicle energy trading pairing method provided in the embodiments of this application. The method may include, but is not limited to, steps S101 to S105.

[0027] Step S101: Obtain real-time vehicle information and transaction request orders for multiple electric vehicles.

[0028] It should be noted that the system collects real-time vehicle status data through onboard terminals or mobile applications deployed on each electric vehicle. This real-time vehicle information includes at least the vehicle identifier, current location coordinates, real-time state of charge (SBC), and parking duration. Parking duration includes the time taken to enter the parking area and the estimated time to leave. The system utilizes an automated data stream loading script to continuously access the collected data and performs rigorous multi-level verification and cleaning: converting all discrete time strings into high-precision time objects; removing invalid data due to sensor malfunctions; filtering logically paradoxical data with negative or zero parking durations; and cleaning abnormal records where the initial SBC value is outside a preset physical reasonable range. Simultaneously, the system receives transaction request orders submitted autonomously by vehicle owners through a smart agent. These transaction request orders include at least a transaction role identifier, the desired transaction amount, and a desired price generated based on a multi-attribute utility function. For vehicle owners on the supply side, the transaction request order includes the floor price for electricity sales determined by them based on the grid purchase opportunity cost, battery depreciation cost, and estimated distance wear cost, as well as the expected electricity sales price generated based on the floor price. For vehicle owners on the demand side, the transaction request order includes the highest psychological price for electricity purchase determined by them based on the commercial fast charging cost and estimated waiting time cost, as well as the expected electricity purchase price generated based on the highest psychological price.

[0029] Step S102: Input the real-time vehicle information into the spatiotemporal state prediction model for state reasoning to obtain the target vehicle information of each electric vehicle during the transaction period.

[0030] The spatiotemporal state prediction model includes a Long Short-Term Memory (LSTM) network layer. This LSTM layer comprises an input layer, a core LSTM hidden layer, and a fully connected regression output layer. At each time step, it precisely regulates the state through forget gates, input gates, and output gates to capture the long-range dependencies of vehicle spatiotemporal behavior. For a given input feature sequence... and the hidden state of the previous moment The mathematical derivation of the model is as follows.

[0031]

[0032] in, For the state representation of the forget gate, This represents the state of the input gate. This represents the state of candidate network nodes. This represents the current state of the network node. This represents the state of the output gate. The state representation that hides the current state. Here is the weight matrix for the forget gate. The weight matrix of the input gate, The weight matrix of candidate network nodes, Here is the weight matrix of the output gate. , , , For the corresponding bias term, It is represented as the hyperbolic tangent activation function.

[0033] It should be noted that during the training phase, the verified and cleaned real-time vehicle information undergoes feature engineering to extract high-precision timestamps for arrival and departure times. Periodic decomposition is then performed on the time features, transforming discrete hour and minute components into continuous sinusoidal and cosine periodic components to eliminate discontinuities in time features across midnight and reduce the model's learning burden. The processed feature vectors, along with the current location coordinates and initial state of charge, are input into a spatiotemporal state prediction model based on a single long short-term memory network layer for training. During the inference phase, the trained spatiotemporal state prediction model predicts the arrival and departure times of each electric vehicle during the target trading period, as well as the estimated state of charge at that time, based on the input feature sequence. The predicted arrival and departure times and estimated state of charge are output as target vehicle information for subsequent supply and demand classification, feasibility assessment, and multi-attribute quotation generation.

[0034] In one possible implementation, the system is trained using the Adam optimizer and a mean squared error loss function. First, the model replaces the traditional stochastic gradient descent optimization method with the Adam optimizer. The Adam optimizer combines momentum and adaptive learning rate mechanisms, enabling it to dynamically adjust the learning rate adaptively for different parameters, exhibiting extremely high convergence efficiency when processing sparse gradients in rural travel data. Since predicting arrival and departure times is essentially a continuous-value regression task, the system uses a standard mean squared error loss function to accurately measure the error between the network's predicted output and the actual data.

[0035] Furthermore, during the training phase, when backpropagating to calculate the gradients of the spatiotemporal state prediction model's parameters, if the L2 norm of the gradient vector exceeds a preset safety threshold, the system will scale it proportionally. This mechanism fundamentally eliminates the gradient explosion problem that might occur under extremely rare abnormal data interference, ensuring the absolute stability of the training process. After each round of model parameter updates, the system comprehensively evaluates the current model's MSE metric on an independent validation set. A tolerance window is set (e.g., 15 rounds). If the loss on the validation set does not show a substantial decrease or improvement for 15 consecutive rounds, the system will forcibly terminate the training process early and automatically roll back, saving the historically best-performing model weights. This mechanism effectively prevents model overfitting, obtaining the most generalizing model with minimal time cost.

[0036] Step S103: Based on the target vehicle information and transaction request orders of each electric vehicle, determine multiple target supply vehicles, the expected supply price of each target supply vehicle, multiple target demand vehicles, and the expected demand price of each target demand vehicle among the multiple electric vehicles.

[0037] It should be noted that the supply and demand roles of vehicles are classified based on the estimated state of charge (SOC) in the target vehicle information. Vehicles with an SOC higher than the preset minimum supply threshold are identified as target supply vehicles, while vehicles with an SOC lower than the preset demand threshold are identified as target demand vehicles. The expected supply price for target supply vehicles is extracted from the transaction request orders submitted by those vehicles, and the expected demand price for target demand vehicles is extracted from the transaction request orders submitted by those vehicles.

[0038] Step S104: Conduct a two-way auction to bid on each expected supply price and each expected demand price to determine multiple winning supply vehicles and multiple winning demand vehicles, and match each winning supply vehicle and each winning demand vehicle with the shortest distance in spatial location to obtain multiple final trading pairs.

[0039] Specifically, the system's two-way auction bidding involves market-clearing by comparing supply-side and demand-side bids. The collected expected supply bids are ranked from low to high, and expected demand bids from high to low. The system compares each price in the ranked queue to find the maximum transaction quantity where the demand-side bid still covers the supply-side asking price. The top-ranked suppliers and demanders are then selected as the winners. After determining the winners, a bipartite graph mapping is constructed based on the location coordinates of the winning supply and demand vehicles. The goal is to minimize the total physical distance traveled between the paired vehicles, achieving spatial matching and forming one-to-one final transaction pairs, thus minimizing the additional driving costs incurred by the transactions.

[0040] Step S105: For each final trading pair, perform game theory calculations on the corresponding expected supply price and expected demand price to generate a settlement price, and generate a transaction execution contract based on the settlement price.

[0041] Specifically, for any successfully matched final transaction pair, the system obtains the supplier's expected electricity sales price and the demander's expected electricity purchase price, and retrieves the system's preset profit distribution tendency weight. The profit distribution tendency weight is a configurable parameter between 0 and 1, used to control the distribution ratio of the economic surplus generated by the transaction between the supply and demand sides. The settlement price is calculated according to the following rules: if the profit distribution tendency weight is the median, the settlement price equals the arithmetic mean of the expected electricity sales price and the expected electricity purchase price, with both supply and demand sides sharing the economic surplus equally; if the profit distribution tendency weight shifts towards 1, the settlement price aligns with the demander's expected electricity purchase price, allocating more economic surplus to the supplier; if the profit distribution tendency weight shifts towards 0, the settlement price aligns with the supplier's expected electricity sales price, allocating more economic surplus to the demander. The settlement price must be higher than the supplier's floor price and lower than the demander's maximum psychological purchase price, ensuring the supplier obtains a positive net profit while the demander achieves charging cost savings. After generating the settlement price, the system determines the final settlement amount, the agreed energy delivery quantity and delivery time window based on the settlement price, and generates a binding transaction execution contract, providing a traceable contractual basis for subsequent energy transmission and fund transfer.

[0042] In this process, the expected supply price and expected demand price for each final transaction pair are used for game theory calculations to generate the settlement price. The formula is expressed as follows:

[0043] in, Weighting of profit allocation To provide the expected price for the car supplied to the winner, Winner's demand for vehicles, expected demand, and pricing. .

[0044] Taking the late-night off-peak hours in remote rural areas of China as an example, the willingness to supply vehicles is often extremely low, creating an absolute seller's market. At this time, the system can dynamically adjust the profit distribution bias weight to 0.8. The final settlement price will then significantly converge towards the extremely high bids from buyers. This means that the vast majority of transaction profits are forcibly allocated to the suppliers. This market-generated expectation of exorbitant profits creates a very strong positive incentive for vehicle suppliers, stimulating them to actively join the seller-to-seller network to provide electricity.

[0045] Furthermore, after the transaction execution contract is completed, the system calculates the actual benefits for each party. Because... The offer price must be higher than the expected price of the car owner who supplied the winning car. The seller's profit is always positive. Similarly, because the price paid by the owner of the winning demand car is always lower than the price they expected to receive. Buyer costs decreased significantly, and customer satisfaction was substantially preserved. The net profit obtained by the car owners supplied by the winner is as follows: The following are the costs that owners of winning demand vehicles can save: .

[0046] In a specific embodiment of steps S101 to S105, vehicle information is inferred through a spatiotemporal state prediction model, enabling the generation of bids from both supply and demand sides to be based on accurate state predictions for future transaction periods, thus improving the reliability of the transaction basis. A two-way auction bidding mechanism replaces the traditional centralized forced matching, with each vehicle independently submitting its desired bid and determining the winner set through market game theory between supply and demand. This fully respects the self-interested decision-making rights of vehicle owners and achieves decentralized, efficient resource matching. Within the winner range, the shortest distance matching in spatial location is further executed, effectively reducing the energy consumption and time loss incurred by vehicles in completing the transaction. An independent game calculation is performed for each final transaction pair to generate a settlement price, ensuring that the transaction price is both above the seller's floor price and below the buyer's psychological price, allowing both parties to obtain economic surplus from the transaction. Finally, a transaction execution contract is automatically generated, providing a traceable contractual basis for energy delivery and fund settlement.

[0047] The process involves inputting real-time vehicle information into a spatiotemporal state prediction model for state reasoning to obtain target vehicle information for each electric vehicle during the trading period. Specifically, this includes the following steps: inputting the vehicle identifier, current location coordinates, real-time charge state, and parking duration of each electric vehicle into the spatiotemporal state prediction model so that the long short-term memory network layer can perform spatiotemporal feature reasoning on the activity trajectory of each electric vehicle to obtain target vehicle information for each electric vehicle during the trading period.

[0048] Specifically, for each electric vehicle, the system extracts the vehicle identifier, current location coordinates, real-time state of charge, and parking duration determined by the entry time and estimated departure time from real-time vehicle information. These fields are integrated into an input feature vector and fed into the spatiotemporal state prediction model. The long short-term memory network layer within the spatiotemporal state prediction model jointly infers the temporal dependencies and spatial location changes in the input feature vector, capturing the vehicle's behavioral patterns during the target trading period. The model outputs the target vehicle information for that electric vehicle during the trading period, including estimated entry time, estimated departure time, and estimated state of charge, providing a basis for subsequent supply and demand classification and price generation.

[0049] Among them, reference Figure 2 As shown, Figure 2 This is provided by the embodiments of this application. Figure 1An optional flowchart for step S103, the method may include, but is not limited to, steps S201 to S202.

[0050] Step S201: Based on the transaction request orders of each electric vehicle, divide each electric vehicle into multiple supply vehicles and multiple demand vehicles.

[0051] Specifically, the system iterates through all electric vehicles that have submitted transaction request orders in the current bidding round, and reads the transaction role identifier field carried in each transaction request order. If the transaction role identifier field indicates that the electric vehicle intends to transfer electricity, it is included in the candidate supply vehicle set L. i =(x i y i If the transaction role identifier field indicates that the electric vehicle's transaction intention is to obtain electricity, then it is included in the candidate demand vehicle set L. j =(x j y j For each vehicle in the candidate supply vehicle set, the expected electricity sales price, available time window, and estimated tradable electricity volume are extracted from its transaction request order; for each vehicle in the candidate demand vehicle set, the expected electricity purchase price, available time window, and demanded electricity volume are extracted from its transaction request order. After the division, the candidate supply vehicle set and the candidate demand vehicle set will serve as the basic participant sets for subsequent feasibility assessment and two-way auction bidding.

[0052] Step S202: Based on the target vehicle information of each supply vehicle and the target vehicle information of each demand vehicle, select multiple target supply vehicles from multiple supply vehicles and select multiple target demand vehicles from multiple demand vehicles.

[0053] Specifically, for each supply vehicle in the candidate supply vehicle set, the estimated state of charge is extracted from its target vehicle information. Supply vehicles with an estimated state of charge higher than the preset minimum supply charge threshold are retained, while supply vehicles with an estimated state of charge lower than the threshold are eliminated. The retained supply vehicles are determined as target supply vehicles. For each demand vehicle in the candidate demand vehicle set, the estimated state of charge is extracted from its target vehicle information. Demand vehicles with an estimated state of charge lower than the preset demand target charge threshold are retained, while demand vehicles with an estimated state of charge higher than the threshold are eliminated. The retained demand vehicles are determined as target demand vehicles.

[0054] Step S203: Determine the expected supply price for each target supply vehicle based on the target vehicle information and transaction request order for each target supply vehicle.

[0055] Specifically, for each target supply vehicle, the estimated state of charge and current location coordinates are extracted from its target vehicle information, and a pre-determined floor price for electricity sales is extracted from its transaction request order. The target supply vehicle generates a desired supply bid based on the floor price and the upper limit of the transferable electricity volume determined by the estimated state of charge. The desired supply bid is not lower than the floor price to ensure that the target supply vehicle can still obtain positive economic benefits after completing the energy transfer.

[0056] Step S204: Determine the expected demand price for each target vehicle based on the target vehicle information and transaction request order for each target demand vehicle.

[0057] Specifically, for each vehicle with target demand, the estimated state of charge (SOC) and current location coordinates are extracted from its target vehicle information, and a pre-determined upper limit for electricity purchase price is extracted from its transaction request order. Based on this upper limit, and considering the electricity demand gap determined by the difference between the estimated SOC and the preset target SOC threshold, the vehicle generates a desired demand quote. This desired quote does not exceed the upper limit to ensure that the cost paid by the vehicle after charging is lower than its maximum acceptable price, thus achieving charging cost savings.

[0058] Reference Figure 3 As shown, Figure 3 This is provided by the embodiments of this application. Figure 2 An optional flowchart in step S202, the method may include, but is not limited to, steps S301 to S303.

[0059] Step S301: Based on the target vehicle information of each supply vehicle, select multiple candidate supply vehicles from the multiple supply vehicles to identify those with valid state of charge.

[0060] Specifically, for each supply vehicle in the supply vehicle set, the estimated state of charge (SOC) is extracted from its target vehicle information and compared with a preset minimum SOC threshold. If the estimated SOC is higher than the minimum SOC threshold, it indicates that the supply vehicle has surplus electricity available for transfer during the target trading period, and it is identified as a candidate supply vehicle and retained. If the estimated SOC is lower than or equal to the minimum SOC threshold, it indicates that the supply vehicle's own electricity is close to the safety limit and it does not meet the objective conditions for participating in energy transfer, so it is removed from the candidate supply vehicle set. This results in multiple candidate supply vehicles, denoted as... ,in, It is the minimum charge threshold supplied to the vehicle.

[0061] Step S302: Based on the target vehicle information of each demand vehicle, select multiple candidate demand vehicles that represent valid state of charge from multiple demand vehicles.

[0062] Specifically, for each vehicle in the demand vehicle set, its estimated state of charge (SBC) is extracted from its target vehicle information and compared with a preset target SBC threshold. If the estimated SBC is lower than the target SBC threshold, it indicates that the vehicle has a charging gap during the target transaction period, and it is identified as a candidate vehicle and retained. If the estimated SBC is higher than or equal to the target SBC threshold, it indicates that the vehicle's battery capacity already meets the target requirements and it does not meet the objective conditions for participating in the charging demand; therefore, it is removed from the candidate vehicle set, resulting in multiple candidate supply vehicles, denoted as... ,in, It is the minimum charge threshold required for the vehicle.

[0063] Step S303: Based on the target vehicle information of each candidate supply vehicle and the target vehicle information of each candidate demand vehicle, conduct a feasibility assessment of the supply and demand capacity between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple target supply vehicles from multiple candidate demand vehicles and multiple target demand vehicles from multiple candidate demand vehicles.

[0064] Specifically, for any pair of candidate supply vehicles and candidate demand vehicles, the system extracts the available time window and estimated state of charge from the target vehicle information of the candidate supply vehicles, and extracts the available time window and estimated state of charge from the target vehicle information of the candidate demand vehicles. It then comprehensively judges the degree of matching between the two in the time dimension and the energy dimension. Candidate supply vehicles that meet the supply-demand matching conditions are designated as official target supply vehicles, and candidate demand vehicles that meet the supply-demand matching conditions are designated as official target demand vehicles. Combinations where supply and demand capabilities do not meet the matching conditions are eliminated.

[0065] In one possible embodiment, refer to Figure 4 As shown, Figure 4 This is provided by the embodiments of this application. Figure 3 An optional flowchart for step S303, the method may include, but is not limited to, steps S401 to S406.

[0066] Step S401: Based on the available time windows of each candidate supply vehicle and each candidate demand vehicle, conduct a time feasibility assessment of the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple time-feasible supply vehicles from multiple candidate demand vehicles and multiple time-feasible demand vehicles from multiple candidate demand vehicles.

[0067] Available time windows include: [ , ], [ , ],in, Representative candidate supply vehicle The time of entering the parking area, Representative candidate supply vehicle The time spent leaving the parking area Representative candidate demand vehicle The time of entering the parking area Representative candidate demand vehicle Time spent leaving the parking area.

[0068] Specifically, for any pair of candidate supply vehicles and candidate demand vehicles, the system extracts the available time windows for both the candidate supply vehicles and the candidate demand vehicles, and determines whether there is an overlap in the time dimension between the two available time windows. Candidate supply vehicles with overlapping time windows are identified as time-feasible supply vehicles, and candidate demand vehicles with overlapping time windows are identified as time-feasible demand vehicles. For pairs where the available time windows do not overlap, it is determined that the energy transaction cannot be completed between the two parties in the time dimension, and these pairs are eliminated.

[0069] It should be noted that the relevant formula for determining whether two available time windows have overlapping intervals in the time dimension is as follows:

[0070]

[0071] in, This represents the shortest service response time. This refers to the duration of overlap between candidate supply vehicles and candidate demand vehicles. The formula above indicates that by calculating the duration of overlap between candidate supply vehicles and candidate demand vehicles, and comparing this duration with the preset shortest service response time, it is determined that the overlap duration must be strictly greater than the shortest service response time. Otherwise, it would be a waste of resources.

[0072] Step S402: Based on the predicted charge status and battery capacity of each candidate supply vehicle and the predicted charge status and battery capacity of each candidate demand vehicle, conduct an energy feasibility assessment of the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple target supply vehicles and multiple target demand vehicles from multiple time-feasible supply vehicles.

[0073] Specifically, for each pair of candidate supply vehicles and candidate demand vehicles that passes the time feasibility assessment, the system determines the available energy transfer capacity of the candidate supply vehicle based on its predicted state of charge and battery capacity, and determines the required supplementary energy transfer capacity of the candidate demand vehicle based on its predicted state of charge and battery capacity. The available energy transfer capacity and the required supplementary energy transfer capacity are compared, and candidate supply vehicles that can form an effective energy transfer are identified as target supply vehicles, and corresponding candidate demand vehicles are identified as target demand vehicles. Combinations that do not meet the matching conditions in the energy dimension are eliminated.

[0074] It should be noted that the relevant formula for comparing the available electricity supply with the required replenishment electricity is as follows:

[0075]

[0076]

[0077] in, This refers to the battery capacity of an electric vehicle. This is expressed as the amount of electricity available for sale. This indicates the amount of electricity required to replenish. The minimum tradable electricity volume is set at this minimum. The system will only process transactions if the available electricity volume for sale or the required supplementary electricity volume exceeds this minimum tradable electricity volume, thus avoiding data waste.

[0078] Reference Figure 5 As shown, Figure 5 This is an optional flowchart provided in the embodiments of this application for determining the expected supply price of each target supply vehicle. The method may include, but is not limited to, steps S501 to S502.

[0079] Step S501: Based on the multi-attribute utility function, calculate the supply distance cost, battery degradation cost, and grid purchase opportunity cost of each target supply vehicle according to the target vehicle information and transaction request order of each target supply vehicle.

[0080] Specifically, for each target supply vehicle, the current location coordinates are extracted from its target vehicle information, and the expected transaction area information is extracted from its transaction request order. Based on the spatial distance between the current location coordinates and the expected transaction area, the supply distance cost is calculated using a multi-attribute utility function. The estimated state of charge and historical charge-discharge cycle count are extracted from its target vehicle information, and the impact of this transaction on the degradation of battery chemical life is evaluated using a multi-attribute utility function and converted into battery degradation cost. The time-of-use electricity price information of the power grid where the target supply vehicle is located is extracted from the transaction request order. Combined with the transaction period, the grid purchase opportunity cost of forgoing electricity sales to or purchases from the grid due to the transfer of electricity is calculated using a multi-attribute utility function.

[0081] It should be noted that the process of calculating the supply distance cost using the multi-attribute utility function is as follows: First, the physical Euclidean distance between the target supply vehicle and the target demand vehicle is calculated. A longer physical Euclidean distance means that the vehicle needs to consume more battery power, experience more mechanical wear, and bear a higher risk of traffic accidents in order to complete the transaction. The system sets an absolute unreachable distance threshold. If this distance threshold is exceeded The transaction cost is then determined to be infinite. Within the threshold, the physical Euclidean distance is linearly normalized and assigned an additional cost of one currency unit. That is, the supply distance cost is .

[0082] Step S502: Sum the supply distance cost, battery degradation cost, and grid purchase opportunity cost for each target supply vehicle to determine the expected supply price for each target supply vehicle.

[0083] Specifically, for each target supply vehicle, the supply distance cost, battery degradation cost, and grid power purchase opportunity cost are added together to obtain the total comprehensive cost of the target supply vehicle. The target supply vehicle uses this total comprehensive cost as a pricing benchmark, with an upward adjustment to generate the expected supply price. The expected supply price is not lower than the total comprehensive cost to ensure that the revenue obtained by the target supply vehicle after completing energy trading can fully cover all cost expenditures and generate a positive profit. The formula for ensuring that the expected supply price is not lower than the total comprehensive cost is expressed as follows:

[0084] in, In order to provide a quote, To cover the opportunity cost of purchasing electricity for the power grid, This is for the cost of battery degradation.

[0085] Reference Figure 6 As shown, Figure 6 This is an optional flowchart provided in the embodiments of this application for determining the expected demand price of each target demand vehicle. The method may include, but is not limited to, steps S601 to S602.

[0086] Step S601: Based on the multi-attribute utility function, calculate the charging station charging cost and charging time cost of each target supply vehicle according to the target vehicle information and transaction request order of each target supply vehicle.

[0087] Specifically, for each target vehicle, the unit charging price of commercial public fast charging stations in its local area is extracted from its transaction request order. This, combined with the energy demand gap determined by the difference between the estimated state of charge in the target vehicle information and the preset target state of charge threshold, is used to calculate the charging station cost if the vehicle were to charge at a commercial public fast charging station using a multi-attribute utility function. Simultaneously, the time span between the current moment and the expected start time of charging is extracted from the target vehicle information and used as the expected waiting time. This waiting time is then converted into a charging time cost using a multi-attribute utility function. The charging station cost and the charging time cost together constitute the cost benchmark for the target vehicle participating in this transaction.

[0088] It should be noted that the process of converting the expected waiting time into charging time cost through the multi-attribute utility function is as follows: the system sets a maximum tolerance time. If this maximum tolerance time is exceeded The transaction is then deemed invalid. Within the acceptable tolerance range, the additional cost factor for the waiting time is converted into currency units. That is, the supply distance cost is .

[0089] Step S602: Calculate the difference between the charging station charging cost and charging time cost for each target supply vehicle to determine the expected demand price for each target supply vehicle.

[0090] Specifically, for each vehicle with target demand, the charging cost at charging stations is used as a price benchmark. The monetary equivalent of the charging time cost is deducted from this benchmark to obtain the expected demand price for that vehicle. The expected demand price should not exceed the charging cost at charging stations, in order to reflect the economic savings that the target vehicle gains by participating in end-to-end energy trading compared to charging at commercial public fast charging stations.

[0091] The formula for ensuring that the expected demand price does not exceed the charging cost of the charging station is expressed as follows:

[0092] in, Quote based on expected demand. The cost of charging at charging stations.

[0093] Reference Figure 7 As shown, Figure 7 This is an optional flowchart provided in the embodiments of this application for conducting a two-way auction bidding for each expected supply price and each expected demand price. The method may include, but is not limited to, steps S701 to S703.

[0094] Step S701: Arrange multiple expected demand quotes in descending order to construct a demand curve vector, and arrange multiple expected supply quotes in ascending order to construct a supply curve vector.

[0095] Specifically, the system collects all expected demand bids submitted by all target vehicles in the current bidding round, sorts them in descending order of price from highest to lowest, and forms a demand curve vector. The bids at the beginning of the vector correspond to the demand side with the strongest willingness to pay. Simultaneously, the system collects all expected supply bids submitted by all target supply vehicles in the current bidding round, sorting them in ascending order of price from lowest to highest, forming a supply curve vector. The bids placed at the beginning of the vector after arrangement correspond to the suppliers with the most significant cost advantage. The demand curve vector and the supply curve vector together constitute the market supply and demand benchmark for two-way auction bidding.

[0096] Step S702: Synchronously traverse and compare the demand curve vector and the supply curve vector to find the largest integer index value that satisfies the buyer's bid covering the seller's asking price.

[0097] Specifically, starting from the initial index position, the process iterates synchronously in ascending order along the sorting direction of the demand curve vector and the supply curve vector, comparing the expected demand price in the demand curve vector with the expected supply price in the supply curve vector at the same index position. If the expected demand price at the current index position is greater than or equal to the expected supply price, the process continues. The process stops when the expected demand price is less than the expected supply price, and the previous index value is determined as the largest integer index value K that satisfies the condition that the buyer's bid covers the seller's asking price. This largest integer index value K defines the maximum number of valid transactions that can be achieved in this round of two-way auction bidding. The criteria for determining the largest integer index value K satisfy the following: .

[0098] Step S703: Based on the largest integer index value, identify the multiple target demand vehicles corresponding to the expected demand quotes ranked in the corresponding positions before the demand curve vector as the winning demand vehicles, and identify the multiple target supply vehicles corresponding to the expected supply quotes ranked in the corresponding positions before the supply curve vector as the winning supply vehicles.

[0099] Specifically, the system uses the largest integer index value K to extract the first K expected demand quotes from the beginning of the demand curve vector, identifying the target demand vehicles corresponding to these K quotes as the winning demand vehicles. Simultaneously, it extracts the first K expected supply quotes from the beginning of the supply curve vector, identifying the target supply vehicles corresponding to these K quotes as the winning supply vehicles. These K pairs of winning supply vehicles and winning demand vehicles constitute the legitimate trading entities in this round of two-way auction. The trading requests of unsuccessful supply vehicles and unsuccessful demand vehicles that do not make it into the first K positions will be carried over to the next bidding round for re-participation in the auction.

[0100] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned end-to-end electric vehicle energy trading and matching method. This electronic device can be any smart terminal, including mobile phones, tablets, and in-vehicle computers.

[0101] Please see Figure 8 , Figure 8 This is a schematic diagram of an optional hardware structure of an electronic device provided in an embodiment of this application. The electronic device includes: The processor 801 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the end-to-end electric vehicle energy trading pairing method provided in the embodiments of this application. The memory 802 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 802 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 802 and is called and executed by the processor 801 to execute the end-to-end electric vehicle energy trading and matching method provided in the embodiments of this application. The 803 input / output interface is used to implement information input and output. The communication interface 804 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 805 transmits information between various components of the device (e.g., processor 801, memory 802, input / output interface 803, and communication interface 804); The processor 801, memory 802, input / output interface 803, and communication interface 804 are connected to each other within the device via bus 805.

[0102] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, provides an end-to-end electric vehicle energy trading matching method.

[0103] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0104] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0105] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0106] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; 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.

[0107] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0108] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0109] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0110] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0111] The units described above 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0112] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0113] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-accessible storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0114] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for end-to-end electric vehicle energy trading and matching, characterized in that, include: Obtain real-time vehicle information and transaction request orders for multiple electric vehicles; The real-time vehicle information is input into the spatiotemporal state prediction model for state reasoning to obtain the target vehicle information of each electric vehicle during the transaction period. Based on the target vehicle information and transaction request orders of each of the electric vehicles, a number of target supply vehicles, the expected supply price of each of the target supply vehicles, a number of target demand vehicles, and the expected demand price of each of the target demand vehicles are determined among the multiple electric vehicles. A two-way auction is conducted on each of the expected supply quotes and each of the expected demand quotes to determine multiple winning supply vehicles and multiple winning demand vehicles. The winning supply vehicles and the winning demand vehicles are then matched in space with the shortest distance to obtain multiple final trading pairs. For each final trading pair, game theory calculations are performed on the corresponding expected supply price and expected demand price to generate a settlement price, and a trading execution contract is generated based on the settlement price.

2. The electric vehicle energy trading matching method according to claim 1, characterized in that, The real-time vehicle information includes vehicle identification, vehicle current location coordinates, real-time state of charge, and vehicle parking duration; the spatiotemporal state prediction model includes a long short-term memory network layer. The step of inputting the real-time vehicle information into a spatiotemporal state prediction model for state reasoning to obtain the target vehicle information of each electric vehicle during the transaction period includes: The vehicle identifier, current location coordinates, real-time state of charge, and parking duration of each electric vehicle are input into the spatiotemporal state prediction model, so that the long short-term memory network layer can perform spatiotemporal feature inference on the activity trajectory of each electric vehicle to obtain the target vehicle information of each electric vehicle during the transaction period.

3. The electric vehicle energy trading matching method according to claim 2, characterized in that, The step of determining multiple target supply vehicles, expected supply prices for each target supply vehicle, multiple target demand vehicles, and expected demand prices for each target demand vehicle from among the multiple electric vehicles based on the target vehicle information and transaction request orders for each of the electric vehicles includes: Based on the transaction request orders for each of the electric vehicles, each of the electric vehicles is divided into multiple supply vehicles and multiple demand vehicles; Based on the target vehicle information of each supply vehicle and the target vehicle information of each demand vehicle, multiple target supply vehicles are selected from the multiple supply vehicles, and multiple target demand vehicles are selected from the multiple demand vehicles. Based on the target vehicle information and transaction request orders for each of the target supply vehicles, determine the expected supply price for each of the target supply vehicles; Based on the target vehicle information and transaction request orders for each of the target demand vehicles, the expected demand price for each of the target demand vehicles is determined.

4. The electric vehicle energy trading matching method according to claim 3, characterized in that, The step of selecting multiple target supply vehicles from the multiple supply vehicles and selecting multiple target demand vehicles from the multiple demand vehicles based on the target vehicle information of each supply vehicle and the target vehicle information of each demand vehicle includes: Based on the target vehicle information of each of the supply vehicles, select multiple candidate supply vehicles that represent a valid state of charge from the multiple supply vehicles; Based on the target vehicle information of each of the demand vehicles, multiple candidate demand vehicles that represent valid state of charge are selected from the multiple demand vehicles. Based on the target vehicle information of each candidate supply vehicle and the target vehicle information of each candidate demand vehicle, a feasibility assessment is performed on the supply and demand capacity between each candidate supply vehicle and each candidate demand vehicle to determine multiple target supply vehicles from the multiple candidate demand vehicles and multiple target demand vehicles from the multiple candidate demand vehicles.

5. The electric vehicle energy trading matching method according to claim 4, characterized in that, The target vehicle information includes the predicted charge status, battery capacity, and available time window from entry to exit of the trading parking area; The step of conducting a feasibility assessment of the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle based on the target vehicle information of each candidate supply vehicle and the target vehicle information of each candidate demand vehicle, in order to determine multiple target supply vehicles from the multiple candidate demand vehicles and multiple target demand vehicles from the multiple candidate demand vehicles, includes: Based on the available time windows of each candidate supply vehicle and each candidate demand vehicle, a time feasibility condition assessment is performed on the supply and demand capacity between each candidate supply vehicle and each candidate demand vehicle to determine multiple time-feasible supply vehicles from the multiple candidate demand vehicles and multiple time-feasible demand vehicles from the multiple candidate demand vehicles. Based on the predicted charge status and battery capacity of each candidate supply vehicle and the predicted charge status and battery capacity of each candidate demand vehicle, an energy feasibility condition assessment is performed on the supply and demand capabilities between each candidate supply vehicle and each candidate demand vehicle, so as to determine multiple target supply vehicles and multiple target demand vehicles from the multiple time-feasible supply vehicles.

6. The electric vehicle energy trading matching method according to claim 3, characterized in that, The step of determining the expected supply price for each of the target supply vehicles based on the target vehicle information and transaction request orders includes: Based on the multi-attribute utility function, the supply distance cost, battery degradation cost, and grid purchase opportunity cost of each target supply vehicle are calculated according to the target vehicle information and transaction request order of each target supply vehicle. The expected supply price for each target supply vehicle is determined by summing the supply distance cost, battery degradation cost, and grid purchase opportunity cost for each target supply vehicle.

7. The electric vehicle energy trading matching method according to claim 3, characterized in that, The step of determining the expected demand price for each of the target demand vehicles based on the target vehicle information and transaction request orders includes: Based on the multi-attribute utility function, the charging station charging cost and charging time cost of each target supply vehicle are calculated according to the target vehicle information and transaction request order of each target supply vehicle. The difference between the charging cost and charging time cost of each target supply vehicle at the charging station is calculated to determine the expected demand price for each target supply vehicle.

8. The electric vehicle energy trading matching method according to claim 1, characterized in that, The step of conducting a two-way auction to determine multiple winning supply vehicles and multiple winning demand vehicles for each of the expected supply and demand quotations includes: A demand curve vector is constructed by arranging the multiple expected demand quotes in descending order, and a supply curve vector is constructed by arranging the multiple expected supply quotes in ascending order. Simultaneously traverse and compare the demand curve vector and the supply curve vector to find the largest integer index value that satisfies the buyer's bid covering the seller's asking price; Based on the maximum integer index value, the multiple target demand vehicles corresponding to the expected demand quotes ranked in the corresponding positions before the demand curve vector are identified as winning demand vehicles, and the multiple target supply vehicles corresponding to the expected supply quotes ranked in the corresponding positions before the supply curve vector are identified as winning supply vehicles.

9. An electronic device, characterized in that, The electronic device is provided with a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the end-to-end electric vehicle energy trading matching method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a processor-executable program, which, when executed by a processor, is the end-to-end electric vehicle energy trading pairing method as claimed in any one of claims 1 to 8.

Images