New energy automobile charging scheme recommendation method and system based on dynamic carbon integral excitation, terminal and medium

Abstract

The invention relates to the field of new energy vehicle charging, and particularly discloses a new energy vehicle charging scheme recommendation method and system based on dynamic carbon integral excitation, a terminal and a medium, and the method comprises the steps: obtaining input data, including user behavior data, vehicle state data, power grid environment data and candidate charging pile data; according to a pre-trained machine learning model, performing feature extraction on time series data and spatial data in the input data to generate a space-time fusion feature vector; and constructing constraint conditions, and generating a recommended charging scheme through a Pareto optimization algorithm on the basis of the space-time fusion feature vector by taking carbon integral maximization, charging cost minimization and charging pile load balancing as targets. According to the method, multiple factors are considered, the accuracy of the recommendation scheme is improved, the utilization rate of renewable energy sources is improved based on dynamic carbon excitation, and more comprehensive optimal configuration of charging resources is realized by taking multiple factors as optimization targets.

Description

Technical Field

[0001] The present application relates to the field of new energy vehicle charging, and specifically to a method, system, terminal and medium for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives. Background Art

[0002] At present, the popularity of new energy vehicles has also brought about a substantial increase in charging demand. How to efficiently and reasonably allocate charging resources, balance the load on the power grid, and promote the use of renewable energy has become an urgent problem to be solved.

[0003] Existing methods for recommending charging plans for new energy vehicles often focus solely on single factors, such as charging cost or charging time, lacking comprehensive consideration of multiple factors, including user needs, vehicle status, and the grid environment. For one thing, these methods may only consider a few factors, such as user location and distance to charging stations, while failing to fully consider multiple factors, including user behavior, vehicle status, grid environment, and charging station load. This can result in suboptimal recommended charging plans that fail to meet user needs in different scenarios. Furthermore, carbon credits, a key means of incentivizing users to participate in environmentally friendly charging, often employ a static carbon credit mechanism in related technologies, failing to dynamically adjust based on real-time grid conditions, such as the proportion of renewable energy and grid load. Consequently, these methods fail to effectively incentivize users to charge during periods with high renewable energy proportions and low grid load, thereby limiting the utilization of renewable energy and the green development of the grid. Furthermore, these related technologies focus solely on minimizing charging cost or time, failing to simultaneously consider multiple optimization objectives, such as maximizing carbon credits, minimizing charging costs, and balancing charging station loads. This results in recommended charging plans that focus on only one aspect while neglecting other important aspects, failing to achieve a comprehensive and optimal allocation of charging resources. Summary of the Invention

[0004] To solve the above problems, the present invention provides a method, system, terminal and medium for recommending new energy vehicle charging plans based on dynamic carbon credit incentives. It considers multiple factors to improve the accuracy of the recommended plans, improves the utilization rate of renewable energy based on dynamic carbon incentives, and uses multiple factors as optimization targets to achieve a more comprehensive optimization configuration of charging resources.

[0005] In a first aspect, the technical solution of the present invention provides a method for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives, comprising the following steps: Obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; Based on the pre-trained machine learning model, feature extraction is performed on the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector; Construct constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, and generate recommended charging plans using the Pareto optimization algorithm based on the spatiotemporal fusion feature vectors; Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

[0006] In an optional embodiment, the user behavior data includes the destination travel distance and historical charging records; the historical charging records include historical charging time, charging pile identification, frequency, and actual carbon credits; Vehicle status data includes vehicle remaining power SoC and battery capacity; Grid environment data includes real-time renewable energy proportion, grid load rate, time period type, and real-time electricity price; The candidate charging pile data includes latitude and longitude location, load rate, new and old pile identification, and available time period.

[0007] In an optional embodiment, feature extraction is performed on the time series data and spatial data in the input data according to a pre-trained machine learning model to generate a spatiotemporal fusion feature vector, specifically including: Use the pre-trained LSTM model to extract time series features from the time series data in the input data and generate a time series feature vector; Use the pre-trained GCN model to extract spatial features from the spatial data in the input data and generate a spatial feature vector; A multi-head attention mechanism is used to fuse the temporal feature vector and the spatial feature vector to generate a spatiotemporal fusion feature vector.

[0008] In an optional embodiment, the constraints include: The estimated charging capacity is between the minimum and maximum charging capacities, the load rate of the charging pile is less than the threshold, and the charging pile is within the available time of the charging period; Among them, the minimum charge capacity = (destination travel distance / vehicle energy consumption) - remaining power; Maximum charge capacity = battery capacity * (1-SoC / 100).

[0009] In an optional embodiment, the objective function constructed with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads is expressed as:

[0010] Where, is the weight coefficient, For the Candidate charging stations.

[0011] In an optional embodiment, the method further includes optimizing the weight coefficients by a reinforcement learning method , specifically including: The state space of reinforcement learning is defined to include the spatiotemporal fusion feature vector, the historical average of actual carbon credits, and the historical average of actual electricity costs; Define the action of reinforcement learning as adjusting the weight coefficient ; Defining the Reward Function for Reinforcement Learning ;in, To predict carbon credits, is the actual carbon credit, To predict charging costs, is the actual charging cost, To predict the charging pile load rate, is the actual charging pile load rate; Update weight coefficients using PPO strategy .

[0012] In an optional embodiment, the method further includes adjusting the time weight coefficient according to the real-time grid load rate, which is expressed as:

[0013] in, is the adjustment coefficient, is the adjusted time weight coefficient, is the real-time grid load rate, is the average load rate of the power grid.

[0014] In a second aspect, the technical solution of the present invention also includes a new energy vehicle charging plan recommendation system based on dynamic carbon credit incentives, comprising: Input data acquisition module, used to obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; The fusion feature extraction module is used to extract features from the time series data and spatial data in the input data based on the pre-trained machine learning model to generate a spatiotemporal fusion feature vector; The recommended charging plan generation module is used to build constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads. It generates recommended charging plans based on the spatiotemporal fusion feature vectors using the Pareto optimization algorithm. Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

[0015] In a third aspect, the technical solution of the present invention provides a terminal, including: A memory, used for storing a new energy vehicle charging solution recommendation program based on dynamic carbon credit incentives; A processor is configured to implement the steps of any of the above-mentioned methods for recommending a new energy vehicle charging solution based on dynamic carbon credit incentives when executing the program for recommending a new energy vehicle charging solution based on dynamic carbon credit incentives.

[0016] In a fourth aspect, the technical solution of the present invention provides a computer-readable storage medium, on which is stored a new energy vehicle charging plan recommendation program based on dynamic carbon credit incentives. When the new energy vehicle charging plan recommendation program based on dynamic carbon credit incentives is executed by a processor, the steps of the new energy vehicle charging plan recommendation method based on dynamic carbon credit incentives as described in any one of the above items are implemented.

[0017] It can be seen from the above technical solutions that this application has the following advantages: (1) Achieve multi-objective optimization, including maximizing carbon credits, minimizing charging costs, and balancing charging pile loads. By encouraging users to charge during periods of sufficient renewable energy supply, this will promote environmental protection, save users money, and improve the overall utilization rate of charging piles. (2) Based on temporal feature extraction, spatial feature extraction and spatiotemporal fusion, it can capture the temporal dependency in the data, consider the geographical location and load conditions of the charging piles, and thus comprehensively consider temporal and spatial factors to improve the accuracy of recommendations; (3) Dynamic time weight coefficient adjustment, based on the real-time grid load rate, enables the recommended plan to adapt more flexibly to the real-time status of the grid, further optimizes charging costs and carbon credits, and provides users with more environmentally friendly, economical and efficient charging plan recommendation services. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for the description. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1A schematic flow chart of a method for recommending a new energy vehicle charging solution based on dynamic carbon credit incentives provided in an embodiment of the present invention.

[0020] Figure 2 This is a schematic block diagram of the structure of a new energy vehicle charging solution recommendation system based on dynamic carbon credit incentives provided by an embodiment of the present invention.

[0021] Figure 3 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the application objectives, features, and advantages of this application more obvious and easy to understand, the technical solutions protected by this application will be clearly and completely described below using specific embodiments and drawings. Obviously, the embodiments described below are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0023] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as those commonly understood by those skilled in the art to which the present invention pertains. The terms used in this application and in the specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention.

[0024] Figure 1 This is a flow chart of a method for recommending a new energy vehicle charging solution based on dynamic carbon credit incentives provided by an embodiment of the present invention. Figure 1 The execution entity may be a system for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives. The method for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives provided in embodiments of the present invention is executed by a computer device. Accordingly, the system for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives runs on the computer device. The order of the steps in this flowchart may be changed, and some steps may be omitted, depending on different needs.

[0025] This method comprehensively considers multiple factors such as user behavior data, vehicle status data, power grid environment data, and candidate charging pile data to achieve comprehensive optimization of the charging plan. At the same time, a dynamic carbon credit mechanism is introduced to dynamically adjust the carbon credit coefficient and time weight coefficient according to the real-time power grid environment to encourage users to charge during periods when the proportion of renewable energy is high and the power grid load rate is low. In addition, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, a recommended charging plan is generated through the Pareto optimization algorithm, and the weight coefficient is further continuously optimized through reinforcement learning methods to adapt to the optimization needs in different scenarios. Figure 1 As shown, the method includes the following steps.

[0026] S1, obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data.

[0027] This step first obtains input data, then performs feature extraction and other processing on these input data, and then generates charging recommendation solutions. The data in this embodiment includes user behavior data, vehicle status data, power grid environment data, candidate charging pile data User behavior data: including the user's destination travel distance and historical charging records; including historical charging time, charging station identification, charging frequency, and actual carbon credits. These data help understand the user's charging habits and needs.

[0028] Vehicle status data: including the vehicle's remaining charge (SoC) and battery capacity. These data are used to assess the vehicle's charging needs and charging capacity.

[0029] Grid environment data: including real-time renewable energy proportion, grid load rate, time period type, real-time electricity price, etc. These data reflect the real-time status of the grid and are used to optimize charging plans, balance grid load, and promote the use of renewable energy.

[0030] Candidate charging pile data: including latitude and longitude location, load rate, identification of new and old piles, and available time periods. This data is the actual charging facility information that needs to be considered when recommending a charging plan.

[0031] S2, based on the pre-trained machine learning model, extracts features from the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector.

[0032] After acquiring the input data, a pre-trained machine learning model is used to extract features from the temporal and spatial data. These features are then fused into a spatiotemporal fusion feature vector using a multi-head attention mechanism. In some optional implementations, a spatiotemporal graph neural network (ST-GNN) is employed, specifically including the following steps.

[0033] S21, use the pre-trained LSTM model to extract time series features from the time series data in the input data and generate a time series feature vector.

[0034] Time series data includes data from the user side, the grid side, and the charging pile side. The user side includes historical charging data and actual carbon credits. The grid side includes real-time renewable energy share, time period type, and real-time electricity price. Real-time renewable energy share and real-time electricity price can be collected every 15 minutes, and time period types include peak, off-peak, and regular. The charging pile side includes load rate and available time period. Load rate can be collected every 15 minutes.

[0035] When extracting time series features, we first construct a time window. For example, we divide a day into 96 15-minute periods. This generates a time series matrix, handles missing values, and then performs one-hot encoding on the period types and normalizes the numerical features. The time series matrix is ​​then fed into the LSTM layer to learn the temporal pattern. The attention mechanism then calculates attention weights for the 96 periods input to the LSTM layer, highlighting the user's most frequent charging periods. This results in the output of a time series feature vector.

[0036] S22, use the pre-trained GCN model to extract spatial features from the spatial data in the input data and generate a spatial feature vector.

[0037] Spatial data includes spatial data on the user side and the charging pile side. The user side includes the destination travel distance and the longitude and latitude of the frequently visited charging piles. The charging pile side includes the longitude and latitude.

[0038] When extracting spatial features, we first construct a spatial graph. For example, candidate charging stations are used as nodes, with features including latitude and longitude location and load rate. The edges between nodes represent the distance weights between charging stations. The spatial graph node features are then fed into the GCN layer to aggregate neighborhood information. The GraphAttention layer then calculates attention weights for adjacent nodes, prioritizing low-load charging stations. This results in the output of a spatial feature vector.

[0039] S23 uses a multi-head attention mechanism to fuse the temporal feature vector and the spatial feature vector to generate a spatiotemporal fusion feature vector.

[0040] The temporal feature vector and the spatial feature vector are concatenated to generate a concatenated vector. The Multi-HeadAttention layer is then used to calculate multi-head attention on the concatenated vector, dynamically adjust the spatiotemporal weights, and output the spatiotemporal fusion feature vector.

[0041] This embodiment combines input data and extracts precise spatiotemporal features through a three-layer architecture of time series, space, and fusion, and then generates a spatiotemporal vector that includes user preferences, low-carbon potential, and load balancing to support charging plan recommendations.

[0042] S3, constructs constraint conditions, takes maximizing carbon credits, minimizing charging costs and balancing charging pile loads as the goals, and generates recommended charging plans through the Pareto optimization algorithm based on the spatiotemporal fusion feature vector.

[0043] It should be noted that the recommended charging plan includes recommended charging time, recommended charging pile, recommended charging amount, expected carbon credits, and expected charging cost.

[0044] Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period, is the amount of renewable energy generated by the grid during the charging period, is the total power generation of the grid during the charging period.

[0045] In some optional implementations, the time weight coefficient Can be pre-configured, Table 1 shows the time weight coefficient Configuration table.

[0046] Table 1: Time weight coefficients Configuration Table

[0047] For example, if a user charges at a charging station in a certain area between 8:00 PM and 10:00 PM, and wind power accounts for 30% of the electricity generated during that period, the carbon credit coefficient for that charging behavior can be calculated as 1.3, based on a preset algorithm. If the user charges 10 kWh during that period, they will receive carbon credits corresponding to 13 kWh. In some optional implementations, the time weight coefficient is calculated based on the real-time grid load rate. Perform dynamic adjustment, expressed as:

[0048] in, is the adjustment coefficient, is the adjusted time weight coefficient, is the real-time grid load rate, is the average load rate of the power grid.

[0049] In some optional implementations, carbon credit weighting is adjusted based on the degree of completion of regional carbon emission targets published by environmental protection authorities. For example, if a region's carbon emission target completion rate for the first quarter of this year is 75%, indicating a low completion rate, the system will dynamically increase the credit weighting for charging during renewable energy periods, such as increasing the carbon credit weighting for peak wind and photovoltaic power generation periods by 0.2 percentage points. This adjusted credit weighting is fed back and applied to the carbon credit generation process.

[0050] In step S3, the system constructs constraints based on the spatiotemporal fusion feature vectors and generates a recommended charging plan using a Pareto optimization algorithm, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads. Under the premise of meeting all constraints, the optimal charging plan recommendation is generated to achieve optimal allocation and efficient utilization of charging resources.

[0051] First, establish constraints. These constraints include the estimated charge capacity being between the minimum and maximum charge capacities, the charging pile load being less than a threshold, and the available charging pile time being within the charging period. The minimum charge capacity is calculated as (destination distance / vehicle energy consumption) - remaining charge; the maximum charge capacity is calculated as battery capacity * (1 - SoC / 100).

[0052] Secondly, in the process of generating the recommended charging plan through the Pareto optimization algorithm, the goals are to maximize carbon credits, minimize charging costs, and balance the load of charging piles. Accordingly, the constructed objective function is expressed as:

[0053] Where, is the weight coefficient, For the Candidate charging stations.

[0054] Under the above constraints and objective function, a recommended charging plan is generated through the Pareto optimization algorithm, which specifically includes the following steps.

[0055] S31, generating an initial population based on the spatiotemporal fusion feature vector, that is, randomly generating several candidate solutions that meet the constraints.

[0056] S32, calculate the objective function value for each candidate solution and perform non-dominated sorting.

[0057] S33, for the solutions in the same non-dominated layer, calculate the congestion degree and select the solution with the highest congestion degree.

[0058] Crowding refers to the distance between adjacent solutions in the target space. The solution with the highest crowding is selected to ensure a uniform distribution of solutions and cover multi-objective trade-off scenarios, such as different combinations of high carbon credits-high cost and low carbon credits-low cost.

[0059] S34, generate the next generation population through crossover and mutation operations, and repeat steps S32 to S34 until convergence.

[0060] The crossover can include your time period + charging station combination crossover, and the variation includes fine-tuning of the charging amount.

[0061] S35, after convergence, the Pareto front contains all non-dominated solutions, according to the current weight coefficient Calculate the objective function value of each frontier solution, and select the solution with the largest objective function value as the recommended charging solution.

[0062] In some optional embodiments, the weight coefficient of the objective function is optimized by reinforcement learning method , specifically including the following steps.

[0063] Step 1: Define the state space of reinforcement learning, including the spatiotemporal fusion feature vector, the historical average value of actual carbon credits, and the historical average value of actual electricity costs.

[0064] Step 2: Define the action of reinforcement learning as adjusting the weight coefficient .

[0065] Step 3: Define the reward function for reinforcement learning ;in, To predict carbon credits, is the actual carbon credit, To predict charging costs, is the actual charging cost, To predict the charging pile load rate, is the actual charging pile load rate.

[0066] It should be noted that the actual value can be obtained through each actual charging process, and the average of multiple actual charging data is taken as the actual value.

[0067] Step 4: Update the weight coefficient using the PPO strategy .

[0068] For example, analyzing a user's charging history over the past year revealed that they typically charged at Charging Station A near their workplace between 7:00 PM and 9:00 PM on weekdays. If the user enters a destination 50 kilometers away the next day and their vehicle's remaining battery is 60%, the charging plan recommendation module will consider the aforementioned factors and perform a multi-objective optimization calculation using a machine learning model. In this example, the user will be recommended to charge at Charging Station A near their workplace between 6:00 PM and 8:00 PM the next day. This ensures that the user earns the highest carbon credits while maintaining reasonable charging costs while ensuring the required travel time.

[0069] In some alternative implementations, blockchain technology can be used to automatically execute carbon credit issuance rules through smart contracts, storing carbon credits on-chain and ensuring data immutability. This can establish a carbon credit trading platform that supports point transactions between users and the exchange of points for the rights of partner merchants. This ensures the transparency and immutability of the point transaction process, with all transaction records permanently stored on the blockchain and accessible to users at any time.

[0070] The above describes in detail an embodiment of a method for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives. Based on the method for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives described in the above embodiment, an embodiment of the present invention also provides a new energy vehicle charging plan recommendation system based on dynamic carbon credit incentives corresponding to this method.

[0071] Figure 2 This is a schematic block diagram of the structure of a new energy vehicle charging plan recommendation system based on dynamic carbon credit incentives provided by an embodiment of the present invention. In this embodiment, the new energy vehicle charging plan recommendation system 200 based on dynamic carbon credit incentives can be divided into multiple functional modules according to the functions it performs, such as Figure 2 The functional modules may include: an input data acquisition module 210, a fusion feature extraction module 220, and a recommended charging plan generation module 230. A module as referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can perform fixed functions, and is stored in a memory.

[0072] The input data acquisition module 210 is used to acquire input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data.

[0073] The fusion feature extraction module 220 is used to extract features from the time series data and spatial data in the input data according to a pre-trained machine learning model to generate a spatiotemporal fusion feature vector.

[0074] The recommended charging plan generation module 230 is used to construct constraint conditions, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, and to generate a recommended charging plan based on the spatiotemporal fusion feature vector using a Pareto optimization algorithm; Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

[0075] The new energy vehicle charging plan recommendation system based on dynamic carbon credit incentives of this embodiment is used to implement the aforementioned new energy vehicle charging plan recommendation method based on dynamic carbon credit incentives. Therefore, the specific implementation methods of this system can be seen in the embodiment part of the new energy vehicle charging plan recommendation method based on dynamic carbon credit incentives in the previous text. Therefore, its specific implementation methods can refer to the descriptions of the corresponding embodiments of each part and will not be elaborated here.

[0076] In addition, since the new energy vehicle charging plan recommendation system based on dynamic carbon credit incentives in this embodiment is used to implement the aforementioned new energy vehicle charging plan recommendation method based on dynamic carbon credit incentives, its function corresponds to that of the aforementioned method and will not be repeated here.

[0077] Figure 3This is a schematic diagram of the structure of a terminal 300 provided in an embodiment of the present invention, comprising: a processor 310, a memory 320, and a communication unit 330. The processor 310 is configured to implement the following steps when implementing a new energy vehicle charging plan recommendation program based on dynamic carbon credit incentives stored in the memory 320: Obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; Based on the pre-trained machine learning model, feature extraction is performed on the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector; Construct constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, and generate recommended charging plans using the Pareto optimization algorithm based on spatiotemporal fusion feature vectors; Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

[0078] The terminal 300 includes a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention; it may be a bus structure or a star structure, and may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0079] Memory 320 can be used to store execution instructions of processor 310. Memory 320 can be implemented by 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 memory, flash memory, magnetic disk, or optical disk. When the execution instructions in memory 320 are executed by processor 310, terminal 300 can perform some or all of the steps in the above-described method embodiments.

[0080] The processor 310 is the control center of the storage terminal. It uses various interfaces and lines to connect various parts of the entire electronic terminal. It runs or executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic terminal and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 310 can only include a central processing unit (CPU). In an embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.

[0081] The communication unit 330 is configured to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals, or send user data to other terminals.

[0082] The present invention also provides a computer storage medium, wherein the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0083] The present invention also provides a computer storage medium, wherein the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0084] The computer storage medium stores a new energy vehicle charging solution recommendation program based on dynamic carbon credit incentives. When the new energy vehicle charging solution recommendation program based on dynamic carbon credit incentives is executed by a processor, the following steps are implemented: Obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; Based on the pre-trained machine learning model, feature extraction is performed on the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector; Construct constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, and generate recommended charging plans using the Pareto optimization algorithm based on spatiotemporal fusion feature vectors; Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

[0085] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software and a necessary general-purpose hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, among other media capable of storing program code, and includes instructions for causing a computer terminal (which can be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention.

[0086] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0087] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0088] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0089] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein, but is intended to be construed in the widest manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for recommending charging plans for new energy vehicles based on dynamic carbon credit incentives, characterized in that: The following steps are involved: Obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; Based on the pre-trained machine learning model, feature extraction is performed on the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector; Construct constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads, and generate recommended charging plans using the Pareto optimization algorithm based on spatiotemporal fusion feature vectors; Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

2. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 1 is characterized in that: User behavior data includes destination travel distance and historical charging records; the historical charging records include historical charging time, charging station identification, frequency, and actual carbon credits; Vehicle status data includes vehicle remaining power SoC and battery capacity; Grid environment data includes real-time renewable energy proportion, grid load rate, time period type, and real-time electricity price; The candidate charging pile data includes latitude and longitude location, load rate, new and old pile identification, and available time period.

3. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 2 is characterized in that: Based on the pre-trained machine learning model, feature extraction is performed on the time series data and spatial data in the input data to generate a spatiotemporal fusion feature vector, specifically including: Use the pre-trained LSTM model to extract time series features from the time series data in the input data and generate a time series feature vector; Use the pre-trained GCN model to extract spatial features from the spatial data in the input data and generate a spatial feature vector; A multi-head attention mechanism is used to fuse the temporal feature vector and the spatial feature vector to generate a spatiotemporal fusion feature vector.

4. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 1 is characterized in that: Constraints include: The estimated charging capacity is between the minimum and maximum charging capacities, the load rate of the charging pile is less than the threshold, and the charging pile is within the available time of the charging period; Among them, the minimum charge capacity = (destination travel distance / vehicle energy consumption) - remaining power; Maximum charge capacity = battery capacity * (1-SoC / 100).

5. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 1 is characterized in that: The objective function constructed with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads is expressed as: Where, is the weight coefficient, For the Candidate charging stations.

6. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 5 is characterized in that: The method also includes optimizing the weight coefficients through reinforcement learning methods , specifically including: The state space of reinforcement learning is defined to include the spatiotemporal fusion feature vector, the historical average of actual carbon credits, and the historical average of actual electricity costs; Define the action of reinforcement learning as adjusting the weight coefficient ; Defining the Reward Function for Reinforcement Learning ;in, To predict carbon credits, is the actual carbon credit, To predict charging costs, is the actual charging cost, To predict the charging pile load rate, is the actual charging pile load rate; Update weight coefficients using PPO strategy .

7. The method for recommending new energy vehicle charging solutions based on dynamic carbon credit incentives according to claim 1 is characterized in that: The method also includes adjusting the time weight coefficient according to the real-time grid load rate, which is expressed as: in, is the adjustment coefficient, is the adjusted time weight coefficient, is the real-time grid load rate, is the average load rate of the power grid.

8. A new energy vehicle charging solution recommendation system based on dynamic carbon credit incentives, characterized in that: include: Input data acquisition module, used to obtain input data, including user behavior data, vehicle status data, power grid environment data, and candidate charging pile data; The fusion feature extraction module is used to extract features from the time series data and spatial data in the input data based on the pre-trained machine learning model to generate a spatiotemporal fusion feature vector; The recommended charging plan generation module is used to build constraints, with the goals of maximizing carbon credits, minimizing charging costs, and balancing charging pile loads. It generates recommended charging plans based on the spatiotemporal fusion feature vectors using the Pareto optimization algorithm. Carbon credits ; Where, is the charge capacity, is the carbon integral coefficient, is the weight coefficient of renewable energy proportion, is the time weight coefficient, is the proportion of renewable energy during the charging period.

9. A terminal, characterized in that: include: A memory, used for storing a new energy vehicle charging solution recommendation program based on dynamic carbon credit incentives; A processor, configured to implement the steps of the method for recommending a new energy vehicle charging solution based on dynamic carbon credit incentives as described in any one of claims 1 to 7 when executing the new energy vehicle charging solution recommendation program based on dynamic carbon credit incentives.

10. A computer-readable storage medium, characterized in that The readable storage medium stores a new energy vehicle charging plan recommendation program based on dynamic carbon credit incentives. When the new energy vehicle charging plan recommendation program based on dynamic carbon credit incentives is executed by the processor, the steps of the new energy vehicle charging plan recommendation method based on dynamic carbon credit incentives as described in any one of claims 1 to 7 are implemented.

Images