Ordered charging and discharging distributed cooperative control method, system and device for electric vehicle group and storage medium
By establishing an electric vehicle user database and a game theory model, formulating a dynamic electricity pricing strategy, and adopting a distributed control algorithm and a feedback learning mechanism, the problems of insufficient communication burden and privacy protection in the charging and discharging control of electric vehicle groups were solved, and the synergistic optimization of system benefits and user satisfaction was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-11
- Publication Date
- 2026-03-13
AI Technical Summary
Existing electric vehicle charging control methods suffer from heavy communication burden, high computational complexity, insufficient privacy protection, difficulty in balancing system benefits and user satisfaction, and lack of effective coordination mechanisms, resulting in low computational efficiency and difficulty in meeting the real-time control requirements of large-scale electric vehicle fleets.
By establishing an electric vehicle user database, designing an incentive registration mechanism and a credit evaluation system, combining machine learning algorithms to predict charging demand, constructing a game theory model and formulating a dynamic electricity pricing strategy, and adopting a distributed control algorithm and a feedback learning mechanism, collaborative optimization between users and the power grid can be achieved.
It improves the system's flexibility and adaptability, meets users' personalized needs, reduces communication burden and privacy leakage risks, enhances system efficiency and user satisfaction, and achieves supply and demand balance and load shaving and valley filling.
Smart Images
Figure CN121663585A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of smart grid and electric vehicle charging control technology, and in particular to an orderly charging and discharging distributed collaborative control method, system, device and storage medium for electric vehicle groups. Background Technology
[0002] In recent years, the number of electric vehicles has grown rapidly. However, the large-scale connection of electric vehicles to the power grid has brought many challenges to the operation of the power system. How to achieve orderly charging and discharging control of electric vehicle groups, while ensuring the stable operation of the system and meeting the charging needs of users, has become an important technical problem.
[0003] In existing technologies, most traditional electric vehicle charging control methods employ centralized control strategies. These strategies require extensive real-time communication and global information exchange, leading to heavy communication burdens, high computational complexity, and insufficient privacy protection. Furthermore, existing charging control methods often neglect personalized user needs and behavioral characteristics, making it difficult to balance system efficiency with user satisfaction. Some decentralized control methods lack effective coordination mechanisms, easily leading to conflicts between local optima and control objectives. They also suffer from low computational efficiency when coordinating the charging and discharging of large-scale electric vehicle fleets, failing to meet real-time control requirements. Existing technologies also suffer from inaccurate user behavior modeling, simplistic incentive mechanisms, and a lack of effective handling of uncertainties. Summary of the Invention
[0004] In view of the above-mentioned problems, the present invention provides an ordered charging and discharging distributed cooperative control method, system, device and storage medium for electric vehicle fleets.
[0005] Therefore, the technical problem solved by this invention is: how to establish a charging and discharging decision model that takes into account user behavior characteristics, design an incentive-compatible decentralized collaborative control mechanism, achieve synergistic optimization of system benefits and user satisfaction, and ensure the convergence and real-time performance of the algorithm.
[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, comprising: Collect basic user information, vehicle parameters, and charging preferences to establish an electric vehicle user database. Use behavioral feature recognition algorithms and classification models to segment users, and design incentive registration mechanisms, credit evaluation systems, and privacy protection frameworks. Based on the electric vehicle user database and combined with historical data, the system uses machine learning algorithms to predict charging demand, assess the available grid capacity and distributed power supply status, analyze demand response potential, and conduct risk assessment. Based on the prediction and evaluation results, a game model between electric vehicle users and grid operators is constructed, the strategy space and payoff function are defined, and the equilibrium solution is solved to obtain the game equilibrium result. Based on the game equilibrium results, a dynamic electricity pricing strategy and a differentiated pricing mechanism are formulated, and price signals are released through multiple channels. At the same time, a price response evaluation and adjustment mechanism is designed. Based on the released price signals and the user's own needs, a decentralized control algorithm is executed to enable the user to formulate and execute a charging plan, while monitoring and handling abnormal situations in real time. Based on system operation data and user feedback, the control effect is evaluated, and system parameters and strategies are optimized through a feedback learning mechanism to achieve continuous improvement.
[0007] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: The process of collecting basic user information, vehicle parameters, and charging preferences to establish an electric vehicle user database, and using behavioral feature recognition algorithms and classification models to segment users, while also designing an incentive registration mechanism, a credit rating system, and a privacy protection framework includes: User behavior is modeled, and the user utility function is defined as a comprehensive consideration of power satisfaction, charging cost, and time preference penalty; the power satisfaction function adopts a piecewise calculation logic based on whether the state of charge has reached the target value; the charging cost function includes electricity price cost and waiting cost; Design a user behavior feature recognition algorithm, analyze user travel patterns, charging habits and price sensitivity, establish a user classification model, and divide users into types based on behavioral features; A credit rating system is established to assess user credibility based on historical participation and performance records; the privacy protection framework adopts differential privacy protection to limit the differences in output results of adjacent datasets, and adds noise to the charging power through a noise addition mechanism.
[0008] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: The process, based on an electric vehicle user database and combined with historical data, uses machine learning algorithms to predict charging demand, assess grid capacity and distributed power generation status, analyze demand response potential, and conduct risk assessments, including: Grid operators consider operating costs with the objective function of minimizing generation costs, network loss costs, and peak costs, while adhering to load balance constraints, namely, the sum of generation power and electric vehicle charging and discharging power equals the sum of base load and network losses; Design a demand response potential assessment method to analyze users' willingness and ability to participate in orderly charging; establish a risk assessment mechanism to identify uncertainties affecting the execution of charging plans; and construct a resource optimization allocation strategy to rationally allocate charging time periods and power limits.
[0009] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: Based on the prediction and evaluation results, a game model is constructed between electric vehicle users and grid operators. The strategy space and payoff function are defined, and the equilibrium solution is solved to obtain the game equilibrium result, including: Define the strategy space and revenue function for grid operators and electric vehicle users, and construct a two-level optimization model. The grid operator is the leader with the goal of maximizing profits, and the electric vehicle user is the follower with the goal of maximizing its own utility. The evolution of the state of charge during the charging process is also considered.
[0010] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: The process of constructing a game model between electric vehicle users and grid operators based on prediction and evaluation results, defining the strategy space and payoff function, and solving for the equilibrium solution to obtain the game equilibrium result also includes: The existence and uniqueness of equilibrium solutions are analyzed using game theory methods, and a distributed optimization algorithm is used to solve the game equilibrium. Convergence criteria and termination conditions are established, including the convergence tolerance of the original residuals and dual residuals, as well as the requirement for the descent of the objective function.
[0011] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: The process of executing a distributed control algorithm based on published price signals and user needs, enabling users to create and execute charging plans, while simultaneously monitoring and handling anomalies in real time includes: Users formulate charging plans based on price signals and their own needs using optimization algorithms, and implement distributed control algorithms. Each electric vehicle executes the charging plan based on local information and coordination signals. An aggregator coordination mechanism is introduced. Aggregators aim to maximize their own profits. There is a profit-sharing mechanism between aggregators and users. Users receive a share of the aggregator's profits exceeding the benchmark profit according to a certain percentage.
[0012] The beneficial effects of this preferred technical solution are as follows: Users can formulate charging plans based on price signals and their own needs, better meeting their own requirements while also responding to grid regulation. The distributed control algorithm improves the system's flexibility and adaptability. The aggregator coordination mechanism integrates electric vehicle resources, enabling large-scale operation, and the profit-sharing mechanism incentivizes users and aggregators to actively participate, improving the overall efficiency of the system.
[0013] As a preferred scheme for an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, wherein: The process of executing a distributed control algorithm based on published price signals and user needs, enabling users to create and execute charging plans, and simultaneously monitoring and handling anomalies in real time, also includes: A real-time monitoring and feedback mechanism is established, and a model predictive control framework is used to track the execution status and deviation of the charging plan. The model predictive control framework achieves rolling optimization updates by minimizing the deviation between the predicted output and the reference trajectory and the cost of the control input, and determines the optimal charging power at the next moment.
[0014] The beneficial effects of this preferred technical solution are as follows: the real-time monitoring and feedback mechanism can promptly detect problems in the execution of the charging plan; the model predictive control framework can adjust the charging power in real time based on the prediction results, ensuring the accuracy and stability of the charging plan; and rolling optimization updates enable the system to continuously optimize the control strategy based on real-time conditions, improving the system's performance and adaptability.
[0015] Secondly, the present invention provides an ordered charging and discharging distributed cooperative control system for electric vehicle fleets, comprising: The user information management and classification module is used to collect basic user information, vehicle parameters and charging preferences, establish an electric vehicle user database, and use behavioral feature recognition algorithms and classification models to classify users. It also designs an incentive registration mechanism, a credit evaluation system and a privacy protection framework. The charging demand forecasting and risk assessment module is used to predict charging demand based on the electric vehicle user database and historical data, and to assess the available grid capacity and distributed power source status through machine learning algorithms. It also analyzes demand response potential and conducts risk assessment. The vehicle-grid game equilibrium solution module is used to construct a game model between electric vehicle users and grid operators based on prediction and evaluation results, define the strategy space and payoff function, and solve for the equilibrium solution to obtain the game equilibrium result. The dynamic electricity pricing and evaluation module is used to formulate dynamic electricity pricing strategies and differentiated pricing mechanisms based on game equilibrium results, and to release price signals through multiple channels, while designing price response evaluation and adjustment mechanisms. The charging plan execution and monitoring module is used to execute a distributed control algorithm based on the published price signals and the user's own needs, enabling the user to formulate and execute a charging plan, while monitoring and handling abnormal situations in real time. The system performance evaluation and optimization module is used to evaluate the control effect based on system operation data and user feedback, and optimize system parameters and strategies through a feedback learning mechanism to achieve continuous improvement.
[0016] Thirdly, the present invention provides a computer device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of an ordered charging and discharging distributed cooperative control method for electric vehicle groups.
[0017] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement steps of an ordered charging and discharging distributed cooperative control method for electric vehicle groups.
[0018] The beneficial effects of this invention are as follows: The decentralized collaborative control mechanism proposed in this invention fully considers the personalized needs and behavioral characteristics of users. Through game theory modeling, it achieves synergistic optimization of user satisfaction and system benefits, and avoids the communication burden and privacy leakage problems of centralized control, thereby improving the scalability and practicality of the system. The proposed incentive-compatible pricing mechanism design ensures that users are motivated to participate in orderly charging. By guiding user behavior through differentiated price signals, it achieves supply and demand balance and load peak shaving and valley filling, taking into account the heterogeneous characteristics of users and improving the accuracy and effectiveness of the control strategy. The proposed decentralized optimization algorithm based on ADMM (Alternating Direction Method of Multipliers) has good convergence and computational efficiency, and can handle the real-time control problem of large-scale electric vehicle groups. The distributed characteristics of the algorithm also reduce the risk of single-point failure and improve the robustness and reliability of the system. Attached Figure Description
[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0020] Figure 1This is an overall flowchart of an ordered charging and discharging distributed cooperative control method for electric vehicle fleets provided by the present invention. Detailed Implementation
[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.
[0022] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, including: S1: Collect basic user information, vehicle parameters and charging preferences, establish an electric vehicle user database, and use behavioral feature recognition algorithms and classification models to segment users. At the same time, design an incentive registration mechanism, a credit evaluation system and a privacy protection framework. S2: Based on the electric vehicle user database and combined with historical data, predict charging demand through machine learning algorithms, assess the available grid capacity and distributed power status, analyze demand response potential, and conduct risk assessment. S3: Based on the prediction and evaluation results, construct a game model between electric vehicle users and grid operators, define the strategy space and payoff function, and solve for the equilibrium solution to obtain the game equilibrium result. S4: Based on the game equilibrium results, formulate dynamic electricity pricing strategies and differentiated pricing mechanisms, release price signals through multiple channels, and design price response evaluation and adjustment mechanisms; S5: Based on the published price signals and the user's own needs, execute the distributed control algorithm to enable the user to formulate and execute the charging plan, while monitoring and handling abnormal situations in real time; S6: Based on system operation data and user feedback, evaluate the control effect and optimize system parameters and strategies through feedback learning mechanisms to achieve continuous improvement.
[0023] It should be noted that steps S1-S6 effectively facilitated interaction and collaboration between electric vehicle users and grid operators. From the collection and management of user information to the accurate prediction of charging demand, the construction and solution of game theory models, the formulation and implementation of dynamic electricity pricing strategies, the execution of charging plans, and the continuous optimization of the system, a complete and closed-loop control system was formed. This system not only fully considers users' personalized needs and charging preferences but also combines the actual operating conditions of the power grid to achieve rational allocation and efficient utilization of resources. This effectively improves the management level of orderly charging and discharging of electric vehicle fleets, reduces grid operating costs and risks, and simultaneously increases user satisfaction and participation, providing strong technical support and guarantees for the large-scale popularization of electric vehicles and the stable operation of the power grid.
[0024] Example 2, refer to Figure 1 As one embodiment of the present invention, based on the previous embodiment, a method for ordered charging and discharging distributed cooperative control of electric vehicle fleets is provided, comprising: In this embodiment, step S1 above involves collecting basic user information, vehicle parameters, and charging preferences to establish an electric vehicle user database. Users are then segmented using behavioral feature recognition algorithms and classification models. Furthermore, an incentive registration mechanism, a credit rating system, and a privacy protection framework are designed, including: Establish an electric vehicle user database to collect basic user information, vehicle parameters, and charging preferences. To more accurately analyze user behavior, model electric vehicle user behavior: The utility function for electric vehicle users is defined as: In the formula, Let i be the utility function for the i-th user. For charging power, For charging time, Let be the power satisfaction function. For the first State of charge at time t, For charging cost function, For time preference penalty function, Adjust the charging time according to the user's preference.
[0025] The user battery satisfaction function uses a piecewise linear function: In the formula, Based on the basic satisfaction coefficient, For the target state of charge, This represents the excess satisfaction coefficient.
[0026] The charging cost function includes electricity cost and waiting cost: In the formula, For the first Electricity price at any time For time intervals, For waiting cost coefficient, For waiting time.
[0027] In another possible implementation, when establishing an electric vehicle user database, one can cooperate with electric vehicle manufacturers to obtain data such as basic user information and vehicle parameters recorded when the vehicle leaves the factory. At the same time, information collection terminals can be set up at charging facilities to collect user charging preference data, such as the duration of each charge and the charging time. This data can then be integrated into the database.
[0028] In another possible implementation, social media platforms and mobile applications can be used to encourage users to proactively upload their basic information, vehicle parameters, and charging preferences when building an electric vehicle user database. Users who upload information can be rewarded, such as with charging coupons. The collected data is then cleaned, organized, and stored in the database.
[0029] In another possible implementation, when modeling electric vehicle user behavior, a questionnaire survey can be used to collect users' weights on battery satisfaction, charging cost, and time preference, and then a user utility function can be constructed based on these weights. For the battery satisfaction function, different segmented calculation logic can be set according to different vehicle types and usage scenarios.
[0030] In another possible implementation, when modeling electric vehicle user behavior, big data analysis of users' historical charging behavior can be used to uncover their potential needs and behavioral patterns, and machine learning algorithms can be used to build a user behavior model. During the construction process, the model parameters are continuously adjusted to more accurately reflect user behavior.
[0031] Design a user behavior feature recognition algorithm to analyze users' travel patterns, charging habits, and price sensitivity. Establish a user classification model to categorize users into different types based on their behavioral characteristics, providing a foundation for personalized services.
[0032] Design a user incentive registration mechanism to attract users to participate in the orderly charging project through preferential policies and service commitments. Establish a user credit evaluation system to assess user credibility based on historical participation and performance records. Construct a privacy protection framework to ensure the security and privacy of users' sensitive information.
[0033] Specifically, the privacy protection framework employs differential privacy protection: In the formula, For privacy protection mechanisms, , For adjacent datasets, For the output set, For privacy budget, This is due to a loss of privacy.
[0034] The privacy protection framework also includes a noise addition mechanism: In the formula, The charging power after adding noise. It is Gaussian noise.
[0035] In this embodiment, step S2 above, based on the electric vehicle user database and combined with historical data, uses machine learning algorithms to predict charging demand, assess the available grid capacity and distributed power supply status, and simultaneously analyze demand response potential and conduct risk assessment, including: Predicting electric vehicle charging demand based on historical data and real-time information, considering travel patterns, weather factors, and holiday impacts. Machine learning algorithms are employed to improve prediction accuracy, establishing short-term, medium-term, and long-term prediction models. The available grid capacity and distributed generation status are assessed, and the system's power supply capacity and constraints are analyzed. In this process, relevant grid system objectives and constraints must also be considered; the grid operator's objective function is designed as follows: In the formula, The cost is the power grid operating cost, where T is the number of time periods. For the power generation cost function, Let be the power generated at time t. For network loss cost function, For network loss, For peak cost function, This represents peak load.
[0036] The power generation cost function uses a quadratic function: In the formula, a, b, and c are power generation cost coefficients.
[0037] The load balance constraint is: In the formula, N represents the number of electric vehicles. The charging and discharging power of the i-th vehicle (charging is positive, discharging is negative). It is the base load.
[0038] In another possible implementation, when assessing the available capacity of the power grid and the status of distributed power sources, smart meters can be used to monitor the power and voltage of each node in the power grid in real time. Combined with historical data and weather forecasts, the grid load situation over a future period can be predicted, thereby assessing the available capacity of the power grid. For distributed power sources, monitoring equipment installed on the power source side can acquire information such as power generation capacity and efficiency in real time, and comprehensively assess their status.
[0039] In another possible implementation, when assessing the available capacity of the power grid and the status of distributed power sources, simulation models of the power grid and distributed power sources can be established. Current operating data and relevant parameters are input to simulate the power grid operation under different conditions, thereby assessing the available capacity of the power grid and the status of distributed power sources. Simultaneously, data sharing and interaction with surrounding power grids can be conducted to obtain more comprehensive power grid information.
[0040] Design a demand response potential assessment method to analyze users' willingness and ability to participate in orderly charging. Establish a risk assessment mechanism to identify uncertainties that may affect the execution of the charging plan. Construct a resource optimization allocation strategy to rationally allocate charging time slots and power limits. Simultaneously, consider a robust optimization model that account for travel uncertainties. In the formula, For an uncertain parameter vector, It is an uncertain set.
[0041] A set of uncertain travel times: In the formula, For the nominal value of an uncertain parameter, For uncertain budget parameters.
[0042] In this embodiment, step S3 above involves constructing a game model between electric vehicle users and grid operators based on the prediction and evaluation results, defining the strategy space and payoff function, and solving for the equilibrium solution to obtain the game equilibrium result, including: A Stackelberg game model is established between electric vehicle users and grid operators, defining the strategy space and payoff function for each party. A two-level optimization model is constructed based on the Stackelberg game, with grid operators as leaders and electric vehicle users as followers. The leader problem is as follows: In the formula, For the profit of power grid operators, The charging power is optimized for the user's response. This represents the total operating cost.
[0043] The follower problem is: Equation of the state of charge: In the formula, For charging efficiency, This refers to the battery capacity.
[0044] Design an incentive-compatible mechanism to ensure users are motivated to honestly report their needs and preferences. Design an incentive-compatible pricing strategy based on the VCG mechanism: In the formula, Let i be the marginal price. Basic electricity price, For social welfare other than user i.
[0045] The social welfare function is defined as: In the formula, For social welfare, This refers to the operating costs of the power grid.
[0046] Dynamic pricing function: In the formula, Total charging power For reference power, , To adjust the parameters.
[0047] In another possible implementation, when constructing the two-layer optimization model, objective functions can be established separately, with the goals of maximizing the profits of the grid operator and maximizing the utility of electric vehicle users. For the constraints on the grid operator, factors such as grid safety requirements and power generation costs can be considered; for the constraints on electric vehicle users, factors such as vehicle state of charge and charging time limitations can be considered. Then, an optimization algorithm is used to solve the two-layer optimization model.
[0048] In another possible implementation, uncertainties such as electricity price fluctuations and changes in user demand can be introduced when constructing the two-layer optimization model to build a stochastic two-layer optimization model. Monte Carlo simulations and other methods can be used to solve the model under different scenarios, yielding more robust optimization results. Simultaneously, the model parameters can be continuously adjusted based on actual operating conditions to improve the model's accuracy and adaptability.
[0049] Game theory methods are used to analyze the existence and uniqueness of equilibrium solutions. A distributed optimization algorithm is employed to solve the game equilibrium, and an efficient iterative algorithm is designed to reduce computational complexity. The Alternating Direction Multiplier Method (ADMM) is used to solve the distributed optimization problem. In the formula, To augment the Lagrange function, Let i be the objective function. For coupling constraint functions, As dual variables, Let be the coupling matrix. This is the penalty parameter.
[0050] ADMM Iterative Update Rules: In the formula, k represents the number of iterations.
[0051] Establish convergence criteria and termination conditions to ensure that the algorithm converges to a satisfactory solution within a reasonable time. Convergence conditions for the ADMM algorithm: In the formula, For the original residual, For dual residuals, , To achieve convergence tolerance.
[0052] Decrease in objective function: In the formula, This is the descent step size parameter.
[0053] Design an adaptive adjustment mechanism for algorithm parameters to dynamically adjust the parameters based on the problem size and convergence status.
[0054] In this embodiment, step S4 above, based on the game equilibrium result, formulates a dynamic electricity pricing strategy and a differentiated pricing mechanism, and releases price signals through multiple channels. Simultaneously, it designs a price response evaluation and adjustment mechanism, including: A dynamic electricity pricing strategy is formulated based on the game equilibrium results, taking into account load balance, network constraints, and user affordability. A differentiated pricing mechanism is designed to develop personalized price signals for different types of users. A price signal dissemination mechanism is established to promptly convey price information to users through multiple channels.
[0055] Design a method for evaluating the effectiveness of price responses and analyze the impact of price signals on user behavior. Establish a price adjustment mechanism to dynamically adjust pricing strategies based on response effectiveness and system status. Construct a price fairness evaluation system to ensure the rationality and acceptability of pricing strategies.
[0056] In this embodiment, step S5 above, based on the published price signal and the user's own needs, executes a distributed control algorithm to enable the user to formulate and execute a charging plan, while simultaneously monitoring and handling abnormal situations in real time, including: Users formulate charging plans based on price signals and their own needs, and optimization algorithms determine the optimal charging strategy. A decentralized control algorithm is implemented, with each electric vehicle executing its charging plan based on local information and coordination signals. This involves an aggregator coordination mechanism, and the aggregator's optimization objective is: In the formula, G represents the aggregator's profit, and G is the collection of electric vehicles managed by the aggregator. For electricity sales price, For electricity purchase price, For operating costs.
[0057] Profit-sharing mechanism between aggregators and users: In the formula, The revenue share for user i To divide the profits proportionally, This is the benchmark profit.
[0058] Establish a real-time monitoring and feedback mechanism to track the execution and deviations of the charging plan. Specifically, a model predictive control framework will be adopted. In the formula, To predict the time domain, To predict the output, For reference trajectory, Q and R are weight matrices used to control the input.
[0059] Rolling optimization updates: In the formula, To determine the optimal charging power for the next moment. This is the current state.
[0060] Design an anomaly handling mechanism to address unexpected situations such as equipment failure, communication interruptions, and changes in user behavior. Establish a coordinated correction mechanism to adjust control strategies promptly when the system deviates from its expected state. Construct a performance evaluation system to assess control effectiveness and system performance in real time.
[0061] In this embodiment, step S6 above, which evaluates the control effect based on system operating data and user feedback, and optimizes system parameters and strategies through a feedback learning mechanism to achieve continuous improvement, includes: Establish a multi-dimensional evaluation index system to assess control effectiveness from the perspectives of system benefits, user satisfaction, and environmental benefits. Regularly analyze changes in user behavior and system operational characteristics to identify areas for optimization and improvement. Establish a feedback learning mechanism to continuously refine control strategies and algorithm parameters based on operational experience.
[0062] Design a user satisfaction survey mechanism to collect user feedback on service quality and price levels. Establish a continuous improvement process to optimize the system design based on evaluation results and user feedback. Build a knowledge accumulation and experience sharing platform to support future system upgrades and wider application.
[0063] Example 3: The above is an illustrative scheme of an ordered charging and discharging distributed cooperative control method for electric vehicle fleets according to this embodiment. It should be noted that the technical solution of an ordered charging and discharging distributed cooperative control system for electric vehicle fleets and the technical solution of the ordered charging and discharging distributed cooperative control method for electric vehicle fleets described above belong to the same concept. Details not described in detail in the technical solution of the ordered charging and discharging distributed cooperative control system for electric vehicle fleets in this embodiment can be found in the description of the technical solution of the ordered charging and discharging distributed cooperative control method for electric vehicle fleets described above.
[0064] This embodiment also provides an ordered charging and discharging distributed cooperative control system for electric vehicle fleets, including: The user information management and classification module is used to collect basic user information, vehicle parameters and charging preferences, establish an electric vehicle user database, and use behavioral feature recognition algorithms and classification models to classify users. It also designs an incentive registration mechanism, a credit evaluation system and a privacy protection framework. The charging demand forecasting and risk assessment module is used to predict charging demand based on the electric vehicle user database and historical data, and to assess the available grid capacity and distributed power source status through machine learning algorithms. It also analyzes demand response potential and conducts risk assessment. The vehicle-grid game equilibrium solution module is used to construct a game model between electric vehicle users and grid operators based on prediction and evaluation results, define the strategy space and payoff function, and solve for the equilibrium solution to obtain the game equilibrium result. The dynamic electricity pricing and evaluation module is used to formulate dynamic electricity pricing strategies and differentiated pricing mechanisms based on game equilibrium results, and to release price signals through multiple channels, while designing price response evaluation and adjustment mechanisms. The charging plan execution and monitoring module is used to execute a distributed control algorithm based on the published price signals and the user's own needs, enabling the user to formulate and execute a charging plan, while monitoring and handling abnormal situations in real time. The system performance evaluation and optimization module is used to evaluate the control effect based on system operation data and user feedback, and optimize system parameters and strategies through a feedback learning mechanism to achieve continuous improvement.
[0065] This embodiment also provides an electronic device applicable to an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement an ordered charging and discharging distributed cooperative control method for electric vehicle fleets, as proposed in the above embodiments.
[0066] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements an ordered charging and discharging distributed cooperative control method for electric vehicle groups as proposed in the above embodiments.
[0067] The storage medium proposed in this embodiment belongs to the same inventive concept as the orderly charging and discharging distributed cooperative control method for electric vehicle groups proposed in the above embodiment. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.
[0068] Example 4, referring to Tables 1-3, is an embodiment of the present invention, providing an ordered charging and discharging distributed cooperative control method for electric vehicle fleets. To verify the beneficial effects of the present invention, a simulation experiment is conducted for scientific demonstration.
[0069] This embodiment uses MATLAB simulation to verify the simulation based on real-world travel data from 1000 electric vehicles in a city, considering different travel patterns on weekdays and weekends. The simulation system includes 600 home charging stations, 200 public charging stations, and 200 workplace charging stations. Electricity pricing adopts a time-of-use pricing mechanism: peak hour price is 0.85 yuan / kWh, weekday price is 0.55 yuan / kWh, and off-peak price is 0.25 yuan / kWh.
[0070] Table 1: Performance Comparison of Different Control Methods
[0071] Table 2: Results of User Behavior Analysis
[0072] Table 3: System Performance Analysis
[0073] Simulation results show that the distributed collaborative control method proposed in this invention significantly outperforms existing methods in terms of user satisfaction, system benefits, and computational efficiency. User satisfaction is increased to 89.6%, the peak-to-valley load difference is reduced by 37.5%, charging costs are reduced by 22.8%, and system benefits are improved by 31.4%. Different types of users exhibit significant differences, with price-sensitive users showing the highest participation rate, while environmentally conscious users, although fewer in number, have the strongest willingness to participate. The system is highly effective in guiding charging during off-peak hours, increasing the charging rate to 85.6% during the 22:00-06:00 period, effectively alleviating peak-hour pressure on the power grid.
[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A distributed cooperative control method for orderly charging and discharging of electric vehicle fleets, characterized in that, include: Collect basic user information, vehicle parameters, and charging preferences to establish an electric vehicle user database. Use behavioral feature recognition algorithms and classification models to segment users, and design incentive registration mechanisms, credit evaluation systems, and privacy protection frameworks. Based on the electric vehicle user database and combined with historical data, the system uses machine learning algorithms to predict charging demand, assess the available grid capacity and distributed power supply status, analyze demand response potential, and conduct risk assessment. Based on the prediction and evaluation results, a game model between electric vehicle users and grid operators is constructed, the strategy space and payoff function are defined, and the equilibrium solution is solved to obtain the game equilibrium result. Based on the game equilibrium results, a dynamic electricity pricing strategy and a differentiated pricing mechanism are formulated, and price signals are released through multiple channels. At the same time, a price response evaluation and adjustment mechanism is designed. Based on the released price signals and the user's own needs, a decentralized control algorithm is executed to enable the user to formulate and execute a charging plan, while monitoring and handling abnormal situations in real time. Based on system operation data and user feedback, the control effect is evaluated, and system parameters and strategies are optimized through a feedback learning mechanism to achieve continuous improvement.
2. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 1, characterized in that, The process of collecting basic user information, vehicle parameters, and charging preferences to establish an electric vehicle user database, and using behavioral feature recognition algorithms and classification models to segment users, while also designing an incentive registration mechanism, a credit rating system, and a privacy protection framework includes: User behavior is modeled, and the user utility function is defined as a comprehensive consideration of power satisfaction, charging cost, and time preference penalty; the power satisfaction function adopts a piecewise calculation logic based on whether the state of charge has reached the target value; the charging cost function includes electricity price cost and waiting cost; Design a user behavior feature recognition algorithm, analyze user travel patterns, charging habits and price sensitivity, establish a user classification model, and divide users into types based on behavioral features; A credit rating system is established to assess user credibility based on historical participation and performance records; the privacy protection framework adopts differential privacy protection to limit the differences in output results of adjacent datasets, and adds noise to the charging power through a noise addition mechanism.
3. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 2, characterized in that, The process, based on an electric vehicle user database and combined with historical data, uses machine learning algorithms to predict charging demand, assess grid capacity and distributed power generation status, analyze demand response potential, and conduct risk assessments, including: Grid operators consider operating costs with the objective function of minimizing generation costs, network loss costs, and peak costs, while adhering to load balance constraints, namely, the sum of generation power and electric vehicle charging and discharging power equals the sum of base load and network losses; Design a demand response potential assessment method to analyze users' willingness and ability to participate in orderly charging; establish a risk assessment mechanism to identify uncertainties affecting the execution of charging plans; and construct a resource optimization allocation strategy to rationally allocate charging time periods and power limits.
4. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 3, characterized in that, Based on the prediction and evaluation results, a game model is constructed between electric vehicle users and grid operators. The strategy space and payoff function are defined, and the equilibrium solution is solved to obtain the game equilibrium result, including: Define the strategy space and revenue function for grid operators and electric vehicle users, and construct a two-level optimization model. The grid operator is the leader with the goal of maximizing profits, and the electric vehicle user is the follower with the goal of maximizing its own utility. The evolution of the state of charge during the charging process is also considered.
5. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 4, characterized in that, The process of constructing a game model between electric vehicle users and grid operators based on prediction and evaluation results, defining the strategy space and payoff function, and solving for the equilibrium solution to obtain the game equilibrium result also includes: The existence and uniqueness of equilibrium solutions are analyzed using game theory methods, and a distributed optimization algorithm is used to solve the game equilibrium. Convergence criteria and termination conditions are established, including the convergence tolerance of the original residuals and dual residuals, as well as the requirement for the descent of the objective function.
6. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 5, characterized in that, The process of executing a distributed control algorithm based on published price signals and user needs, enabling users to create and execute charging plans, while simultaneously monitoring and handling anomalies in real time includes: Users formulate charging plans based on price signals and their own needs using optimization algorithms, and implement distributed control algorithms. Each electric vehicle executes the charging plan based on local information and coordination signals. An aggregator coordination mechanism is introduced. Aggregators aim to maximize their own profits. There is a profit-sharing mechanism between aggregators and users. Users receive a share of the aggregator's profits exceeding the benchmark profit according to a certain percentage.
7. The ordered charging and discharging distributed cooperative control method for electric vehicle fleets as described in claim 6, characterized in that, The process of executing a decentralized control algorithm based on published price signals and user needs, enabling users to create and execute charging plans, while simultaneously monitoring and handling anomalies in real time, also includes: A real-time monitoring and feedback mechanism is established, and a model predictive control framework is used to track the execution status and deviation of the charging plan. The model predictive control framework achieves rolling optimization updates by minimizing the deviation between the predicted output and the reference trajectory and the cost of the control input, and determines the optimal charging power at the next moment.
8. An ordered charging and discharging distributed cooperative control system for electric vehicle fleets, employing the method described in any one of claims 1 to 7, characterized in that, include: The user information management and classification module is used to collect basic user information, vehicle parameters and charging preferences, establish an electric vehicle user database, and use behavioral feature recognition algorithms and classification models to classify users. It also designs an incentive registration mechanism, a credit evaluation system and a privacy protection framework. The charging demand forecasting and risk assessment module is used to predict charging demand based on the electric vehicle user database and historical data, and to assess the available grid capacity and distributed power source status through machine learning algorithms. It also analyzes demand response potential and conducts risk assessment. The vehicle-grid game equilibrium solution module is used to construct a game model between electric vehicle users and grid operators based on prediction and evaluation results, define the strategy space and payoff function, and solve for the equilibrium solution to obtain the game equilibrium result. The dynamic electricity pricing and evaluation module is used to formulate dynamic electricity pricing strategies and differentiated pricing mechanisms based on game equilibrium results, and to release price signals through multiple channels, while designing price response evaluation and adjustment mechanisms. The charging plan execution and monitoring module is used to execute a distributed control algorithm based on the published price signals and the user's own needs, enabling the user to formulate and execute a charging plan, while monitoring and handling abnormal situations in real time. The system performance evaluation and optimization module is used to evaluate the control effect based on system operation data and user feedback, and optimize system parameters and strategies through a feedback learning mechanism to achieve continuous improvement.
9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, It stores computer-executable instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 7.