Nash negotiation-based energy management scheduling method for light storage and charging integrated energy station

Abstract

The invention discloses a Nash negotiation-based energy management scheduling method for an optical storage and charging integrated energy station, and the method comprises the steps: constructing a multi-target optimization model which comprehensively considers the interest demands of an electric vehicle user, a charging station operator and a power grid operator; determining participants of Nash negotiation as a charging station operator, a power grid operator and an electric vehicle user, and determining a strategy space of each participant; on the basis of the Nash negotiation theory, all the benefit subjects are regarded as negotiation participants, utility functions of all the benefit subjects are constructed, and negotiation reference points of all the benefit subjects are determined; solving a Nash equilibrium solution of the multi-objective optimization model, and determining an optimal energy management scheduling strategy through iterative negotiation; and according to the optimal energy management scheduling strategy, real-time control and scheduling are carried out on the optical storage and charging integrated energy station. According to the invention, the flexibility, adaptability and economy of an energy scheduling scheme can be improved, and multi-party win-win and stable and efficient operation of the system are realized.

Description

Technical Field

[0001] The present invention belongs to the field of power systems, and in particular relates to an energy management and scheduling method for a photovoltaic storage and charging integrated energy station based on Nash negotiation. Background Art

[0002] As the global energy transition progresses, traditional fossil fuels are gradually transitioning to renewable energy. Photovoltaic power generation, as a key component of renewable energy, has garnered widespread attention. Against this backdrop, integrated photovoltaic (PV)-storage-charging (ESC) charging stations, as an emerging energy management model, are gaining increasing attention from both academia and industry. These stations combine PV generation, energy storage systems, and electric vehicle (EV) charging facilities. Their core advantage lies in their ability to maximize solar power generation, reducing the burden on traditional power grids. They also leverage energy storage systems to balance power supply during peak charging demand periods, mitigating the impact of grid fluctuations. These systems not only optimize energy efficiency and reduce operating costs, but also enhance the convenience of EV charging. While this model demonstrates significant potential in both theory and practice, its energy management and optimal scheduling face a series of critical issues and challenges that need to be addressed.

[0003] The performance and management of energy storage systems are directly related to the overall operational efficiency of integrated solar-storage-charging charging stations. The charging and discharging scheduling of energy storage equipment must not only consider the battery lifecycle and efficiency, but also respond in real time to changes in photovoltaic power generation and electric vehicle demand. Electric vehicle charging demand fluctuates significantly, and the uncertainty of charging times and locations adds additional complexity to the energy scheduling of electric vehicle charging stations. Without effective charging demand forecasting and optimized scheduling strategies, charging stations may face energy shortages, excessive loads, or poor charging station service quality. Integrated solar-storage-charging charging stations must not only achieve self-scheduling but also interact with the power grid. Grid load fluctuations can impact charging station operations. Optimizing the scheduling of charging piles to prevent charging stations from placing excessive pressure on the power grid is another key issue facing integrated solar-storage-charging charging stations.

[0004] The operation of integrated solar-storage-charging (PSC) charging stations involves multiple stakeholders, including electric vehicle users, charging station operators, grid operators, and energy storage system managers. Each stakeholder has distinct and often conflicting goals and interests. For example, electric vehicle users desire fast charging, charging station operators seek to optimize equipment efficiency, the grid seeks to balance loads and maintain stable operation, and the energy storage system needs to maximize its service life and efficiency. Existing scheduling optimization methods often fail to consider the coordination of these interests and are limited to optimizing a single objective or maximizing a single entity. Scheduling strategies that consider multiple stakeholders involve optimizing multiple objectives, such as minimizing operating costs, maximizing EV charging benefits, and balancing grid load. Designing efficient algorithms to solve these multi-objective optimization problems and ensuring their computational efficiency and accuracy in practical applications is another major challenge in this field. Effectively solving multi-objective optimization problems that both improve system efficiency and ensure rapid response in dynamic environments is key to the successful application of integrated solar-storage-charging (PSC) charging stations.

[0005] In summary, energy management and optimal scheduling for integrated photovoltaic (PV) and energy storage charging stations is a complex and multifaceted issue, encompassing challenges ranging from the uncertainty of photovoltaic power generation, the charge and discharge characteristics of energy storage systems, the dynamic changes in electric vehicle charging demand, to grid load fluctuations and electricity market prices. Addressing these issues is not only crucial for improving the operational efficiency and economic benefits of charging stations but also has far-reaching implications for promoting the widespread adoption of green energy and the widespread adoption of electric vehicles. Therefore, designing efficient energy management and optimal scheduling strategies has become a core issue that urgently needs to be addressed in this field. Summary of the Invention

[0006] The purpose of this invention is to provide an energy management and scheduling method for a photovoltaic, storage and charging integrated energy station based on Nash negotiation, which helps charging stations to rationally schedule electric vehicle charging, photovoltaic power generation and energy storage discharge to achieve optimal economic benefits and ensure the sustainability and profitability of power station operations. At the same time, power stations can flexibly adjust strategies according to actual changes in electricity prices to optimize operating profits.

[0007] In order to achieve the above object, the solution of the present invention is:

[0008] A method for energy management and scheduling of a photovoltaic storage and charging integrated energy station based on Nash negotiation includes the following steps:

[0009] Step 1: Construct a multi-objective optimization model that comprehensively considers the interests of electric vehicle users, charging station operators, and power grid operators;

[0010] Step 2: Determine the participants in the Nash negotiation as charging station operators, power grid operators, and electric vehicle users, and determine the strategy space of each participant;

[0011] Step 3: Based on Nash negotiation theory, consider each stakeholder as a negotiation participant, construct the utility function of each stakeholder, and determine the negotiation benchmark of each stakeholder;

[0012] Step 4: solving the Nash equilibrium solution of the multi-objective optimization model and determining the optimal energy management scheduling strategy through iterative negotiation;

[0013] Step 5: According to the optimal energy management scheduling strategy, the photovoltaic storage and charging integrated energy station is controlled and scheduled in real time.

[0014] In step 1 above, the multi-objective optimization model includes three objectives: maximizing the overall operating revenue of the integrated solar-storage-charging power station, reducing the charging costs of electric vehicle owners, and reducing the degree of grid load fluctuation;

[0015] Among them, to maximize the overall operating income of the integrated photovoltaic storage and charging power station, the objective function is:

[0016] R station =R EV +R grid -C ess -C pv -C purchase +λE pv,util

[0017] Among them, R station is the charging station target, R PV is the revenue from providing charging services to electric vehicles, R grid is the revenue from selling electricity to the grid, C ess is the total operating cost of the energy storage system, C pv is the operation and maintenance cost of the photovoltaic power generation system, C purchase is the cost of purchasing electricity from the grid, E pv,util is the goal of maximizing the photovoltaic consumption of renewable energy, and λ is the conversion coefficient;

[0018] Among them, to reduce the charging cost of electric vehicle owners, the objective function is:

[0019]

[0020] Among them, T is the scheduling period, C EV is the goal of the electric vehicle owner; ω1, ω2, ω3 are weight coefficients, which respectively represent the importance of charging cost, charging power matching and charging waiting time in satisfaction; P ev,t is the charging power of the electric vehicle, Price ev,t is the time-sharing charging price; Pev * Represents the predicted charging demand power value at the time, T wait is the average charging waiting time;

[0021] Among them, to reduce the degree of grid load fluctuation, the objective function is:

[0022]

[0023] Among them, L fluct is the load fluctuation index of the power grid, P grid,t is the power interaction with the grid at time t.

[0024] In step 2 above, the strategy space of Nash negotiation is determined, including:

[0025] Determine the strategy space for charging station operators, including the charge and discharge control strategy of the energy storage system and the energy interaction strategy of the power grid;

[0026] Determine the grid operator's strategy including electricity price incentives or constraints for integrated energy stations;

[0027] Determining the strategy of electric vehicle users includes choosing the charging time and charging power.

[0028] In step 3 above, the utility function of each stakeholder is constructed, including:

[0029] The utility function of the charging station operator is constructed as:

[0030]

[0031] Among them, U station Utility for charging station operators, It is an alternative location for charging stations;

[0032] The utility function of electric vehicle users is constructed as follows:

[0033]

[0034] Among them, U EV Utility for electric vehicle users, An alternative point for car owners;

[0035] The utility function of the grid operator is constructed as,

[0036]

[0037] Among them, U grid Utility for grid operators, It is an alternative point for the power grid.

[0038] In step 4 above, the optimal energy management scheduling strategy is determined through Nash negotiation, including:

[0039] Construct the objective function of Nash negotiation solution;

[0040] Construct the decision variables and constraints of Nash negotiation;

[0041] On the basis of satisfying their own constraints, each stakeholder adjusts their own strategy through multiple iterative negotiations until a Nash equilibrium is reached, and the strategy combination at this time is used as the optimal energy management scheduling strategy.

[0042] Among them, the objective function of constructing the Nash negotiation solution includes:

[0043]

[0044] Among them, U total is the product of the three-party utility, which is the objective function of the optimization algorithm. The larger the better. λ1, λ2, and λ3 are the weights of charging station operators, electric vehicle users, and power grid operators, respectively. R station For charging station targets, It is an alternative location for charging stations; is the car owner's alternative point, C EV Target for tram owners; is the alternative point of the power grid, L fluct It is the load fluctuation index of the power grid.

[0045] Among them, each party adjusts its strategy through multiple iterative negotiations until a Nash equilibrium is reached, including:

[0046] The improved gradient descent method is used to solve the problem, which includes the following steps:

[0047] S201: Initialize decision variables;

[0048] S202: Calculate the utility function and its gradient under the current decision variable;

[0049] S203: Update decision variables according to gradient and learning rate;

[0050] S204: Check the convergence conditions. If not, return to step S202 for the next iteration. If converged, output the optimal solution.

[0051] Among them, the respective strategies were adjusted through multiple iterative negotiations, including:

[0052] Obtain historical and real-time data from the integrated solar-storage-charging energy station, including storage state-of-charge data and grid electricity price information;

[0053] Obtaining photovoltaic power generation prediction data and charging demand prediction data based on the historical data and real-time data;

[0054] Iterative negotiation is performed based on the photovoltaic power generation prediction data and the charging demand prediction data.

[0055] Among them, obtaining photovoltaic power generation prediction data includes:

[0056] According to the prediction model P based on neural network PV,t * =f PV (I t ,T t ,H t ) to obtain photovoltaic power generation prediction data, where P PV,t * represents the predicted value of photovoltaic power generation at time t, I t ,T t ,H t They represent the meteorological parameters such as light intensity, temperature, and humidity at time t, respectively. PV is the photovoltaic power prediction function;

[0057] Obtain charging demand forecast data, including,

[0058] According to the charging demand prediction model P ev * =g C (t,n EV,t ,E req,t ) obtain charging demand forecast data, where P ev * represents the charging demand power forecast value at the moment, t is the time variable, n ev,t represents the number of electric vehicles at time t, E req,t represents the average charging energy demand of a single electric vehicle at time t, g C is the charging demand prediction function.

[0059] The above method also includes real-time monitoring of the operating parameters of the photovoltaic storage and charging integrated energy station, including photovoltaic power generation power, energy storage system charge state, charging power, and interaction power with the power grid; when the actual operating parameters deviate from the predetermined strategy or the actual utility value of each stakeholder decreases, the charging power distribution of the charging facilities and the energy interaction power with the power grid are adjusted according to the power grid electricity price and energy storage status.

[0060] After adopting the above scheme, the present invention has the following advantages:

[0061] Based on Nash negotiation, the present invention provides a new coordination mechanism for multi-party conflicts of interest in charging station energy management, enabling all parties to achieve a win-win situation on a fair basis. Unlike traditional centralized optimization methods, the present invention optimizes the decision-making of all parties through a game theory model, thereby improving the satisfaction of all parties and the overall efficiency of the system. The present invention comprehensively considers the multi-dimensional coordinated scheduling of photovoltaics, energy storage, charging demand and power grids, and optimizes the adaptability and flexibility of the scheduling strategy. Through the combination of dynamic optimization and Nash negotiation, the system can respond to changes in charging demand in real time, adjust the strategies of all parties, and improve operational efficiency. At the same time, the algorithm of the present invention optimizes the decision-making in multi-agent games through a negotiation mechanism, effectively controls computational complexity, and meets real-time scheduling requirements. Compared with traditional static scheduling methods, the present invention significantly improves the flexibility, adaptability and real-time performance of the scheduling system, ensures the short-term benefits of the system, return on investment and equipment life, and improves stability and response speed in a dynamic environment.

[0062] The improvements of the present invention are embodied in the following aspects:

[0063] 1) Nash negotiation model based on multi-objective optimization:

[0064] This paper constructs a multi-objective optimization model that comprehensively considers the interests of multiple stakeholders, including electric vehicle users, charging station operators, and power grid operators. This model quantifies the objectives of each party, such as the operating revenue of charging stations, the charging costs and satisfaction of electric vehicle users, and load fluctuations in the power grid. It models the energy management and scheduling problem of integrated solar-storage-charging energy stations as a multi-objective optimization problem, and clearly defines the objective functions and constraints for each stakeholder.

[0065] 2) Application of Nash negotiation mechanism in energy management scheduling:

[0066] This paper introduces Nash negotiation theory, treating all stakeholders as negotiating participants. By constructing a utility function and determining a negotiation benchmark, it achieves a balanced distribution of interests and collaborative decision-making among all parties. By applying Nash negotiation theory to the energy management and scheduling of integrated solar-storage-charging energy stations, this paper uses iterative negotiation to find the Nash equilibrium point, thereby determining the optimal energy management and scheduling strategy.

[0067] 3) Coordinated dispatch framework of comprehensive energy elements:

[0068] The present invention incorporates key energy factors such as the uncertainty of photovoltaic power generation, the charging and discharging characteristics of the energy storage system, the dynamic charging requirements of electric vehicles, and the volatility of the grid load into a unified scheduling framework for optimization. By collaboratively optimizing the operating strategies of these factors, energy utilization efficiency and overall system performance are improved.

[0069] 4) Improved heuristic algorithm solution:

[0070] To address large-scale computational demands, this paper proposes an improved gradient descent method that combines the momentum method and RMSprop to improve the algorithm's convergence speed and stability, while ensuring accuracy and reducing computational complexity. An improved heuristic algorithm (e.g., a gradient descent method combining momentum and RMSprop) is used to solve the constructed multi-objective optimization model and Nash negotiation model, enabling efficient real-time energy management scheduling.

[0071] 5) Real-time monitoring and dynamic adjustment mechanism:

[0072] The present invention also establishes a real-time monitoring system for the energy management scheduling strategy, which can monitor key operating parameters and the utility of all parties, and dynamically adjust and optimize according to the actual operating conditions to ensure that the system always operates in a near-optimal state. It can dynamically adjust the scheduling strategy according to the actual operating status and the utility feedback of all parties, thereby improving the adaptability and robustness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 This is a schematic diagram of the energy scheduling optimization process based on Nash negotiation;

[0074] Figure 2 It is a schematic diagram of the dynamic adjustment and game process of Nash negotiation;

[0075] Figure 3 This is a detailed diagram of the solution process of the improved gradient descent method;

[0076] Figure 4 It is a schematic diagram of energy management scheduling strategy implementation and monitoring. DETAILED DESCRIPTION

[0077] The technical solutions and beneficial effects of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0078] Figure 1 This paper demonstrates the overall process of an energy management and scheduling method for a photovoltaic, energy storage, and charging integrated energy station based on Nash negotiation. First, data collection and analysis are performed to provide a data foundation for subsequent model development. Next, a multi-objective optimization model that considers the interests of multiple parties is constructed. Next, the Nash negotiation participants and their strategy space are determined. Based on Nash negotiation theory, the model is solved to obtain the optimal scheduling strategy. Finally, this strategy is implemented, and real-time monitoring and dynamic adjustments are performed to ensure efficient and stable system operation.

[0079] In the energy management and dispatch system for integrated solar-storage-charging energy stations, the Nash negotiation model is introduced to coordinate the behavioral choices of the grid operator, electric vehicle users, and charging station operators under multi-objective optimization. As game participants, each possesses independent utility functions and strategy spaces. The grid focuses on system load balancing and minimizing volatility, electric vehicle users focus on charging convenience and economic efficiency, and charging station operators aim to maximize revenue.

[0080] In this model, each participant first generates a set of feasible strategies based on the current system state and forecast data. The energy management system then calculates the joint utility value of each strategy based on the utility functions of each participant. To achieve Pareto optimal resource allocation, the system iteratively adjusts each participant's strategy through a Nash negotiation mechanism until an equilibrium solution is reached that satisfies all parties. This equilibrium point represents the optimal scheduling strategy for the three parties for the current period and is output to the execution system as part of the scheduling module.

[0081] In the entire Nash negotiation model, information flows from each participant, feeding their preference functions and constraints into the negotiation center (the EMS scheduler). The scheduler, through multiple iterations, continuously searches for the point of maximum joint utility that satisfies the constraints and outputs the final scheduling instructions. This structure ensures that the scheduling strategy is not only globally optimal but also feasible and fair in practical engineering operations.

[0082] like Figure 2 The Nash negotiation process diagram shown in Figure 2 illustrates how electric vehicle users, charging station operators, and grid operators in a solar-storage-charging integrated energy station reach a Nash equilibrium through dynamic game-playing. In the energy management and scheduling of a solar-storage-charging integrated energy station, multiple stakeholders, including the grid, energy storage equipment, electric vehicle charging stations, and vehicle owners, participate in decision-making and optimization. Each stakeholder has its own independent decision space and utility function. These utility functions quantify each stakeholder's goals and preferences, helping to balance the interests of all parties.

[0083] The power grid's strategy space encompasses regulating system loads, charging and discharging instructions for energy storage devices, and coordinated interactions with other entities. Its utility function primarily considers operating costs, load balancing, and system stability, with the goal of ensuring a stable power supply, mitigating fluctuations, and optimizing resource allocation. The energy storage device's strategy space focuses on charging and discharging strategies and capacity adjustment. Its utility function is based on charging and discharging costs, energy loss, and health status, with the goal of extending battery life and minimizing energy loss. The strategy space for electric vehicle charging stations encompasses charging time, power selection, and policy scheduling. Its utility function considers charging efficiency, owner needs, and operating costs, with the goal of improving charging efficiency and minimizing impact on the grid. The owner's strategy space encompasses charging time, power selection, and whether to accept charging station scheduling. Its utility function focuses on charging demand, convenient time slots, and cost minimization, with the goal of meeting demand and reducing charging costs. Using a game theory framework, the strategy spaces and utility functions of each entity can be optimized collaboratively across multiple entities. Ultimately, scheduling is optimized through a central control system, ensuring a balanced benefit for all parties and maximizing system economics and stability.

[0084] The process is centered around multiple rounds of iterative negotiation. Each stakeholder first proposes a scheduling plan based on an initial strategy, including the charging station operator's energy storage charging and discharging plan, the grid operator's electricity pricing incentive strategy, and the EV user's charging demand response. After collecting all the strategies, the central controller calculates the utility function of the three parties under the current strategy combination and assesses whether it meets the Nash equilibrium condition—that is, if any party unilaterally changes its strategy, its own utility cannot be improved. If equilibrium is not reached, the parties adjust their strategies based on the utility gradient: the charging station operator optimizes the energy storage scheduling plan, the grid operator adjusts the time-of-use electricity pricing parameters, and the EV user updates the charging time selection. The adjusted strategies are fed back to the central controller via the communication network for a new round of utility calculation, forming a closed-loop iteration. When the product of the three-party utility functions reaches its maximum value and meets the convergence threshold, the negotiation terminates and the optimal energy management strategy is output. A dynamic learning rate mechanism accelerates convergence throughout the process. The grid operator's load fluctuation indicator serves as a key constraint, providing real-time feedback to the strategy adjustment process to ensure system stability.

[0085] In Nash negotiations, the dynamic adjustment of each party's utility function and the game process are based on the strategic interactions between the participants. This process is described within the framework of game theory, where each party's utility function reflects its preferences, goals, and constraints in the negotiation. The core goal of the game is to find an equilibrium solution through the strategic interactions of the participants, so that each party's utility function is optimized under given conditions.

[0086] The dynamic adjustment of the negotiation process involves repeated rounds of negotiation, with participants adjusting their strategies based on the outcomes of previous rounds and changes in the other party's strategies. In the initial stages, participants may express their expectations and make preliminary decisions based on their own utility functions. These decisions may be influenced by each party's initial estimates of factors such as resource allocation, price, and risk. As the game progresses, participants gradually adjust their strategies through information sharing, compromise, and concessions, aiming to reach a mutually beneficial agreement.

[0087] As the game progresses, each party adjusts its utility function in response to the other's strategies. For example, in the power dispatch of a solar-storage-charging integrated energy station, different stakeholders—the grid, energy storage systems, and electric vehicle charging facilities—will adjust their respective utility functions based on actual demand and changing market conditions. The grid may prioritize system stability and power supply reliability, while the energy storage system focuses on the efficiency of power storage and release. Charging facilities may focus on charging efficiency and cost.

[0088] During the game, each party's utility function continuously adjusts as negotiations progress. This dynamic adjustment reflects the gradual optimization process in the game, until each party's utility function reaches a stable equilibrium point while satisfying the constraints. This equilibrium point is the Nash equilibrium, meaning that each party cannot improve its utility by unilaterally changing its own strategy given the strategies of the other parties.

[0089] The key to this game lies in the transmission and sharing of information, as well as how each party makes rational decisions within the game. Ultimately, after multiple rounds of the game, all parties will find a solution that benefits all parties by adjusting their strategies and utility functions.

[0090] The present invention adopts the improved gradient descent method to solve, Figure 3 A detailed flowchart illustrates the specific steps and logical sequence for solving the optimization problem using an improved gradient descent method in energy management and scheduling for a solar-storage-charging integrated energy station. First, the diagram shows the initialization of decision variables. This step explicitly sets initial values ​​for decision variables such as grid transaction power, photovoltaic power generation, energy storage system charging power, discharge power, and vehicle owner charging power. These initial values ​​form the foundation for subsequent calculations and iterations.

[0091] Next, the step of calculating the utility function and its gradient for the current decision variable is performed. During this process, the utility function gradients are calculated for the grid, the PV storage system, and the vehicle owner. The grid utility function gradient is calculated using a specific formula, taking into account relevant factors. The PV storage utility function gradient also follows a corresponding formula, combining the operating parameters and objective function of the PV storage system. The vehicle owner utility function gradient is also calculated using a corresponding formula, taking into account various factors during the charging process. These gradient calculations provide a directional basis for subsequent decision variable updates.

[0092] The decision variables are then updated based on the calculated gradient and the set learning rate. The learning rate plays a key role in this process, controlling the step size of each update. According to the gradient descent update formula, the decision variables are adjusted in the direction of the gradient, so that the objective function value gradually converges toward the optimal solution. During the update process, the learning rate must be carefully balanced. If the learning rate is too large, it may lead to excessive updates and prevent the algorithm from convergence; if the learning rate is too small, the convergence speed will become very slow.

[0093] After updating the decision variables, the step of checking the convergence conditions is entered. The convergence condition is set as the maximum number of iterations reaching a predetermined upper limit or the change in the objective function is less than a predetermined threshold. If the current situation does not meet the convergence condition, that is, the maximum number of iterations has not been reached and the change in the objective function is still greater than the predetermined threshold, then the process returns to the step of calculating the utility function and its gradient under the current decision variable for the next iterative calculation. By continuously repeating this process, the decision variables are gradually adjusted so that the objective function value is continuously optimized. When the convergence condition is met, the algorithm ends and the optimal solution is output. This optimal solution is the optimal strategy for energy management and scheduling of the photovoltaic storage and charging integrated energy station.

[0094] also, Figure 3 The paper also demonstrates the improvements of the modified gradient descent method over the standard gradient descent method. Combining the concepts of the momentum method and RMSprop, the learning rate of each parameter is adjusted by calculating the moving average of the gradient and the moving average of the squared gradient. Setting the decay rates of the momentum and squared gradient, and correcting for deviations, the algorithm's performance is further optimized, its convergence speed and stability are improved, and this ensures that optimal solutions can be efficiently obtained while maintaining computational accuracy when solving complex optimization problems such as energy management and scheduling in integrated solar-energy-storage-charging stations.

[0095] Figure 4The energy scheduling strategy implementation and monitoring flow chart describes a systematic energy scheduling process, involving aspects such as strategy implementation, real-time monitoring and scheduling adjustment. The core components of the integrated energy station include photovoltaic power generation systems, energy storage systems, charging facilities and connections to the power grid. The photovoltaic power generation system transmits the generated electricity to the energy storage system and charging facilities through specific lines, providing green energy for the entire system. The energy storage system plays a key role in energy buffering and regulation. It can not only store the excess electricity generated by the photovoltaic power generation system, but also release electricity when the system needs it to meet the electricity demand of the charging facilities. Charging facilities are directly aimed at electric vehicle users and are the terminal link for converting electricity into vehicle power.

[0096] First, the system formulates an energy scheduling strategy based on predetermined scheduling objectives and requirements, determining the optimal scheduling plan for batteries, photovoltaics, energy storage, and loads. Based on the optimal energy management scheduling strategy derived from Nash negotiation, the control system precisely controls each component. For example, for the energy storage system, the operating state of the energy storage converter is controlled according to its charging and discharging strategy. When there is excess photovoltaic power generation, the energy storage converter is activated to store excess energy. When charging demand is high and photovoltaic power generation is insufficient, the energy storage converter is controlled to release energy. Regarding the connection to the grid, the power flow of the power converter is adjusted according to the energy interaction strategy with the grid, purchasing electricity from the grid for storage when electricity prices are low, and selling electricity to the grid when electricity prices are high and energy storage is sufficient. For charging facilities, the charging power of each charging station is rationally allocated according to the scheduling strategy, prioritizing critical charging needs and improving charging efficiency and user satisfaction.

[0097] During the implementation phase, the system will issue dispatch instructions to each energy unit through the dispatch control center based on the plan, ensuring optimal energy allocation and scheduling. Simultaneously, a real-time monitoring module monitors the operating status of each energy unit, including energy storage device power, load demand, and photovoltaic power generation, and transmits this data to the monitoring platform. Power generation is monitored at the photovoltaic system to understand energy production; state of charge is monitored at the energy storage system to assess the level of stored energy; charging power is monitored at the charging facility to understand electric vehicle charging demand and usage; and grid interaction power is monitored at the connection point to understand energy exchange with the grid. The monitoring platform dynamically adjusts based on the collected data, providing early warnings and responses to unexpected situations. The monitoring system also calculates the actual utility of each stakeholder and compares it with the negotiated optimal utility. If monitored system operating parameters deviate from the predetermined targets or fluctuate abnormally within a certain period of time, the system automatically activates an adjustment mechanism to restore stable system operation by optimizing the dispatch strategy. For example, when a sudden drop in photovoltaic power generation causes excessive discharge pressure on the energy storage system, the system will appropriately adjust the charging power distribution of the charging facilities, give priority to the charging needs of important users, and decide whether to increase the power purchased from the grid based on the grid electricity price and energy storage status to maintain the stable operation of the integrated energy station and get as close as possible to the optimal energy management scheduling target.

[0098] Finally, the dispatch system continuously adjusts and optimizes its dispatch strategy based on feedback from monitoring data to ensure efficient and stable energy management and dispatch across the entire system. This process emphasizes real-time data, flexible dispatch instructions, and timely feedback and adjustment mechanisms to ensure the efficient operation of the energy dispatch system.

[0099] An embodiment of the present invention provides an energy management and scheduling method for a photovoltaic storage and charging integrated energy station based on Nash negotiation, which specifically includes the following steps:

[0100] S101: Data Collection and Analysis

[0101] Collects power forecast data for photovoltaic power generation systems, state-of-charge data for energy storage systems, charging demand forecast data for charging facilities, and grid electricity pricing information from integrated photovoltaic, storage, and charging stations. The collected data is preprocessed and analyzed to remove abnormal data. Based on historical and real-time data, photovoltaic power forecast models and charging demand forecast models are established to provide accurate data support for subsequent energy management and scheduling decisions.

[0102] S102: Establishing a multi-objective optimization model

[0103] Electric vehicle users, charging station operators, and grid operators are three different stakeholders in an integrated solar-storage-charging power station, each with its own distinct interests. Based on cooperative game theory, the game relationship between the interests of all parties is considered, and the benefits of all parties are maximized by optimizing their utilities. The problem is modeled as a multi-objective optimization problem with three main objectives: maximizing the overall operating benefits of the integrated solar-storage-charging power station, including the depreciation cost of the energy storage system, the grid's electricity purchase costs, the electricity sales revenue, and the photovoltaic power generation revenue; reducing the charging costs of electric vehicle owners, by taking into account the expenses incurred by the owners during the charging process and ensuring their interests; and reducing the degree of grid load fluctuation, by controlling the peak-to-valley difference of the grid and reducing grid load fluctuations.

[0104] S103: Determine the participants and strategy space of Nash negotiation

[0105] Identify the key stakeholders in the integrated solar-storage-charging energy station, including charging station operators, grid operators, and electric vehicle users. For each stakeholder, determine their strategy space in the energy management and scheduling process. For example, the charging station operator's strategy might include the energy storage system's charge and discharge control strategy and its energy interaction strategy with the grid. The grid operator's strategy might involve pricing incentives or constraints for the integrated energy station. The electric vehicle user's strategy might involve selecting charging time and power.

[0106] S104: Model Solution Based on Nash Negotiation

[0107] Based on Nash negotiation theory, a utility function is constructed for each stakeholder. Each stakeholder, while satisfying their own constraints, negotiates and negotiates to determine the optimal energy management scheduling strategy. During the negotiation process, based on the objectives in the multi-objective optimization model, each stakeholder's strategy is continuously adjusted until a Nash equilibrium is reached, where each stakeholder's utility function is optimal under the current strategy and cannot be further improved by unilaterally changing the strategy.

[0108] S105: Energy management scheduling strategy implementation and monitoring

[0109] Based on the optimal energy management and scheduling strategy derived from Nash negotiation, the photovoltaic power generation system, energy storage system, and charging facilities in the integrated photovoltaic, energy storage, and charging station are controlled and scheduled in real time. Simultaneously, a monitoring system is established to monitor the station's operating status in real time, including parameters such as power generation, energy storage system state of charge, charging power, and the actual utility of each stakeholder. If actual operating conditions deviate from the predetermined strategy, timely adjustments and optimizations are implemented to ensure that the integrated energy station always operates according to the optimal energy management and scheduling plan.

[0110] Furthermore, the step S101 includes:

[0111] The photovoltaic power generation prediction model adopts a prediction model based on a neural network, and the specific formula is: P PV,t * =f PV (I t ,T t ,H t ), where P PV,t * represents the predicted value of photovoltaic power generation at time t, I t ,T t ,H t They represent the meteorological parameters such as light intensity, temperature, and humidity at time t, respectively. PV is the photovoltaic power generation power prediction function.

[0112] A charging demand prediction model is established using historical charging data (including charging time distribution, charging power requirements, etc.) and reservation information of electric vehicle users. Preferably, a model is constructed using a time series analysis method:

[0113] P ev * =g C (t,n EV,t ,E req,t )

[0114] Among them, P ev * represents the charging demand power forecast value at the moment, t is the time variable, n ev,t represents the number of electric vehicles at time t, E req,t represents the average charging energy demand of a single electric vehicle at time t, g C is the charging demand prediction function.

[0115] Furthermore, the step S102 includes:

[0116] Define the participants and the objective function

[0117] The goal of the charging station is to maximize overall operating revenue and renewable energy consumption:

[0118] R station =R EV +R grid -C ess -C pv -C purchase +λE pv,util

[0119] Where: R EV is the revenue from providing charging services to electric vehicles,

[0120] Among them, P ev,tis the charging power of the electric vehicle, Price ev,t It is the time-sharing charging price.

[0121] R grid is the revenue from selling electricity to the grid,

[0122] Among them, P grid,t is the power sold by the grid to the outside world, Price sell,t is the electricity selling price of the power grid.

[0123] C ess =C ess,om +C ess,dep

[0124] Among them, C ess is the total operating cost of the energy storage system, C ess,om is the operation and maintenance cost of the energy storage system, which can be expressed as a function related to the charging and discharging power of the energy storage system, C ess,dep is the depreciation cost of the energy storage system, which is related to the capacity and service life of the energy storage system.

[0125] C pv =C pv,om

[0126] Among them, C pv =C pv,om The operation and maintenance cost of the photovoltaic power generation system can be estimated based on the installed capacity and operating time of the photovoltaic power generation system.

[0127] C purchase is the cost of purchasing electricity from the grid,

[0128] Among them, Price buy,t is the unit price of electricity purchased from the grid at time t, P grid,t is the power purchased from the grid at time t.

[0129] E pv,util is the goal of maximizing the photovoltaic consumption of renewable energy, λ is the conversion coefficient,

[0130]

[0131] Among them, T is the scheduling period, P PV,t is the photovoltaic power generation power at time t, P ESS,t is the charging and discharging power of the energy storage system at time t (charging is positive and discharging is negative).

[0132] The charging and discharging power of the energy storage system can be controlled by the decision variables. Assume that the charging power of the energy storage system is P ess,ch (t), the discharge power is P ess,dis(t), controlling these two parameters to regulate the charging and discharging of the energy storage system.

[0133] The goal of electric vehicle owners is to maximize their satisfaction with charging services. The charging cost for electric vehicle owners is determined by the amount of electricity they charge and the charging fee. The formula for the owner's charging fee can be expressed as:

[0134]

[0135] Where: ω1, ω2, ω3 are weight coefficients, which respectively represent the importance of charging cost, charging power matching and charging waiting time in satisfaction, T wait The average charging waiting time.

[0136] In order to reduce the fluctuation of grid load, it is necessary to reduce the amplitude of load fluctuation in the power interaction between the power station and the grid. Set the load fluctuation index of the grid to L fluct , then its goal is to minimize the load fluctuation of the power grid, which can be expressed by the following formula:

[0137]

[0138] Among them, P grid,t is the amount of power interaction (purchase or sale) with the grid at time t. By adjusting the power interaction with the grid, load fluctuations are minimized.

[0139] Furthermore, the step S103 includes:

[0140] Participating entities:

[0141] Charging station operator: Responsible for the comprehensive management and scheduling of energy within the station to maximize operational benefits.

[0142] Grid operators: Focus on the stable operation of the grid, load balance, and efficient allocation of power resources, hoping to achieve grid-side optimization goals through interaction with integrated energy stations.

[0143] Electric vehicle users: pursue the lowest charging cost, fastest charging speed, and convenient and reliable charging services.

[0144] Strategy Space:

[0145] Charging station operators:

[0146] Power control is performed through charging and discharging of the energy storage system, and whether to purchase or sell electricity from the grid and the amount of electricity purchased or sold are determined based on operating costs and energy storage status.

[0147] Power grid companies:

[0148] Set electricity price ranges for integrated energy stations at different time periods to guide their electricity consumption behavior and achieve peak shaving and valley filling of grid load.

[0149] Electric vehicle users:

[0150] Charging time selection strategy: On the premise of meeting your own travel needs, choose a period with lower electricity prices and shorter charging waiting time for charging.

[0151] Furthermore, the step S104 includes:

[0152] Step 401: Establish negotiation benchmarks for each party

[0153] Each party involved will set a bottom line (alternative) point during the "negotiation" process, which is the worst outcome they can accept when no agreement is reached.

[0154] Alternative points for charging stations: In the worst case, the revenue of the power station may be zero, or it may only cover its operating costs, such as the cost of the energy storage system C storage and the grid electricity purchase cost C purchase .

[0155]

[0156] Alternative points for car owners: The worst case scenario for car owners may be that the charging cost is very high, the waiting time is very long, and the charging power matching is very low.

[0157]

[0158] Alternative points for the power grid: The worst case scenario for the power grid is that the load fluctuation is very large, that is, the power grid must dispatch and adjust the load on a large scale to balance the power demand. At this time, the burden on the power grid is very heavy and the load fluctuation is large.

[0159]

[0160] The power grid will reduce fluctuations by adjusting the grid load, so in the worst case, the fluctuation level is high.

[0161] Step 402: Constructing a negotiation utility function

[0162] The utility function is constructed according to the needs and goals of the negotiation. The utility function, that is, the objective function of each participant, determines how to generate utility in the negotiation.

[0163] Power plant utility function:

[0164] Each participant will have a guaranteed utility in the absence of cooperation. The utility function of the power station can be expressed as:

[0165]

[0166] This represents the difference between the utility of the power plant and its actual operating profit and the worst-case profit. The utility of the power plant is maximized through negotiation.

[0167] Car owner utility function:

[0168] The utility function of the car owner is expressed as:

[0169]

[0170] This represents the difference between the owner's utility and their actual charging cost and the worst-case charging cost.

[0171] Grid utility function:

[0172] The utility function of the power grid is expressed as:

[0173]

[0174] This represents the difference between the utility of the grid and its worst-case load fluctuations and the actual fluctuations.

[0175] Step 403: Construct the objective function of the Nash negotiation solution

[0176] The goal of Nash negotiation is to find an optimization strategy that maximizes the utility product of each party. Its objective function can be expressed as:

[0177]

[0178] in:

[0179] U i is the utility function of the ith party.

[0180] is the non-cooperation utility of the i-th party, that is, the bottom line utility.

[0181] The negotiation solution is the product of the utility functions of all participants, and the weighted balance of the interests of all parties. Assuming that the weights of the power station, the car owner, and the grid are λ1, λ2, and λ3 respectively, the final objective function can be expressed as:

[0182] maxU total =λ1·U station ·λ2·U EV ·λ3·U grid

[0183] Substitute the utility function of each participant into:

[0184]

[0185] (1) Define the decision variables of the game

[0186] In this problem, the decision variables include:

[0187] Power P traded by the grid grid (t)

[0188] Photovoltaic power generation power P PV (t),

[0189] The charging power of the energy storage system is P ch (t),

[0190] The discharge power is P dis (t)

[0191] The owner's charging power P EV (t)

[0192] (2) Constraints

[0193] To ensure that the behaviors of the grid, solar storage, and vehicle owners are within reasonable physical and economic limits, the optimization problem needs to satisfy a series of constraints:

[0194] Power balance constraint: supply and demand balance between the grid, solar storage and car owners, such as: P PV (t)+P grid (t)+P dis (t)-P ch (t)-P EV (t) = 0

[0195] Power purchase constraints of the power grid: P grid (t)≤P max

[0196] Photovoltaic power generation constraint P PV (t)≤P PV,max

[0197] And energy storage charge and discharge constraints: P charge ≤P max,charge , P discharge ≤P max,discharge

[0198]

[0199] Among them, E(t) is the energy storage system power at that moment, E min 、E max is the lower and upper limits of the energy storage system capacity, P ch,max 、P dis,max is the maximum value of charging and discharging power of the energy storage system, η ch ,η dis Charging and discharging efficiency of energy storage systems.

[0200] Charging restrictions for car owners:

[0201]

[0202] Among them, P EV,max is the maximum value of the electric vehicle charging power, t=t i,start , t i,end is the starting charging time and the ending charging time of the i-th electric vehicle, E i,req is the power demand of the i-th electric vehicle, and Δt is the time interval.

[0203] Nash negotiation process

[0204] Each stakeholder adjusts its strategy through multiple iterations of negotiation, while satisfying its own constraints. In each iteration, the utility function of each stakeholder is calculated, and according to the Nash negotiation principle, a strategy combination is sought that maximizes the product of their utility functions. Strategies are continuously adjusted iteratively until a Nash equilibrium is reached, where each stakeholder's utility function is optimal under the current strategy combination, and unilaterally changing the strategy of any one stakeholder will not improve its own utility.

[0205] Step 404: Solve using heuristic algorithm

[0206] Given that energy management in integrated solar-storage-charging stations involves multi-party game-playing and complex dynamic constraints, traditional optimization methods often face challenges such as excessive computational effort and slow response. To improve computational efficiency, this paper employs an improved heuristic algorithm, the modified gradient descent method, to address the computational complexity issues associated with large-scale system optimization. This algorithm achieves real-time optimization while ensuring accuracy and efficiently coordinates the strategies of all parties through a Nash negotiation mechanism.

[0207] The improved gradient descent method is used to solve the above problem. The specific steps are:

[0208] S201: Initialize decision variable P grid (0), P PV (0), P ch (0), P dis (0), P EV (0).

[0209] S202: Calculate the utility function and its gradient under the current decision variable.

[0210] S203: Update the decision variables according to the gradient and learning rate.

[0211] S204: Check the convergence conditions. If not, return to step S202 for the next iteration. If converged, output the optimal solution.

[0212] The decision variables are adjusted according to the direction of the gradient so that the objective function value gradually converges towards the optimal solution. The update formula for gradient descent is:

[0213]

[0214] in:

[0215] X (k) is the parameter value (decision variable) at the kth iteration.

[0216] η is the learning rate, which controls the step size of each update.

[0217] is the gradient of the objective function f(X) with respect to X.

[0218] The gradient of the grid utility function is calculated as:

[0219]

[0220] The gradient of the light storage utility function is calculated as:

[0221]

[0222] The gradient of the owner's utility function is calculated as:

[0223]

[0224] According to the basic formula of the gradient descent method, the update rule of the decision variable in each iteration is:

[0225]

[0226] In the gradient descent method, choosing an appropriate step size (learning rate) is crucial. If the step size is too large, it may lead to excessive updates and cause convergence failure; if the step size is too small, the convergence speed will be very slow. Although the standard gradient descent method is simple and intuitive, its convergence speed and stability are often affected. This paper combines the ideas of momentum method and RMSprop, and adjusts the learning rate of each parameter by calculating the moving average of the gradient and the moving average of the square of the gradient.

[0227]

[0228] in:

[0229] m t is the moving average of the gradient.

[0230] v t is the moving average of the squared gradient.

[0231] β1 and β2 are the decay rates of momentum and gradient squared, respectively, and are set to numbers close to 1, optionally 0.999.

[0232] and It is a correction for deviation.

[0233] The convergence judgment condition is: the maximum number of iterations reaches a predetermined upper limit or the change of the objective function is less than a predetermined threshold (that is, when |f(X (k+1) )-f(X (k) )|<ε).

[0234] Furthermore, the step S105 includes:

[0235] Based on the optimal energy management scheduling strategy derived from the Nash negotiation, the integrated energy station operator uses a control system to schedule the photovoltaic power generation system, energy storage system, and charging facilities in real time. For example, the operating status of the energy storage converter is controlled according to the energy storage system's charging and discharging strategy; the power flow of the power converter connected to the grid is adjusted according to the grid energy interaction strategy; and the charging power of each charging pile is rationally allocated based on the charging facility scheduling strategy. A comprehensive monitoring system is established to monitor the integrated energy station's operating parameters in real time, including photovoltaic power generation, energy storage system state of charge, charging power, and grid interaction power. Simultaneously, the actual utility of each stakeholder is calculated and compared with the negotiated optimal utility. If actual operating parameters deviate from the predetermined strategy or utility values ​​decrease, the cause is promptly analyzed and the strategy adjusted. For example, if a sudden drop in photovoltaic power leads to excessive discharge pressure on the energy storage system, the charging power allocation of the charging facilities is appropriately adjusted to prioritize critical charging needs. Based on the grid electricity price and energy storage status, the system determines whether to increase power purchased from the grid to maintain stable operation of the integrated energy station and achieve the optimal energy management scheduling goal as closely as possible.

[0236] In summary, the present invention provides an energy management and scheduling method for a photovoltaic, storage and charging integrated energy station based on Nash negotiation, which collects and analyzes data and establishes a corresponding prediction model; comprehensively considers the demands of multiple stakeholders such as electric vehicle users, charging station operators, and power grid operators, and constructs a multi-objective optimization model; after determining the Nash negotiation subjects and their respective strategy spaces, a utility function is constructed based on Nash negotiation theory, incorporating factors such as the uncertainty of photovoltaic power generation, the charging and discharging characteristics of the energy storage system, the dynamic charging requirements of electric vehicles, and the volatility of the power grid load into the same scheduling framework, optimizing the decision-making of all parties through a game theory model, and each subject determines the optimal energy management scheduling strategy through negotiation until a Nash equilibrium is reached; for large-scale calculations, the present invention adopts an improved heuristic algorithm for solution, which ensures that the calculation accuracy is maintained while significantly reducing the calculation complexity. This improves the flexibility, adaptability and economy of the energy scheduling scheme.

[0237] It will be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention may be implemented in various computer languages, for example, the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0238] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0239] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0240] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0241] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0242] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for energy management and scheduling of a photovoltaic energy storage and charging integrated energy station based on Nash negotiation, characterized by: The following steps are included: Step 1: Construct a multi-objective optimization model that comprehensively considers the interests of electric vehicle users, charging station operators, and power grid operators; Step 2: Determine the participants in the Nash negotiation as charging station operators, power grid operators, and electric vehicle users, and determine the strategy space of each participant; Step 3: Based on Nash negotiation theory, consider each stakeholder as a negotiation participant, construct the utility function of each stakeholder, and determine the negotiation benchmark of each stakeholder; Step 4: solving the Nash equilibrium solution of the multi-objective optimization model and determining the optimal energy management scheduling strategy through iterative negotiation; Step 5: According to the optimal energy management scheduling strategy, the photovoltaic storage and charging integrated energy station is controlled and scheduled in real time.

2. The method according to claim 1, wherein: In step 1, the multi-objective optimization model includes three objectives: maximizing the overall operating income of the integrated photovoltaic storage and charging power station, reducing the charging costs of electric vehicle owners, and reducing the degree of grid load fluctuation; Among them, to maximize the overall operating income of the integrated photovoltaic storage and charging power station, the objective function is: R station =R EV +R grid -C ess -C pv -C purchase +λE pv,util Among them, R station is the charging station target, R PV is the revenue from providing charging services to electric vehicles, R grid is the revenue from selling electricity to the grid, C ess is the total operating cost of the energy storage system, C pv is the operation and maintenance cost of the photovoltaic power generation system, C purchase is the cost of purchasing electricity from the grid, E pv,util is the goal of maximizing the photovoltaic consumption of renewable energy, and λ is the conversion coefficient; Among them, to reduce the charging cost of electric vehicle owners, the objective function is: Among them, T is the scheduling period, C EV is the goal of the electric vehicle owner; ω1, ω2, ω3 are weight coefficients, which respectively represent the importance of charging cost, charging power matching and charging waiting time in satisfaction; P ev,t is the charging power of the electric vehicle, Price ev,t is the time-sharing charging price; P ev * Represents the predicted charging demand power value at the time, T wait is the average charging waiting time; Among them, to reduce the degree of grid load fluctuation, the objective function is: Among them, L fluct is the load fluctuation index of the power grid, P grid,t is the power interaction with the grid at time t.

3. The method according to claim 1, wherein: In step 2, the strategy space of Nash negotiation is determined, including: Determine the strategy space for charging station operators, including the charge and discharge control strategy of the energy storage system and the energy interaction strategy of the power grid; Determine the grid operator's strategy including electricity price incentives or constraints for integrated energy stations; Determining the strategy of electric vehicle users includes choosing the charging time and charging power.

4. The method according to claim 3, wherein: In step 3, the utility function of each stakeholder is constructed, including: The utility function of the charging station operator is constructed as: Among them, U station Utility for charging station operators, It is an alternative location for charging stations; The utility function of electric vehicle users is constructed as follows: Among them, U EV Utility for electric vehicle users, An alternative point for car owners; The utility function of the grid operator is constructed as, Among them, U grid Utility for grid operators, It is an alternative point for the power grid.

5. The method according to claim 1, wherein: In step 4, the optimal energy management scheduling strategy is determined through Nash negotiation, including: Construct the objective function of Nash negotiation solution; Construct the decision variables and constraints of Nash negotiation; On the basis of satisfying their own constraints, each stakeholder adjusts their own strategy through multiple iterative negotiations until a Nash equilibrium is reached, and the strategy combination at this time is used as the optimal energy management scheduling strategy.

6. The method according to claim 5, wherein: Construct the objective function of the Nash negotiation solution, including, Among them, U total is the product of the three-party utility; λ1, λ2, and λ3 are the weights of charging station operators, electric vehicle users, and power grid operators respectively; R station For charging station targets, It is an alternative location for charging stations; is the car owner's alternative point, C EV Target for tram owners; is the alternative point of the power grid, L fluct It is the load fluctuation index of the power grid.

7. The method according to claim 5, wherein: Adjust their respective strategies through multiple iterative negotiations until a Nash equilibrium is reached. include, The improved gradient descent method is used to solve the problem, which includes the following steps: S201: Initialize decision variables; S202: Calculate the utility function and its gradient under the current decision variable; S203: Update decision variables according to gradient and learning rate; S204: Check the convergence conditions. If not, return to step S202 for the next iteration. If converged, output the optimal solution.

8. The method according to claim 5, wherein: Adjust their respective strategies through multiple iterative negotiations, including, Obtain historical and real-time data from the integrated solar-storage-charging energy station, including storage state-of-charge data and grid electricity price information; Obtaining photovoltaic power generation prediction data and charging demand prediction data based on the historical data and real-time data; Iterative negotiation is performed based on the photovoltaic power generation prediction data and the charging demand prediction data.

9. The method according to claim 8, wherein: Obtain photovoltaic power generation forecast data, including, According to the prediction model P based on neural network PV,t * =f PV (I t ,T t ,H t ) to obtain photovoltaic power generation prediction data, where P PV,t * represents the predicted value of photovoltaic power generation at time t, I t ,T t ,H t They represent the meteorological parameters such as light intensity, temperature, and humidity at time t, respectively. PV is the photovoltaic power prediction function; Obtain charging demand forecast data, including, According to the charging demand prediction model P ev * =g C (t,n EV,t ,E req,t ) obtain charging demand forecast data, where P ev * represents the charging demand power forecast value at the moment, t is the time variable, n ev,t represents the number of electric vehicles at time t, E req,t represents the average charging energy demand of a single electric vehicle at time t, g C is the charging demand prediction function.

10. The method according to claim 1, wherein: The method also includes real-time monitoring of the operating parameters of the photovoltaic, storage and charging integrated energy station, including photovoltaic power generation power, energy storage system charge state, charging power and interaction power with the power grid; when the actual operating parameters deviate from the predetermined strategy or the actual utility value of each stakeholder decreases, the charging power distribution of the charging facility and the energy interaction power with the power grid are adjusted according to the power grid electricity price and energy storage status.

Images