Multi-agent game energy scheduling method and system for high-speed service area micro-grid

Abstract

The invention provides a multi-agent game energy scheduling method and system for a high-speed service area micro-grid, and relates to the technical field of energy scheduling, and the method comprises the steps: building a real-time operation state in the operation process of the high-speed service area micro-grid; constructing a multi-subject energy interaction model; generating a dynamic contract electricity price strategy by a high-speed service area operation main body under the limitation of a system operation constraint set; and performing cooperative scheduling control based on the power purchase proportion and the system operation constraint set. According to the method and the device, the technical problem that static scheduling is difficult to carry out real-time collaborative optimization and stable operation of the micro-grid in the high-speed service area is further influenced due to the fact that the high-speed service area is large in electrical load fluctuation, insufficient in renewable energy consumption capability and difficult in multi-main-body collaborative optimization in the prior art is solved, and by simulating and optimizing interaction between different main bodies, the stable operation of the micro-grid in the high-speed service area is improved. Cooperative scheduling control is realized, and the stability of the high-speed service area micro-grid is improved.

Description

A multi-agent game-theoretic energy dispatching method and system for microgrids in high-speed service areas Technical Field

[0001] This application relates to the field of energy dispatching technology, specifically to a multi-agent game energy dispatching method and system for a high-speed service area microgrid. Background Technology

[0002] Highway service areas, as unique transportation nodes, exhibit distinctive electricity consumption characteristics. Their loads show significant tidal fluctuations, with substantial differences in electricity demand between holidays and weekdays, and the charging load is highly random. Simultaneously, photovoltaic (PV) power generation in service areas is greatly affected by weather and geographical location, resulting in unstable output. This poses a greater challenge to the energy management of microgrids in highway service areas. Existing methods are mostly focused on PV-storage-charging microgrids in industrial parks or communities. Their dispatch strategies are often based on centralized optimization by a single entity or static time-of-use pricing mechanisms, focusing on single-entity or localized optimization. This approach struggles to adapt to the complex dynamic characteristics unique to highway service areas, such as drastic tidal load fluctuations, high randomness in PV power generation, and short user dwell times. Furthermore, the lack of effective market or game-theoretic mechanisms to coordinate the interests of multiple stakeholders, including service area operators and charging station operators, puts pressure on the safe and stable operation of the distribution network.

[0003] In summary, existing technologies suffer from technical problems such as large fluctuations in electricity load in highway service areas, insufficient renewable energy absorption capacity, and difficulties in multi-entity collaborative optimization, which make it difficult to achieve real-time collaborative optimization of static dispatch, thus affecting the stable operation of microgrids in highway service areas. Summary of the Invention

[0004] The purpose of this application is to provide a multi-agent game-theoretic energy dispatching method and system for microgrids in high-speed service areas, in order to solve the technical problems in the prior art where large fluctuations in electricity load in high-speed service areas, insufficient renewable energy absorption capacity, and difficulties in multi-agent collaborative optimization make it difficult to achieve real-time collaborative optimization of static dispatching, thereby affecting the stable operation of microgrids in high-speed service areas.

[0005] To achieve the above objectives, this application provides a multi-agent game-theoretic energy scheduling method and system for microgrids in high-speed service areas.

[0006] Firstly, this application provides a multi-agent game-theoretic energy dispatching method for a highway service area microgrid. This method is implemented through a multi-agent game-theoretic energy dispatching system for a highway service area microgrid. The method includes: during the operation of the highway service area microgrid, collecting real-time output status of the photovoltaic power generation system, state of charge of the energy storage system, base load status of the service area, and electric vehicle charging load demand status to establish a real-time operating status; and constructing a multi-agent energy interaction model based on the energy interaction relationship between the highway service area operator and the charging station operator. This multi-agent energy interaction model is used to characterize the energy of each operator under different energy source conditions. The system acquires permissions, adjustable power ranges, and grid-connected power constraints. Based on the multi-entity energy interaction model and the real-time operating status, it generates a system operation constraint set to constrain the energy flow allocation between the microgrid's source, load, and storage network. Under the constraints of the system operation constraint set, the highway service area operator generates a dynamic contract electricity price strategy. Based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, and under the constraints of the system operation constraint set, the charging station operator determines the electricity purchase ratio of the charging load from the highway service area microgrid versus the public grid. According to the electricity purchase ratio and the system operation constraint set, it performs coordinated scheduling and control of photovoltaic power generation allocation, energy storage system charging and discharging power, charging load power supply path, and interaction power between the highway service area microgrid and the public grid.

[0007] Optionally, the multi-agent energy interaction model is constructed based on Stackelberg game theory, in which the high-speed service area operator sets the dynamic contract electricity price as the leader, and the charging station operator adjusts the electricity purchase ratio according to the dynamic contract electricity price and the real-time electricity price of the public grid. The Stackelberg game theory is solved by backward induction to find the equilibrium.

[0008] Optionally, the system operation constraint set includes the microgrid energy balance equation, energy storage system constraints, and electricity purchase ratio binding constraints, wherein: the microgrid energy balance equation is expressed as: ;in, for Photovoltaic power generation at any given time for The discharge power of the energy storage system at any given time. for The charging power of the energy storage system at any time. for Power purchased by the power grid at any given time for Flexible load power at any time for The power transmitted from the operator of the highway service area to the main body of the charging station.

[0009] Optionally, the constraints of the energy storage system include a dynamic state of charge model, charge / discharge power constraints, and capacity constraints, wherein: the dynamic state of charge model is expressed as: ;in, for The state of charge of the energy storage system at all times. for The state of charge of the energy storage system at all times. For energy storage systems at time steps The total energy of discharge, The rated capacity of the energy storage system, For discharge efficiency, For energy storage systems at time steps Total input electrical energy, For charging efficiency; charging and discharging power constraints satisfy , and These are the minimum and maximum discharge power of the energy storage system, respectively; the capacity constraint satisfies... , and These are the minimum and maximum capacities of the energy storage system, respectively.

[0010] Optionally, the dynamic contract electricity pricing strategy is constructed based on time-of-use pricing mechanism and supply and demand status awareness.

[0011] Optionally, the multi-party game energy dispatch is executed periodically within a rolling time window, so that the power purchase ratio and the interactive power are dynamically updated as the photovoltaic output and charging load change.

[0012] Optionally, in determining the electricity purchase ratio, the electricity purchase ratio can be coupled with the local photovoltaic absorption capacity of the highway service area to limit the purchased power or grid-connected power in the case of photovoltaic surplus.

[0013] Secondly, this application also provides a multi-agent game-theoretic energy dispatching system for a high-speed service area microgrid, used to execute the multi-agent game-theoretic energy dispatching method for a high-speed service area microgrid as described in the first aspect. The multi-agent game-theoretic energy dispatching system for a high-speed service area microgrid includes: a status acquisition module, used to acquire the real-time output status of the photovoltaic power generation system, the state of charge of the energy storage system, the base load status of the service area, and the charging load demand status of electric vehicles during the operation of the high-speed service area microgrid, and to establish a real-time operating status; and a model building module, used to construct a multi-agent energy interaction model based on the energy interaction relationship between the high-speed service area operator and the charging station operator. The multi-agent energy interaction model is used to characterize the energy acquisition rights, adjustable power range, and grid-connected power constraints of each operator under different energy source conditions. The system comprises the following modules: a system operation constraint generation module, used to generate a system operation constraint set for constraining the energy flow allocation between the microgrid's source, load, and storage network based on the multi-entity energy interaction model and the real-time operating status; a strategy generation module, used by the highway service area operator to generate a dynamic contract electricity price strategy under the constraints of the system operation constraint set; a ratio determination module, used by the charging station operator to determine the electricity purchase ratio of the charging load from the highway service area microgrid versus the public grid, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, under the constraints of the system operation constraint set; and a collaborative scheduling control module, used to perform collaborative scheduling control on the photovoltaic power generation allocation, energy storage system charging and discharging power, charging load power supply path, and interaction power between the highway service area microgrid and the public grid, based on the electricity purchase ratio and the system operation constraint set.

[0014] One or more technical solutions provided in this application have at least the following technical effects or advantages: By collecting real-time output status of the photovoltaic power generation system, state of charge of the energy storage system, base load status of the service area, and charging load demand status of electric vehicles during the operation of the microgrid in the highway service area, a real-time operating status is established; based on the energy interaction relationship between the highway service area operator and the charging station operator, a multi-entity energy interaction model is constructed, which characterizes the energy acquisition rights, adjustable power range, and grid-connected power constraints of each entity under different energy source conditions; based on the multi-entity energy interaction model and the real-time operating status... The system generates a set of system operation constraints to constrain the energy flow allocation between the microgrid's source, load, and storage network. Under these constraints, the highway service area operator generates a dynamic contract electricity price strategy. The charging station operator, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, determines the electricity purchase ratio between the charging load and the public grid, constrained by the system operation constraints. Based on this purchase ratio and the system operation constraints, the system coordinates and controls the photovoltaic power generation allocation, the energy storage system's charging and discharging power, the charging load's power supply path, and the interaction power between the highway service area microgrid and the public grid. In other words, by collecting the real-time operating status of the photovoltaic power generation system, a multi-entity energy interaction model is used to optimize the coordinated scheduling of various energy sources in the microgrid. This establishes an interactive relationship between the service area operator and the charging station operator. Furthermore, by optimizing the energy acquisition permissions, power ranges, and grid connection power constraints among the entities, collaborative optimization among them is achieved. Based on the operation constraints and real-time electricity price, the electricity purchase ratio of the charging load is flexibly adjusted, improving energy utilization efficiency while ensuring the stability of the highway service area microgrid.

[0015] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0017] Figure 1 is a flowchart illustrating a multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid according to this application.

[0018] Figure 2 is a schematic diagram of the structure of a multi-agent game energy dispatching system for a high-speed service area microgrid according to this application.

[0019] Figure labeling: 11. Status acquisition module, 12. Model building module, 13. Running constraint generation module, 14. Strategy generation module, 15. Proportion determination module, 16. Cooperative scheduling and control module. Detailed Implementation

[0020] This application provides a multi-agent game-theoretic energy dispatching method and system for microgrids in highway service areas. It addresses the technical problems in existing technologies where large fluctuations in electricity load, insufficient renewable energy absorption capacity, and difficulties in multi-agent collaborative optimization in highway service areas hinder real-time collaborative optimization of static dispatching, thus affecting the stable operation of microgrids in these areas. By collecting real-time operating status data of the photovoltaic power generation system, a multi-agent energy interaction model is used to optimize the coordinated dispatching of various energy sources within the microgrid. An interactive relationship is established between the service area operator and the charging station operator. Furthermore, by optimizing the energy acquisition permissions, power ranges, and grid-connected power constraints among the operators, collaborative optimization among them is achieved. Based on the operating constraint set and real-time electricity price, the electricity purchase ratio of the charging load is flexibly adjusted, improving energy utilization efficiency while ensuring the stability of the microgrid in the highway service area.

[0021] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0022] Example 1, please refer to Figure 1. This application provides a multi-agent game energy dispatch method for a high-speed service area microgrid. The multi-agent game energy dispatch method for a high-speed service area microgrid is applied to a multi-agent game energy dispatch system for a high-speed service area microgrid. The multi-agent game energy dispatch method for a high-speed service area microgrid specifically includes the following steps: S100: During the operation of the high-speed service area microgrid, the real-time output status of the photovoltaic power generation system, the state of charge of the energy storage system, the basic load status of the service area, and the charging load demand status of electric vehicles are collected to establish a real-time operating status.

[0023] Specifically, the integrated photovoltaic-storage-charging-DC-flexible microgrid in the highway service area adopts a common AC bus topology and mainly consists of a photovoltaic system, an energy storage system, a charging pile system, flexible loads, and grid connection components. The energy management system is the core control unit, which collects real-time data from various systems through various devices, and coordinates and schedules various devices in combination with control strategies and optimization algorithms to ensure the safe, stable, and economical operation of the system; it also has remote monitoring and data analysis functions to provide decision support for energy management.

[0024] Intelligent sensing and metering devices are installed at key nodes of the microgrid. Photovoltaic inverters continuously measure and report their output power; the energy storage battery management system accurately calculates and communicates their state of charge (SOC); smart meters installed at the main incoming line of the service area's power distribution room monitor the total power of the base load in real time; and each DC charging pile reports its current actual power being used to charge vehicles or its standby status. Real-time output status, the SOC of the energy storage system, the base load status of the service area, and the charging load demand status of electric vehicles are collected in real time via a high-speed communication network to the energy management system located in the service area's data center or in the cloud. These heterogeneous data are timestamped, verified, and integrated at extremely high frequencies to form a real-time operating status. This transforms the dispersed, simulated states in the physical system into centralized, accurate digital information, solving the problems of data lag, insufficient accuracy, and missing dimensions inherent in traditional manual meter reading or periodic reporting methods. The high-frequency data acquisition capability allows the system to promptly detect sudden drops in photovoltaic output due to cloud cover or sudden surges in charging load, gaining valuable time for rapid adjustments to scheduling strategies.

[0025] S200: Based on the energy interaction relationship between the operator of the highway service area and the operator of the charging station, a multi-entity energy interaction model is constructed. The multi-entity energy interaction model is used to characterize the energy acquisition rights, adjustable power range and grid-connected power constraints of each entity under different energy source conditions.

[0026] Furthermore, S200 of this application includes: the multi-agent energy interaction model is constructed based on Stackelberg game, wherein the high-speed service area operator sets the dynamic contract electricity price as the leader, and the charging station operator adjusts the electricity purchase ratio according to the dynamic contract electricity price and the real-time electricity price of the public grid. The Stackelberg game is solved for equilibrium by backward induction.

[0027] Specifically, the service area operator is clearly defined as the leader, with its decision variable being the dynamic contract electricity price curve over a future period; the charging station operator is defined as the follower, with its decision variable being the corresponding electricity purchase ratio curve. An objective function is established for the service area operator to maximize total operating revenue, derived from selling electricity to charging stations and to the grid, with costs including electricity purchase from the grid and energy storage losses. An objective function is established for the charging station operator to minimize total electricity purchase costs. Real-time operating status is used as known parameters, and the rigid constraints of the physical system, such as energy balance, energy storage charging and discharging power and capacity limitations, and grid interaction power limitations, are transformed into mathematical inequality constraints, collectively constituting the feasible region that both parties must adhere to in their decisions.

[0028] The algorithm employs backward induction to simulate the decision-making process of a charging station: upon receiving a tentative electricity price signal, it immediately calculates an optimal electricity purchase ratio based on its own cost minimization model. This feedback is then substituted into the service area operator's model to evaluate its own profitability under this pricing strategy. Through iterative optimization, such as gradient descent and intelligent optimization algorithms, a pricing strategy is ultimately found that maximizes the service area operator's profitability while maintaining the charging station's response at this price. At this point, the Stackelberg game equilibrium solution is obtained, representing the optimal strategy combination for both parties.

[0029] The energy interaction between highway service area operators and charging station operators is essentially an electricity trading relationship. Service areas act as electricity producers and wholesalers, while charging stations are retailers and major customers. This interaction determines how electricity is priced, transmitted, and the rights and obligations of both parties. A multi-entity energy interaction model characterizes the energy acquisition rights, adjustable power ranges, and grid-connected power constraints of each entity under different energy source conditions. Energy acquisition rights indicate who has the right to use photovoltaic power, energy storage power, and grid power; adjustable power ranges include the charging station's total load being adjustable between 200kW and 800kW, and the service area's power purchase / sale from the grid not exceeding 500kW; grid-connected power constraints represent the safety limits and protocol requirements that must be met for power exchange between the microgrid and the public grid connection point.

[0030] Stackelberg game is a classic master-follower dynamic game model that simulates a sequential decision-making process: the leader acts first, formulating strategies such as pricing; followers observe the leader's actions and then formulate their own optimal strategies, such as purchasing quantity. Both sides rationally pursue their own profit maximization, eventually reaching a stable state where neither side benefits from unilaterally changing its strategy—this is game equilibrium.

[0031] The fundamental theorem of static Stackelberg game theory states that a Stackelberg equilibrium exists if the objective functions of both the principal and the slave are continuous, and the slave's optimal response mapping is non-empty, compact, and upper-semi-continuous. The Stackelberg equilibrium solution is unique when the slave's utility function is strictly concave and the principal's utility function remains strictly concave after substitution. Let the service area operator be the principal, with the contract electricity price as the decision variable; and the charging station be the slave, with its purchased electricity as the decision variable. Define the objective functions of both parties as continuous and strictly concave functions, respectively, and the feasible region as a closed convex set. This two-level optimization problem is solved using backward induction: first, the contract electricity price is fixed, and the slave's optimal response electricity purchase is calculated; then, this is fed back into the principal's optimization problem to search for the optimal contract electricity price. Since the inner problem has a closed-form solution and the outer problem is a univariate nonlinear optimization, it can converge efficiently through gradient descent or enumeration. Simulation results show that the algorithm converges on average within 5 iterations, verifying the stability and computational feasibility of the strategy.

[0032] The complex problem of coordinating the interests of multiple stakeholders is transformed into a standardized mathematical optimization problem that can be solved automatically, and an incentive-compatible market mechanism is designed. Through Stackelberg game theory and dynamic pricing, the designed rules ensure that charging stations pursue the lowest-cost behavior, i.e., buying more cheap photovoltaic electricity, which is perfectly aligned with the service area operator's goal of maximizing revenue, i.e., selling more photovoltaic electricity and reducing low-price grid connection, as well as the system's goal of increasing photovoltaic power consumption.

[0033] S300: Based on the multi-subject energy interaction model and the real-time operating status, generate a set of system operation constraints to constrain the energy flow distribution of the microgrid source-load-storage network.

[0034] Furthermore, S300 of this application includes: a system operation constraint set including a microgrid energy balance equation, energy storage system constraints, and electricity purchase ratio binding constraints, wherein: the microgrid energy balance equation is expressed as: ;in, for Photovoltaic power generation at any given time for The discharge power of the energy storage system at any given time. for The charging power of the energy storage system at any time. for Power purchased by the power grid at any given time for Flexible load power at any time for The power transmitted from the operator of the highway service area to the main body of the charging station.

[0035] Specifically, based on a multi-agent energy interaction model and real-time operating status, a system operation constraint set is generated to constrain the energy flow allocation between the microgrid's source-load-storage network. This system operation constraint set is a set of mandatory, inviolable physical and operational rules expressed mathematically, specifying the boundary conditions for energy flow and state changes. It ensures that any optimization or game-theoretic decision-making scheme is physically feasible, within the safety limits of the equipment, and complies with grid connection requirements.

[0036] The energy balance equation of a microgrid is the most fundamental real-time power conservation law in power system operation. It requires that at any given moment, the total power generated by all power sources within the microgrid must be equal to the total power consumed by all loads (including storage devices). It is the mathematical cornerstone for maintaining the stability of system voltage and frequency. Any scheduling scheme that violates this equation means that it is physically impossible to implement.

[0037] Energy storage system constraints are a set of mathematical rules describing the safe, efficient, and long-life operating boundaries of energy storage devices. These rules ensure that energy storage scheduling does not exceed its hardware capabilities and follows scientific battery management strategies. Energy storage system constraints include dynamic state of charge (SOC) models, charge / discharge power constraints, and capacity constraints. The dynamic state of charge model is a recursive calculation formula describing how battery charge evolves over time. Based on the current charge level, the charge / discharge energy during the current period, and battery efficiency, it accurately calculates the charge level at the start of the next period, characterizing the memory effect of energy storage and the inevitable losses during energy conversion. Charge / discharge power constraints limit the maximum and minimum output (discharge) or input (charge) power that the energy storage system's power converter can deliver at a single moment, preventing equipment damage due to overload. Capacity constraints stipulate that the state of charge (SOC) of the energy storage battery must be strictly maintained within a preset safety window (e.g., 20% to 90%). Preventing permanent damage to the battery due to over-discharge or the risk of thermal runaway due to overcharging is a core constraint for ensuring battery lifespan.

[0038] The electricity purchase ratio binding constraint is a strategic, mandatory rule designed to maximize the absorption of local photovoltaic (PV) power. It directly links the proportion of electricity purchased by a charging station from the service area to the physical state of the microgrid, i.e., the surplus of local PV power generation. The aim is to guide energy flow through rules, prioritizing the local use of clean electricity. To achieve the strategic goal of absorbing every kilowatt-hour of local PV power, the operator establishes a business rule: when PV power generation exceeds the service area's basic electricity demand, the charging load should be prioritized to use this surplus green electricity. This strategy is translated into precise mathematical language, such as the electricity purchase ratio must be greater than or equal to the calculated value of (current PV output - base load) / maximum possible demand of the charging station, forming an electricity purchase ratio binding constraint, which is then added to the system operation constraint set.

[0039] The energy balance equation is taken as the fundamental equality constraint that all scheduling schemes must satisfy. Any scheduling plan generated by game theory algorithms must strictly hold true after being substituted into this equation. The energy balance equation of the integrated photovoltaic-storage-charging-DC-flexible microgrid in the high-speed service area is the basic constraint for achieving stable system operation, reflecting the supply and demand relationship of energy in each component. At any time t, the system's energy balance equation is: ;in, for Photovoltaic power generation at any given time for The discharge power of the energy storage system at any given time. for The charging power of the energy storage system at any time. for Power purchased by the power grid at any given time for Flexible load power at any time for The power transmitted from the highway service area operator to the charging station operator at time t. At time t, the total power provided by photovoltaic power generation, energy storage discharge, and grid power purchase should be equal to the total power consumed by electric vehicle charging, flexible load consumption, and energy storage charging to ensure the energy supply and demand balance of the system. In other words, for any candidate scheduling scheme generated by upper-level game theory or optimization algorithms, the values ​​of all power variables, including photovoltaic, energy storage, grid interaction, and load, must be equal on both sides after being substituted into this equation; otherwise, the scheme will be directly judged as invalid.

[0040] As an independent subsystem, the charging pile's charging power at time t is the sum of the power transmitted from the service area to the charging station and the power purchased directly from the grid by the charging station at time t. This indicates that the charging power of the charging station is jointly provided by the power transmitted from the service area and the power purchased directly from the grid, reflecting the energy source composition of the charging station.

[0041] By constructing a system operation constraint set, the solution efficiency and reliability are improved, making it possible to quickly find the global optimal or satisfactory solution under complex multivariable and multi-constraint conditions. This provides computational support for the online real-time operation of microgrids in high-speed service areas.

[0042] Furthermore, this application also includes the following steps: the energy storage system constraints include a dynamic state of charge model, charge / discharge power constraints, and capacity constraints, wherein: the dynamic state of charge model is expressed as: ;in, for The state of charge of the energy storage system at all times. for The state of charge of the energy storage system at all times. For energy storage systems at time steps The total energy of discharge, The rated capacity of the energy storage system, For discharge efficiency, For energy storage systems at time steps Total input electrical energy, For charging efficiency; charging and discharging power constraints satisfy , and These are the minimum and maximum discharge power of the energy storage system, respectively; the capacity constraint satisfies... , and These are the minimum and maximum capacities of the energy storage system, respectively.

[0043] Specifically, the State of Charge (SOC) dynamic model is a state update formula that describes how the "charge" of an energy storage battery precisely evolves over time. It is not a separate equation, but rather two calculation formulas that switch depending on whether the energy storage is in charging or discharging mode. Its core principle is energy conservation: the remaining charge at the current moment equals the charge at the previous moment, plus the net energy stored (considering efficiency losses) or minus the net energy released (considering efficiency losses) during this period. SOC is the ratio of the energy currently stored in the battery to its rated total capacity, expressed as a percentage, and is the most critical state indicator for measuring how much charge the energy storage still has. Rated capacity is the total electrical energy that the energy storage system can store or release under specific conditions, such as a specific discharge rate and temperature. It is usually measured in kilowatt-hours (kWh) and is the denominator in calculating SOC, as well as a core parameter for measuring the scale of energy storage. Charge / discharge efficiency is the proportion of energy lost during the process of storing or retrieving energy from the battery due to chemical conversion, heat loss, converter conversion, etc. For example, a charging efficiency of 95% means that out of 100 kWh of electricity input from the grid, only 95 kWh can be effectively stored in the battery.

[0044] The dynamic model of the state of charge is represented as follows: ;in, for The state of charge of the energy storage system at all times. for The state of charge of the energy storage system at all times. For energy storage systems at time steps The total energy of discharge, The rated capacity of the energy storage system, For discharge efficiency, For energy storage systems at time steps Total input electrical energy, For charging efficiency.

[0045] The charge / discharge power constraint limits the maximum rate at which the power converter of an energy storage system can transfer energy per unit time. It restricts the instantaneous capability of the energy storage system, ensuring its operation within the safe limits of the hardware design. The charge / discharge power constraint satisfies... , and These refer to the minimum and maximum discharge power of the energy storage system, respectively. The maximum / minimum charge / discharge power is typically determined by the rated power of the energy storage converter. For example, a 500kW / 1MWh system... Typically 500kW, with positive for discharging and negative for charging. It can be 0 (standby) or a negative value (minimum allowed charging power).

[0046] Capacity constraints stipulate that the state of charge (SOC) of energy storage batteries must be strictly maintained within a preset safe operating window. This is a protective limitation on the accumulated SOC of the energy storage system and directly relates to battery life and safety. Capacity constraints must be met. , and These are the minimum and maximum capacities of the energy storage system, respectively. The maximum / minimum capacity is typically not equal to 0% and 100%. To prevent overcharging and over-discharging of the battery and protect its chemical structure, a conservative buffer zone is typically set up in actual operation, such as... =20%, =90%, sacrificing some usable capacity, but gaining longer cycle life and higher security in return.

[0047] The dynamic state of charge (SOC) model, charge / discharge power constraints, and capacity constraints work together to achieve a closed-loop management process in the energy management system. During initialization, fixed parameters such as the rated capacity, efficiency, power limits, and SOC safety window of the energy storage are loaded. During operation, the current SOC is continuously read through the battery management system. The current SOC is linked to the decision variable, the current charging power or current discharging power. The optimization algorithm must simultaneously calculate its impact on future SOC while determining the charging / discharging power. The range of values ​​for the current charging or discharging power is directly limited to ensure that the command does not exceed the converter's capacity. The SOC of the energy storage system at time t, calculated by the dynamic model, is constrained to ensure it falls within the safe range. In each scheduling period, such as the next 24 hours, at 15-minute intervals, the optimization algorithm solves for the optimal scheduling plan while satisfying all the above constraints. After the plan is generated, the cumulative impact of executing the plan on the SOC is simulated prospectively to ensure that the SOC trajectory does not reach the minimum and maximum capacity limits of the energy storage system throughout the entire scheduling cycle. The scheduling command is then issued to the energy storage converter for execution. After a period of time Δt, the system updates the SOC value precisely according to the actual measured charge and discharge energy and the dynamic model formula, which serves as the initial state for the next optimization cycle, thus starting a new round of optimization.

[0048] Charging and discharging power constraints prevent acute faults such as converter damage or battery thermal runaway caused by overcurrent; capacity constraints effectively slow down battery capacity degradation and significantly extend its service life by preventing overcharging and over-discharging, directly reducing the project's total lifecycle cost. The dynamic model couples scheduling decisions across different time periods, making optimization no longer an isolated period optimization but a globally coordinated optimization, ensuring that tonight's charging decisions take into account tomorrow morning's discharging needs, thus generating a temporally coherent optimal scheduling sequence.

[0049] S400: A dynamic contract electricity price strategy generated by the highway service area operator under the constraints of the system operation set.

[0050] Furthermore, S400 of this application includes: the dynamic contract electricity price strategy is constructed based on the time-of-use electricity price mechanism and supply and demand status perception.

[0051] Specifically, dynamic contract pricing is a pricing scheme for service area operators to sell electricity to charging station operators. It not only follows the general framework of time-of-use pricing in the public grid but also allows for sensitive adjustments based on real-time supply and demand conditions within the microgrid, measured in seconds or minutes. Time-of-use pricing is a pricing policy established by the public grid based on the overall electricity load patterns of society, with different prices at different times, typically divided into peak, flat, and valley periods, with the highest price during peak hours and the lowest during valley hours. Supply and demand perception refers to the service area operator's precise understanding of the real-time balance between its microgrid's generation capacity and electricity demand. This includes real-time photovoltaic output, the current state of charge (SOC) of energy storage, the service area's base load, and the total power demand of charging stations. This is the core basis and driving signal for dynamic contract pricing to function effectively.

[0052] First, based on the current time, the core operating mode is matched, such as 23:00-7:00 for nighttime charging; 8:00-11:00 and 15:00-19:00 for self-consumption of photovoltaic power during off-peak hours; and 1:00-15:00 and 19:00-22:00 for peak-hour charging replenishment. Based on the collected real-time operating status data, fine adjustments are made on top of the basic pricing tendency. The adjustments follow a core economic logic: electricity prices are positively correlated with the scarcity of local power resources. When photovoltaic power generation is high and there is a surplus, i.e., supply exceeds demand, and local power resources are abundant, EMS will proactively lower the contract price to make it more attractive in order to maximize consumption and avoid curtailment. When photovoltaic power generation is insufficient, energy storage SOC is low, and load is high, i.e., supply is less than demand, and local power resources are tight, it is necessary to rely on higher-cost energy storage discharge or purchase electricity from the grid during peak hours. In this case, to reflect the true cost of power supply and guide charging stations to use grid electricity rationally, EMS will appropriately raise the contract price. The price adjustment process is not arbitrary, but rather strictly constrained by the entire system's operational constraints. For example, the pricing strategy must ensure that the final energy dispatch result meets the safety constraints of the energy storage SOC, the grid interaction power limit, and passes the energy balance equation test. Therefore, the pricing decision itself is often a sub-problem embedded in a larger-scale optimization problem. By solving the entire game model, i.e., considering the optimal response of the charging stations, the optimal contract electricity price curve is derived. The generated optimal (or suboptimal) dynamic contract electricity price will serve as a clear bidding signal, sent to the charging station operators, and initiating the subsequent game response phase.

[0053] Based on the electricity consumption characteristics and energy supply features of highway service areas, microgrids can be divided into the following four core operating modes: Nighttime charging mode is during off-peak hours when grid electricity prices are low and there is no photovoltaic power generation. The microgrid mainly relies on grid power supply to meet basic loads such as lighting and security in the service area, while also charging the energy storage system. The charging power is adjusted according to the current SOC (State of Charge) of the energy storage and the duration of off-peak electricity to ensure that it is charged to a higher SOC level during off-peak hours. If there is a need for vehicle charging, grid power is used first to control costs.

[0054] In the self-consumption mode of photovoltaic (PV) power generation, the PV system begins to generate electricity as solar radiation increases, at which point the load is stable. The microgrid prioritizes using PV power to supply charging piles and flexible loads. When there is a surplus in PV output, it is used to charge energy storage, maximizing local consumption. If there is still a surplus and the self-consumption revenue is higher than the revenue from selling electricity, electricity can be sold to the grid.

[0055] Peak-hour charging replenishment mode is used during peak electricity demand and when electricity prices are high, potentially resulting in insufficient photovoltaic output. When photovoltaic output cannot meet the load, energy storage discharges in conjunction with photovoltaic power supply; if this is still insufficient, electricity is purchased from the grid, and peak shaving and valley filling are achieved through energy storage discharge, reducing costs. Charging piles prioritize the use of photovoltaic and energy storage energy, with the grid supplementing any shortfall.

[0056] A multi-stakeholder interactive model runs throughout the entire operation and is the core of collaborative optimization. Service area operators take the lead, developing dynamic contract electricity prices based on photovoltaic output, energy storage SOC, load, and electricity prices. Charging station operators, combining contract electricity prices with real-time grid electricity prices, flexibly choose to purchase photovoltaic and energy storage energy from service areas or directly from the grid, balancing the interests of all parties. The Energy Management System (EMS) monitors the status of each entity and energy flow in real time, adjusting control strategies to ensure system stability.

[0057] The generation and implementation of dynamic contract pricing strategies translate the complex physical system states, such as photovoltaic output and energy storage SOC, into market price signals that charging stations can directly understand and respond to, automatically guiding charging loads to track photovoltaic output over time. Lower electricity prices when there is abundant photovoltaic power stimulate increased electricity consumption; higher electricity prices when there is less photovoltaic power suppress some demand or shift it to the grid.

[0058] S500: The charging station entity determines the electricity purchase ratio between the charging load obtaining electricity from the high-speed service area microgrid and the public grid, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, under the constraints of the system operation constraint set.

[0059] Specifically, after receiving the dynamic contract electricity price strategy from the service area operator, the charging station compares it with the real-time electricity price of the public grid. Simultaneously, the charging station has its own charging load demand and aims to minimize its own electricity purchase costs while meeting that demand. However, the charging station's decision-making is not entirely free; it must adhere to a set of system operating constraints, especially the electricity purchase ratio constraint. When there is a surplus in photovoltaic power, the electricity purchase ratio must not fall below a certain lower limit. Comparing the dynamic contract electricity price strategy with the real-time electricity price of the public grid, if the dynamic contract electricity price strategy is lower than the real-time electricity price of the public grid, purchasing electricity from the service area is more economical, and the station tends to increase the proportion of power purchased from the service area microgrid; conversely, it tends to decrease the proportion of power purchased from the service area microgrid. The electricity purchase ratio is the proportion of the total charging power of the charging station that is purchased from the service area microgrid.

[0060] The decision-making process for charging stations is not entirely free; it must adhere to the set of system operation constraints defined in the preceding steps. Among these constraints, the electricity purchase ratio binding constraint is a key constraint that directly affects the decision-making process. It is stated as follows: when there is a surplus in the photovoltaic power generation in the service area, the electricity purchase ratio must not be lower than a lower limit calculated from the photovoltaic surplus and charging demand.

[0061] Obtain the dynamic contract electricity price strategy and the real-time electricity price of the public grid, along with the current constraint lower bound. If the dynamic contract electricity price strategy is lower than the real-time electricity price of the public grid, purchasing electricity from the service area is cheaper, and the charging station tends to increase the proportion of electricity purchased. In this case, the optimal solution is a purchase ratio of 1, meaning all electricity is purchased from the service area, but this must satisfy the lower bound of the purchase ratio ≤ 1, which usually holds true. If the dynamic contract electricity price strategy is higher than the real-time electricity price of the public grid, purchasing electricity from the grid is cheaper, and the charging station tends to decrease the purchase ratio. In this case, the optimal solution should take the minimum allowed value. If the lower bound of the purchase ratio = 0, then all electricity is purchased from the grid. If the dynamic contract electricity price strategy equals the real-time electricity price of the public grid, the costs on both sides are the same, so the value of the purchase ratio has no impact on the cost, and usually an intermediate value is sufficient.

[0062] Charging stations do not need to receive direct power control commands from service areas. Instead, they act as independent market entities, making decisions that are most beneficial to themselves based on clear price signals and rules. When the public grid electricity price is abnormally high, the service area can provide a relatively stable contract electricity price through energy storage discharge. By comparing prices, charging stations will naturally increase the proportion of local electricity purchases. This is equivalent to using market mechanisms to provide a hedging option for charging loads and mitigate the impact of external market risks on the cost of terminal charging.

[0063] S600: Based on the power purchase ratio and the system operation constraint set, perform coordinated scheduling and control of photovoltaic power generation allocation, energy storage system charging and discharging power, charging load power supply path, and interaction power between the microgrid in the high-speed service area and the public power grid.

[0064] Specifically, based on the electricity purchase ratio, system operating constraints, and the latest real-time operating status, an optimization problem is constructed with the goal of minimizing operating costs. This problem uses photovoltaic (PV) grid-connected power, energy storage charging and discharging power, and grid interaction power as decision variables. The power delivered to charging stations in the energy balance equation is a known quantity calculated from the electricity purchase ratio and total charging demand: Power delivered to charging stations = Electricity purchase ratio × Total charging demand. Under the premise of satisfying all energy storage constraints and grid interaction power constraints, the algorithm solves how to combine PV power allocation, energy storage charging and discharging, and grid buying and selling to meet the determined base load and power supply to charging stations at the lowest cost or highest return. For example, when PV output is sufficient and the electricity purchase ratio is high, PV power is prioritized for meeting the base load and power supply to charging stations, with the remainder used for charging energy storage. If there is still a surplus and selling electricity is cost-effective, it is fed into the grid. If PV power is insufficient, the algorithm compares the energy storage discharging cost with the grid purchasing cost to decide whether to supplement with energy storage discharging or purchase electricity from the grid.

[0065] After the solution is obtained, the results are converted into specific equipment setpoints. Power reference values ​​or operating mode commands, such as maximum power point tracking or power-limited operation, are sent to the photovoltaic inverter; explicit charging and discharging power commands, such as charging at 300kW or discharging at 200kW, are sent to the energy storage converter; and power exchange commands, such as absorbing 50kW from the grid or feeding 100kW back to the grid, are sent to the connection point controller between the microgrid and the grid. Through internal metering and switching control, the power mix ratio obtained by the charging pile is ensured to be consistent with the purchased power ratio. The equipment executes the commands while continuously monitoring whether the actual operating power is consistent with the setpoint, and feeding back the actual SOC, actual interactive power, and other statuses to the system as the initial state for the next scheduling cycle, forming a closed-loop control.

[0066] The abstract game equilibrium and optimization objectives are translated into a set of physical control commands that can be directly executed by power equipment, completing a closed loop from the information layer, decision-making layer to the physical layer. This ensures that, under the premise of satisfying all constraints, multiple dimensions of objectives such as cost, photovoltaic consumption, and equipment and grid stability are synergistically optimized and balanced. The generation of all commands has been rigorously verified by the energy balance equation and various equipment safety constraints, fundamentally eliminating power imbalance, equipment overload, or battery damage caused by improper scheduling, thus enabling the reliable implementation of intelligent scheduling.

[0067] Furthermore, this application also includes the following steps: multi-agent game energy dispatch is executed periodically within a rolling time window, so that the power purchase ratio and interactive power are dynamically updated as photovoltaic output and charging load change.

[0068] Specifically, when the internal clock reaches a preset cycle point, a new scheduling cycle is immediately initiated. First, the latest real-time operating status is collected, including the current photovoltaic output, energy storage SOC, and actual load. This data is then used to overwrite the first point of the previous cycle's predicted data, serving as the absolutely accurate initial state for this rolling optimization. Based on the latest measured data, the photovoltaic output prediction model and the charging load prediction model are invoked to regenerate a prediction curve covering the entire rolling time window. The predicted data is then used to reconstruct a complete optimization model containing the objective function and all constraints. The rolling time window is a range of future time covered by the optimization calculation at any given decision point. This time window is not fixed but rolls forward over time. For example, optimization is always performed starting from the current moment for the next 4 hours, but it is re-executed every 15 minutes. Therefore, the starting point of each optimization moves forward by 15 minutes, and the end of the window also rolls forward by 15 minutes accordingly, always maintaining coverage of the next 4 hours.

[0069] In the newly predicted model, the Stackelberg game algorithm is re-run to find the optimal dynamic contract electricity price sequence and the corresponding optimal power purchase ratio sequence for each time period within the entire rolling window. Finally, only the decision results of the first time period in the rolling window—that is, the next period from T to T+1 to be executed—including power purchase ratio, energy storage charging and discharging power, and grid interaction power, are locked and issued as actual control commands. The issued commands are executed, and the system enters a waiting state until the next cycle's clock trigger point arrives. This process is repeated, with rolling optimization based on the latest actual state to generate commands for the next time period, and so on. The core outputs of the scheduling strategy—the power purchase ratio and the interaction power with the public grid—are dynamically updated and recalculated and adjusted in each new execution cycle, ensuring continuous synchronization between the scheduling strategy and the actual state of the high-speed service area microgrid.

[0070] By introducing a periodic execution and dynamic update mechanism within a rolling time window, it is ensured that at any given moment, the executed strategy is a locally optimal solution calculated based on the latest and most accurate information. The system no longer rigidly executes an outdated plan, but rather, like an autonomous vehicle, continuously perceives the environment, recalculates and fine-tunes its direction every second, and always proceeds along the optimal trajectory under current conditions, demonstrating strong environmental adaptability.

[0071] Furthermore, this application also includes the following steps: in the process of determining the electricity purchase ratio, the electricity purchase ratio is coupled with the local photovoltaic absorption capacity of the highway service area to limit the purchased power or grid-connected power in the case of photovoltaic surplus.

[0072] Specifically, the electricity purchase ratio is forcibly linked to the local photovoltaic (PV) absorption capacity of highway service areas, directly limiting the value of one quantity to the magnitude of the other. The local PV absorption capacity, i.e., surplus power, is the ratio of the service area's maximum power transmission capacity to charging stations. The electricity purchase ratio must be greater than or equal to this ratio, falling within the range of 0 to 1. The ratio of surplus power to the transmission limit calculates a theoretical minimum absorption ratio requirement. When the PV surplus is large, this ratio may be greater than 1, but due to the constraint that the electricity purchase ratio ≤ 1, the constraint effectively applies as an electricity purchase ratio = 1, meaning full absorption is mandatory. This constraint is incorporated into its Stackelberg game model. Operators know that once an electricity price is set, charging stations must meet at least the minimum electricity purchase ratio when making decisions. Therefore, operators anticipate and utilize this when setting prices. For example, when there is a PV surplus, even if operators set the electricity price slightly higher than the grid price, but still within a reasonable range, the constraint forces charging stations to purchase a significant proportion of local PV power because the electricity purchase ratio must be sufficiently high. The cost minimization model for charging stations is no longer a simple comparison of service area electricity prices and grid electricity prices. It has become a constrained optimization. That is, even if the service area electricity price is slightly higher than the grid electricity price, as long as it is not high enough for the charging station to feel that the penalty for violating the constraint is lower, the charging station must choose the minimum electricity purchase ratio instead of not purchasing at all, thus ensuring the bottom line of photovoltaic power consumption.

[0073] By coupling the electricity purchase ratio with the photovoltaic (PV) grid integration capacity, this rule ensures that PV grid integration does not fall below a reasonable level determined by physical conditions, even when economic incentives are weak or distorted. For service area operators, this constraint provides a definite expectation of the minimum purchase volume for charging stations when formulating electricity pricing strategies, reducing market risk. For the entire microgrid, the PV power integration path is more certain, reducing power fluctuations caused by the randomness of game outcomes and making energy dispatch more stable and reliable.

[0074] In summary, the multi-agent game-theoretic energy dispatching method for highway service area microgrids provided in this application has the following technical effects: By collecting real-time output status of the photovoltaic power generation system, state of charge of the energy storage system, base load status of the service area, and charging load demand status of electric vehicles during the operation of the highway service area microgrid, a real-time operating status is established; based on the energy interaction relationship between the highway service area operator and the charging station operator, a multi-agent energy interaction model is constructed. This model characterizes the energy acquisition rights, adjustable power ranges, and grid-connected power constraints of each agent under different energy source conditions; based on this multi-agent energy interaction model… Based on the real-time operating status, a system operation constraint set is generated to constrain the energy flow allocation between the microgrid's source, load, and storage network. Under the constraints of this system operation constraint set, the highway service area operator generates a dynamic contract electricity price strategy. The charging station operator, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, determines the electricity purchase ratio of the charging load from the highway service area microgrid versus the public grid, under the constraints of the system operation constraint set. According to the purchase ratio and the system operation constraint set, coordinated scheduling control is performed on the photovoltaic power generation allocation, the energy storage system's charging and discharging power, the charging load's power supply path, and the interaction power between the highway service area microgrid and the public grid. In other words, by collecting the real-time operating status of the photovoltaic power generation system, a multi-entity energy interaction model is used to optimize the coordinated scheduling of various energy sources in the microgrid, establishing an interactive relationship between the service area operator and the charging station operator. Furthermore, by optimizing the energy acquisition permissions, power ranges, and grid-connected power constraints among the entities, coordinated optimization among them is achieved. Based on the operating constraint set and the real-time electricity price, the electricity purchase ratio of the charging load is flexibly adjusted, improving energy utilization efficiency while ensuring the stability of the highway service area microgrid.

[0075] Example 2: Based on the same inventive concept as the multi-agent game energy dispatch method for a highway service area microgrid in Example 1, this application also provides a multi-agent game energy dispatch system for a highway service area microgrid. Please refer to Figure 2. The multi-agent game energy dispatch system for a highway service area microgrid includes: a status acquisition module 11, used to acquire the real-time output status of the photovoltaic power generation system, the state of charge of the energy storage system, the service area basic load status, and the electric vehicle charging load demand status during the operation of the highway service area microgrid, and establish a real-time operating status; and a model construction module 12, used to construct a multi-agent energy interaction model based on the energy interaction relationship between the highway service area operator and the charging station operator. The multi-agent energy interaction model is used to characterize the energy acquisition rights, adjustable power range, and grid connection power of each operator under different energy source conditions. The system includes: a rate constraint relationship; an operation constraint generation module 13, used to generate a system operation constraint set for constraining the energy flow allocation of the microgrid source-load-storage network based on the multi-entity energy interaction model and the real-time operating status; a strategy generation module 14, used by the highway service area operator to generate a dynamic contract electricity price strategy under the constraints of the system operation constraint set; a ratio determination module 15, used by the charging station operator to determine the electricity purchase ratio of the charging load from the highway service area microgrid and from the public grid under the constraints of the system operation constraint set, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid; and a collaborative scheduling control module 16, used to perform collaborative scheduling control on the photovoltaic power generation allocation, energy storage system charging and discharging power, charging load power supply path, and interaction power between the highway service area microgrid and the public grid, according to the electricity purchase ratio and the system operation constraint set.

[0076] Furthermore, the model building module 12 in the multi-agent game energy dispatch system of a high-speed service area microgrid is also used for: the multi-agent energy interaction model is built based on Stackelberg game, wherein the high-speed service area operator sets the dynamic contract electricity price as the leader, and the charging station operator adjusts the electricity purchase ratio according to the dynamic contract electricity price and the real-time electricity price of the public grid. The Stackelberg game solves the equilibrium through backward induction.

[0077] Furthermore, the operational constraint generation module 13 in the multi-agent game energy dispatch system for a high-speed service area microgrid is also used to: express the microgrid energy balance equation as: ;in, for Photovoltaic power generation at any given time for The discharge power of the energy storage system at any given time. for The charging power of the energy storage system at any time. for Power purchased by the power grid at any given time for Flexible load power at any time for The power transmitted from the operator of the highway service area to the main body of the charging station.

[0078] Furthermore, the operational constraint generation module 13 in the multi-agent game energy dispatch system of a high-speed service area microgrid is also used for: energy storage system constraints including a dynamic state of charge model, charging and discharging power constraints, and capacity constraints, wherein: the dynamic state of charge model is expressed as: ;in, for The state of charge of the energy storage system at all times. for The state of charge of the energy storage system at all times. For energy storage systems at time steps The total energy of discharge, The rated capacity of the energy storage system, For discharge efficiency, For energy storage systems at time steps Total input electrical energy, For charging efficiency; charging and discharging power constraints satisfy , and These are the minimum and maximum discharge power of the energy storage system, respectively; the capacity constraint satisfies... , and These are the minimum and maximum capacities of the energy storage system, respectively.

[0079] Furthermore, the strategy generation module 14 in the multi-agent game energy dispatch system of a high-speed service area microgrid is also used for: the dynamic contract electricity price strategy is constructed based on the time-of-use electricity price mechanism and supply and demand status perception.

[0080] Furthermore, the multi-agent game energy dispatch system for a high-speed service area microgrid also includes: the multi-agent game energy dispatch is executed periodically within a rolling time window, so that the power purchase ratio and the interactive power are dynamically updated as the photovoltaic output and charging load change.

[0081] Furthermore, the multi-agent game energy dispatch system for a high-speed service area microgrid also includes: in the process of determining the power purchase ratio, the power purchase ratio is coupled with the local photovoltaic absorption capacity of the high-speed service area to limit the purchased power or grid-connected power in the case of photovoltaic surplus.

[0082] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The multi-agent game energy scheduling method and specific examples of a high-speed service area microgrid in the foregoing embodiment one are also applicable to the multi-agent game energy scheduling system of a high-speed service area microgrid in this embodiment. Through the foregoing detailed description of the multi-agent game energy scheduling method of a high-speed service area microgrid, those skilled in the art can clearly understand the multi-agent game energy scheduling system of a high-speed service area microgrid in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0083] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0084] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A multi-agent game-theoretic energy scheduling method for a microgrid in a high-speed service area, characterized in that, include: During the operation of the microgrid in the highway service area, the real-time output status of the photovoltaic power generation system, the state of charge of the energy storage system, the basic load status of the service area, and the charging load demand status of electric vehicles are collected to establish the real-time operation status. Based on the energy interaction relationship between the highway service area operator and the charging station operator, a multi-entity energy interaction model is constructed. The multi-entity energy interaction model is used to characterize the energy acquisition rights, adjustable power range, and grid-connected power constraints of each entity under different energy source conditions. Based on the multi-agent energy interaction model and the real-time operating status, a system operation constraint set is generated to constrain the energy flow allocation between the microgrid's source, load, and storage network. Under the constraints of this system operation constraint set, the highway service area operator generates a dynamic contract electricity price strategy. Based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, and constrained by the system operation constraint set, the charging station operator determines the electricity purchase ratio between the charging load and the power obtained from the highway service area microgrid and the public grid. According to the purchase ratio and the system operation constraint set, coordinated scheduling and control are performed on the photovoltaic power generation allocation, the energy storage system's charging and discharging power, the charging load's power supply path, and the interaction power between the highway service area microgrid and the public grid.

2. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 1, characterized in that, The multi-agent energy interaction model is based on Stackelberg game theory, in which the highway service area operator sets the dynamic contract electricity price as the leader, and the charging station operator adjusts the electricity purchase ratio according to the dynamic contract electricity price and the real-time electricity price of the public grid. The Stackelberg game theory is solved by backward induction to find the equilibrium.

3. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 1, characterized in that, The system operation constraint set includes the microgrid energy balance equation, energy storage system constraints, and electricity purchase ratio binding constraints, wherein: the microgrid energy balance equation is expressed as: ;in, for Photovoltaic power generation at any given time for The discharge power of the energy storage system at any given time. for The charging power of the energy storage system at any time. for Power purchased by the power grid at any given time for Flexible load power at any time for The power transmitted from the operator of the highway service area to the main body of the charging station.

4. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 3, characterized in that, Energy storage system constraints include a dynamic state of charge model, charge / discharge power constraints, and capacity constraints. The dynamic state of charge model is expressed as: ;in, for The state of charge of the energy storage system at all times. for The state of charge of the energy storage system at all times. For energy storage systems at time steps The total energy of discharge, The rated capacity of the energy storage system, For discharge efficiency, For energy storage systems at time steps Total input electrical energy, For charging efficiency; charging and discharging power constraints satisfy ， and These are the minimum and maximum discharge power of the energy storage system, respectively; the capacity constraint satisfies... ， and These are the minimum and maximum capacities of the energy storage system, respectively.

5. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 1, characterized in that, The dynamic contract electricity pricing strategy is built upon time-of-use pricing mechanisms and supply and demand status awareness.

6. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 1, characterized in that, Multi-party game energy dispatch is executed periodically within a rolling time window, so that the power purchase ratio and interactive power are dynamically updated as photovoltaic output and charging load change.

7. The multi-agent game-theoretic energy scheduling method for a high-speed service area microgrid as described in claim 1, characterized in that, In determining the electricity purchase ratio, the electricity purchase ratio is coupled with the local photovoltaic absorption capacity of the highway service area to limit the purchased power or grid-connected power in the case of photovoltaic surplus.

8. A multi-agent game-theoretic energy dispatching system for a microgrid in a high-speed service area, characterized in that, The steps for implementing the multi-agent game-theoretic energy dispatching method for a high-speed service area microgrid according to any one of claims 1 to 7, wherein the multi-agent game-theoretic energy dispatching system for a high-speed service area microgrid includes: a status acquisition module, used to acquire the real-time output status of the photovoltaic power generation system, the state of charge of the energy storage system, the base load status of the service area, and the charging load demand status of electric vehicles during the operation of the high-speed service area microgrid, and establish a real-time operating status; a model building module, used to construct a multi-agent energy interaction model based on the energy interaction relationship between the high-speed service area operator and the charging station operator, wherein the multi-agent energy interaction model is used to characterize the energy acquisition rights, adjustable power range, and grid-connected power constraints of each operator under different energy source conditions; and an operation constraint generation module, used for... Based on the multi-entity energy interaction model and the real-time operating status, a system operation constraint set is generated to constrain the energy flow allocation of the microgrid's source-load-storage network. A strategy generation module is used by the highway service area operator to generate a dynamic contract electricity price strategy under the constraints of the system operation constraint set. A ratio determination module is used by the charging station operator to determine the electricity purchase ratio between the charging load obtaining electricity from the highway service area microgrid and from the public grid, based on the dynamic contract electricity price strategy and the real-time electricity price of the public grid, under the constraints of the system operation constraint set. A collaborative scheduling control module is used to perform collaborative scheduling control on the photovoltaic power generation allocation, energy storage system charging and discharging power, charging load power supply path, and the interaction power between the highway service area microgrid and the public grid, based on the electricity purchase ratio and the system operation constraint set.

Images