Non-cooperative game optimization method and device of multi-virtual power plant and load aggregator, medium

Abstract

The application discloses a kind of non-cooperative game optimization method and equipment of multiple virtual power plants and load aggregator, medium, and it is related to electric power system and energy market optimization scheduling technical field.The method comprises: establishing the energy transaction system consisting of multiple virtual power plants and load aggregator;The internal energy unit of each virtual power plant in energy transaction system is modeled respectively to obtain the state and / or constraint model of internal energy unit;Adopting ladder type carbon pricing mechanism, carbon emission cost is included in virtual power plant revenue function, and load aggregator revenue function is established by demand response to maximize comprehensive benefit;Non-cooperative game model is constructed, and C-ADMM algorithm is used to iteratively solve non-cooperative game model under the condition of meeting energy balance and transaction constraints, Nash equilibrium is realized.The application improves the economy, low carbon and distributed collaborative ability of multi-agent energy transaction system.

Description

Technical Field

[0001] This invention relates to the field of power system and energy market optimization scheduling technology, specifically to a non-cooperative game optimization method, equipment, and medium for multiple virtual power plants and load aggregators. Background Technology

[0002] With the rapid development of new power systems, virtual power plants (VPPs) incorporating photovoltaic, wind power, energy storage, and integrated energy equipment are gradually becoming important means to improve the absorption capacity of new energy sources and energy utilization efficiency. Meanwhile, load aggregators (LAs) participate in demand response and energy trading by aggregating industrial, commercial, and residential loads, enabling user-side resources to participate in the electricity market more flexibly. Against this backdrop, multi-energy coupled operation modes such as electricity, heat, natural gas, and hydrogen energy are constantly developing, and collaborative trading and distributed optimization between multiple VPPs and LAs are gradually becoming important directions in integrated energy system research.

[0003] However, most existing technologies focus on single VPP scheduling or centralized energy optimization, lacking a unified modeling of the non-cooperative competitive relationship between multiple VPPs and energy storage (LA). Furthermore, they fail to adequately consider the synergistic effects of gas turbine hydrogen co-firing, carbon capture and storage (CCS) devices, and tiered carbon trading mechanisms, making it difficult to effectively balance system economics with low-carbon operation requirements. In addition, traditional centralized solution methods suffer from high computational complexity, poor privacy protection, and low distributed collaborative efficiency in multi-entity, multi-constraint, and high-dimensional energy trading scenarios, making them unsuitable for practical applications in large-scale multi-energy market environments. Summary of the Invention

[0004] This invention addresses the problem that current collaborative methods between VPPs and LAs are insufficient to meet practical application needs. It provides a non-cooperative game optimization method, equipment, and medium for multiple virtual power plants and load aggregators, thereby improving the economy, low carbon emissions, and distributed collaborative capabilities of multi-entity energy trading systems.

[0005] On the one hand, a non-cooperative game optimization method for multiple virtual power plants and load aggregators is provided, characterized in that the method includes:

[0006] Establish an energy trading system consisting of multiple virtual power plants and load aggregators;

[0007] The internal energy units of each virtual power plant in the energy trading system are modeled to obtain the state and / or constraint models of the internal energy units.

[0008] A tiered carbon pricing mechanism is adopted, incorporating carbon emission costs into the revenue function of virtual power plants. Combined with the gas turbine, hydrogen co-firing, electricity purchase, and state and / or constraint models of the internal energy units, the carbon cost of each virtual power plant is calculated, and a revenue function of the virtual power plant is constructed based on the carbon cost. In addition, revenue is obtained through demand response, and a revenue function of load aggregator is established with the goal of maximizing comprehensive benefits.

[0009] A non-cooperative game model is constructed based on the revenue function of the virtual power plant, the revenue function of the load aggregator, the strategy of the virtual power plant, and the strategy of the load aggregator. The C-ADMM algorithm is then used to iteratively solve the non-cooperative game model under the conditions of energy balance and transaction constraints to achieve Nash equilibrium.

[0010] In some embodiments, the internal energy unit includes at least: a photovoltaic generator, a wind turbine, a gas turbine, an electrical energy storage system, and a carbon capture and storage device.

[0011] Specifically, the internal energy units of each virtual power plant in the energy trading system are modeled to obtain the state and / or constraint models of the internal energy units, including:

[0012] For photovoltaic generators and wind turbines, output constraint models are established respectively, wherein the actual output of the photovoltaic generator does not exceed the predicted output of photovoltaic power generation, and the actual output of the wind turbine does not exceed the predicted output of wind power generation.

[0013] For gas turbines, an output power and constraint model is established under the condition of co-firing natural gas and hydrogen:

[0014] ;

[0015] ;

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] in, This represents the gas-to-electricity heat energy conversion coefficient of a gas turbine. For the energy conversion efficiency of gas turbines, , The gas turbine at time The amount of natural gas and hydrogen imported. , These are the electrical power and thermal power output by the gas turbine, respectively. For the start-up and shutdown variables of the gas turbine, This is the maximum intake air volume of the gas turbine. This represents the ramping limit of a gas turbine. , These are the minimum and maximum values ​​of the heat-to-electricity ratio of a gas turbine, respectively. This represents the maximum hydrogen blending ratio for the gas turbine. Carbon dioxide produced by gas turbine The amount, This is the conversion factor between natural gas and carbon dioxide per unit volume;

[0021] For carbon capture and storage devices, establish a model constrained by fixed energy consumption and carbon capture efficiency:

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] in, For carbon capture and storage devices at all times Energy consumption For the start-up and shutdown variables of carbon capture and storage devices, For the fixed energy consumption of carbon capture and storage devices, , Carbon captured by carbon capture and storage devices Energy consumption and capture per unit mass Energy consumption To capture The efficiency coefficient, For maximum energy consumption, The ramp-up limit for carbon capture and storage devices;

[0027] For energy storage systems, a charge / discharge state and state of charge constraint model is established:

[0028] ;

[0029] ;

[0030] ;

[0031] ;

[0032] ;

[0033] ;

[0034] in, For the energy storage system at all times The state of charge, , These are the charging power and discharging power of the energy storage system, respectively. The energy loss rate of the electric energy storage system. , These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. , These represent the maximum charging power and the maximum discharging power of the energy storage system, respectively. , These are the 0 / 1 variables for charging and discharging, respectively. , These represent the maximum and minimum values ​​of the state of charge of the energy storage system, respectively. The interval time;

[0035] The method further includes modeling user-reducible load to describe load changes following demand response:

[0036] ;

[0037] ;

[0038] ;

[0039] in, , Virtual power plants At any moment Load before and after participating in demand response; For loads that can be reduced, The proportion of load that can be reduced, , These are conventional electrical load and thermal load, respectively.

[0040] In some embodiments, the carbon cost of each virtual power plant is calculated using the following formula:

[0041] ;

[0042] ;

[0043] ;

[0044] in, It represents the penalty price for the carbon emissions of a virtual power plant at each stage. This is the actual carbon emissions of the virtual power plant. It is the carbon emission allocation of the virtual power plant. It is the length of the carbon step interval. It is the carbon emission distribution factor of the gas turbine. , These are the electrical power and thermal power output by the gas turbine, respectively. This refers to the electricity purchased by the virtual power plant from the electricity market. It is the carbon emission allocation factor that the virtual power plant purchases from the electricity market. For gas turbine The amount, Capture for carbon capture and storage devices Energy consumption This represents the carbon emission coefficient corresponding to the virtual power plant's purchase of electricity in the electricity market;

[0045] The revenue function of the virtual power plant is:

[0046] ;

[0047] ;

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] ;

[0054] ;

[0055] in, To find the maximum value function, It is a virtual power plant Revenue from selling electrical and thermal power to users It is a virtual power plant With virtual power plants Energy trading revenue between them It is the cost of electricity purchased by the virtual power plant from the grid. It is the cost for the virtual power plant to purchase natural gas from the natural gas pipeline network. It is the cost of purchasing heat power through a heating network. This refers to the cost for the virtual power plant to purchase hydrogen from the hydrogen network. It is the cost of various devices in the virtual power plant. It is the fee paid to the distributed network when virtual power plants trade energy. It is the carbon cost of a virtual power plant; and It is a virtual power plant at all times The electrical and thermal power sold to users; and These are the energy prices for electrical power and thermal power, respectively. and It is a virtual power plant With virtual power plants The price of electrical and thermal power transactions between them; and It refers to the electrical and thermal power traded between virtual power plants; and It is the price at which a virtual power plant buys and sells electricity from the grid; and It is a virtual power plant Electricity purchased and sold from the power grid; It is the price at which the virtual power plant purchases natural gas from the natural gas pipeline network; This is the amount of natural gas that the virtual power plant purchases from the natural gas pipeline network; It is the price at which the virtual power plant purchases hydrogen from the hydrogen network; It is a virtual power plant The amount of hydrogen purchased from the hydrogen network; and It refers to the price of buying and selling heat power through a heat network; and It is a virtual power plant Heat power purchased and sold from the heating network; These are the cost coefficients of various devices in a virtual power plant; and It is the price coefficient for electricity and heat paid to the distributed network; , These represent the natural gas and hydrogen input quantities for the gas turbine, respectively. This refers to the thermal power output of the gas turbine. This refers to the number of virtual power plants traded. It refers to a time period;

[0056] The revenue function for the load aggregator is:

[0057] ;

[0058] ;

[0059] in, It's about virtual power plants. The user's energy use utility function; It is the benefit of user participation in demand response; It is the cost for users to purchase energy from virtual power plants. , It is the demand response price. This is a load that can be reduced.

[0060] In some embodiments, the user energy usage utility function is:

[0061] ;

[0062] in, , , , Indicating regarding virtual power plants The weighting coefficient of energy use utility , Virtual power plants At any moment The conventional electrical and thermal loads.

[0063] In some embodiments, the energy balance and trading constraints include:

[0064] Power balance constraints:

[0065] ;

[0066] in, For the output of photovoltaic generators; For the output of wind turbines; The electrical output of the gas turbine; For the discharge output of the energy storage system;

[0067] Thermal power balance constraint:

[0068] ;

[0069] in, The thermal power output of the gas turbine.

[0070] The natural gas balance constraint:

[0071] ;

[0072] in, This refers to the natural gas input to the gas turbine.

[0073] The hydrogen balance constraint:

[0074] ;

[0075] in, This refers to the amount of hydrogen input to the gas turbine.

[0076] Energy trading constraints:

[0077] ;

[0078] ;

[0079] ;

[0080] in, , Virtual power plants With virtual power plants The maximum trading volume of electrical and thermal power between them; , Virtual power plants With virtual power plants The volume of electrical and thermal power exchanged between them.

[0081] In some embodiments, the C-ADMM algorithm is used to iteratively solve the non-cooperative game model under the conditions of energy balance and trading constraints to achieve Nash equilibrium, including:

[0082] The multiple virtual power plants and the load aggregator are defined as participants in a non-cooperative game, wherein each of the multiple virtual power plants participates in energy trading as an independent whole, and the load aggregator participates in energy trading and demand response management as a user-side agent.

[0083] The energy trading relationship between the multiple virtual power plants and the load aggregator is constructed as a non-cooperative game model to describe the competition and cooperation among the market participants:

[0084] ;

[0085] in, Representing a game theory model, For the set of game participants, For virtual power plants Strategy set, For a set of load aggregators, , These are the revenue functions for virtual power plants and load aggregators, respectively.

[0086] Define virtual power plant (VPS) strategies and load aggregator strategies, wherein the VPS strategy includes energy equipment output and trading capabilities, and the load aggregator strategy includes demand response and trading capabilities with VPS.

[0087] ;

[0088] ;

[0089] in, , These refer to the natural gas and hydrogen input rates for the gas turbine, respectively. Captured by carbon capture and storage devices quantity, , These refer to the charging power and discharging power of the energy storage system, respectively. To respond to load demand, , These are respectively electrical power trading volume and thermal power trading volume;

[0090] Using the C-ADMM algorithm, each virtual power plant and load aggregator continuously adjusts its own strategy, and revenue is calculated using the revenue functions of virtual power plants and load aggregators. When all participants no longer adjust their strategies, the energy trading market reaches Nash equilibrium.

[0091] ;

[0092] in, , Virtual power plants The optimal set of strategies for load aggregators under Nash equilibrium.

[0093] In some embodiments, the C-ADMM algorithm is employed, whereby each virtual power plant and load aggregator continuously adjusts its own strategy, and revenue is calculated using the revenue functions of virtual power plants and load aggregators. When all participants cease adjusting their strategies, the energy trading market reaches Nash equilibrium, including:

[0094] The C-ADMM algorithm is used to solve the non-cooperative game model of multiple virtual power plants and load aggregators in a distributed manner. The dual variable is updated by the average value of the variable, and energy trading consensus is achieved without introducing auxiliary variables. The energy trading problem is decomposed into a sub-problem of updating trading power and a sub-problem of updating trading price.

[0095] The virtual power plant adjusts its internal energy output and trading power based on the current trading price to achieve power decision updates. The objective function for updating the trading power in this power decision update is:

[0096] ;

[0097] in, Update the objective function for trading power. , Virtual power plants With virtual power plants Electricity and heat transactions between them at any time The corresponding dual variable, and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and These are the energy prices for electrical power and thermal power, respectively. For penalty parameters, The overall operating cost of a virtual power plant. This refers to the number of virtual power plants traded. It refers to a time period;

[0098] Based on the updated power trading results, the dual variables corresponding to the power trading price and the heat trading price are updated:

[0099] ;

[0100] ;

[0101] in, , These are the updated dual variables for electricity and heat trading, respectively.

[0102] The convergence of the C-ADMM algorithm is determined by the residual value. When the residual value of market participants is less than a preset threshold after iterative updates, the C-ADMM algorithm is considered to have converged and reached a global Nash equilibrium.

[0103] Secondly, a non-cooperative game optimization device for multiple virtual power plants and load aggregators is provided, the device comprising:

[0104] The energy trading system construction module is used to: establish an energy trading system consisting of multiple virtual power plants and load aggregators;

[0105] The internal energy unit modeling module is used to: model the internal energy units of each virtual power plant in the energy trading system to obtain the state and / or constraint models of the internal energy units;

[0106] The revenue function construction module is used to: adopt a tiered carbon pricing mechanism to incorporate carbon emission costs into the revenue function of virtual power plants, and combine the state and / or constraint models of gas turbines, hydrogen co-firing and electricity purchase, as well as the internal energy units, to calculate the carbon cost of each virtual power plant and construct the revenue function of virtual power plants based on the carbon cost; and to establish the revenue function of load aggregator with the goal of maximizing comprehensive benefits by obtaining revenue through demand response.

[0107] The Nash equilibrium calculation module is used to: construct a non-cooperative game model based on the revenue function of the virtual power plant and the revenue function of the load aggregator, the strategy of the virtual power plant and the strategy of the load aggregator, and use the C-ADMM algorithm to iteratively solve the non-cooperative game model under the conditions of satisfying energy balance and transaction constraints to achieve Nash equilibrium.

[0108] Thirdly, an electronic device is provided, including a processor and a memory, the memory storing a computer program, wherein the processor executes the computer program to implement the method described in any of the above embodiments.

[0109] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and the computer program, when executed by a processor, implements the method described in any of the preceding embodiments.

[0110] Compared with existing technologies, this invention has the following advantages and beneficial effects: By constructing a non-cooperative game model between multiple VPPs and LAs, and combining it with a gas turbine hydrogen co-firing model, a carbon capture and storage model, a tiered carbon trading mechanism, and a demand response model, the invention achieves coordinated and optimized operation of multiple energy systems under low-carbon constraints. Simultaneously, the C-ADMM algorithm is used to solve the game model in a distributed manner, achieving global Nash equilibrium under limited information interaction conditions, thereby improving the economy, low-carbon nature, and distributed coordination capabilities of the multi-entity energy trading system. Attached Figure Description

[0111] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0112] Figure 1 This is a schematic diagram of a non-cooperative game-based energy trading framework for multiple VPPs and LAs according to an embodiment of the present invention.

[0113] Figure 2 This is a flowchart of a non-cooperative game optimization method for multiple virtual power plants and load aggregators according to an embodiment of the present invention.

[0114] Figure 3 This is a structural block diagram of a non-cooperative game optimization device for multiple virtual power plants and load aggregators according to an embodiment of the present invention.

[0115] Figure 4 This is a schematic diagram of a non-cooperative game optimization device for multiple virtual power plants and load aggregators according to an embodiment of the present invention. Detailed Implementation

[0116] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0117] On the one hand, the present invention provides a non-cooperative game optimization method for multiple virtual power plants and load aggregators. Figure 1 This is a schematic diagram of a non-cooperative game-based energy trading framework for multiple VPPs and LAs according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating a non-cooperative game optimization method for multiple virtual power plants and load aggregators according to an embodiment of the present invention. (Reference) Figure 1 and Figure 2 The non-cooperative game optimization method for multiple virtual power plants and load aggregators includes steps S10 to S40.

[0118] In S10, an energy trading system is established, consisting of multiple virtual power plants (VPPs) and load aggregators (LAs). VPPs can include photovoltaic generators, wind turbines, gas turbines, energy storage systems, and carbon capture and storage devices for multi-energy production and trading; LAs can manage industrial, commercial, and residential user loads through edge computing terminals, participate in demand response, and conduct energy trading.

[0119] like Figure 1 As shown, a VPP includes various energy production units for electricity, heat, hydrogen, and natural gas, such as photovoltaic (PV), wind power (WT), gas turbines (GT), energy storage systems (ESS), and carbon capture and storage (CCS). Multiple VPPs aim to maximize their respective benefits. LA manages users' demand response (DR) and energy trading information by configuring edge computing terminals (ECTs) for three types of users (industrial, commercial, and residential users). LA aims to maximize overall benefits. To protect user privacy, market participants only obtain partial solutions to their optimization strategies through limited trading information.

[0120] In S20, the energy units inside the VPP are modeled separately, including PV / WT output constraints, gas turbine hydrogen co-firing and power constraints, CCS energy consumption and carbon capture efficiency constraints, ESS charge and discharge and state of charge constraints, and user-reducible load (DR) model.

[0121] Specifically, an output constraint model is established for PV and WT, for example, the actual output of PV does not exceed the predicted output:

[0122] ; (1)

[0123] in, For PV at any time Actual output For PV at any time The predicted output. WT is constrained similarly.

[0124] Modeling the output power and constraints of a gas turbine under natural gas and hydrogen co-firing conditions:

[0125] ; (2)

[0126] ; (3)

[0127] ; (4)

[0128] ; (5)

[0129] ; (6)

[0130] ; (7)

[0131] in, This represents the gas-to-electricity heat energy conversion coefficient of a gas turbine. For the energy conversion efficiency of gas turbines, , The gas turbine at time The amount of natural gas and hydrogen imported. , These are the electrical power and thermal power output by the gas turbine, respectively. For the start-up and shutdown variables of the gas turbine, This is the maximum intake air volume of the gas turbine. This represents the ramping limit of a gas turbine. , These are the minimum and maximum values ​​of the heat-to-electricity ratio of a gas turbine, respectively. This represents the maximum hydrogen blending ratio for the gas turbine. Carbon dioxide produced by gas turbine The amount, This is the conversion factor between natural gas and carbon dioxide per unit volume.

[0132] Modeling the CCS, including fixed energy consumption and carbon capture efficiency constraints:

[0133] ; (8)

[0134] ; (9)

[0135] ; (10)

[0136] ; (11)

[0137] in, For carbon capture and storage devices at all times Energy consumption For the start-up and shutdown variables of carbon capture and storage devices, For the fixed energy consumption of carbon capture and storage devices, , Carbon captured by carbon capture and storage devices Energy consumption and capture per unit mass Energy consumption To capture The efficiency coefficient, For maximum energy consumption, This represents the ramp-up limit for carbon capture and storage (CCS) devices.

[0138] Constraints are imposed on the charge / discharge state and state of charge of the ESS:

[0139] ;(12)

[0140] ; (13)

[0141] ;(14)

[0142] ; (15)

[0143] ; (16)

[0144] ; (17)

[0145] in, For the energy storage system at all times The state of charge, , These are the charging power and discharging power of the energy storage system, respectively. The energy loss rate of the electric energy storage system. , These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. , These represent the maximum charging power and the maximum discharging power of the energy storage system, respectively. , These are the 0 / 1 variables for charging and discharging, respectively. , These represent the maximum and minimum values ​​of the state of charge of the energy storage system, respectively. This refers to the interval time.

[0146] Model the load that users can reduce to describe the load changes after demand response:

[0147] ; (18)

[0148] ; (19)

[0149] ; (20)

[0150] in, , Virtual power plants At any moment Load before and after participating in demand response; For loads that can be reduced, The proportion of load that can be reduced, , Virtual power plants At any moment The conventional electrical and thermal loads.

[0151] In S30, a tiered carbon pricing mechanism is adopted, incorporating carbon emission costs into the revenue function of virtual power plants. Combined with the gas turbine, hydrogen co-firing, electricity purchase, and the state and / or constraint models of the internal energy units, the carbon cost of each virtual power plant is calculated, and the revenue function of the virtual power plant is constructed based on the carbon cost. In addition, revenue is obtained through demand response, and the revenue function of the load aggregator is established with the goal of maximizing comprehensive benefits.

[0152] There are two main pricing mechanisms for carbon trading: fixed-price carbon trading and tiered carbon trading. Compared to fixed-price carbon trading, tiered carbon trading can have a more significant impact on carbon emission reduction and has a better incentive effect on energy entities. Based on the tiered carbon trading pricing mechanism, the carbon trading costs for VPPs are as follows:

[0153] (twenty one)

[0154] (twenty two)

[0155] (twenty three)

[0156] in, It represents the penalty price for the carbon emissions of a virtual power plant at each stage. This is the actual carbon emissions of the virtual power plant. It is the carbon emission allocation of the virtual power plant. It is the length of the carbon step interval. It is the carbon emission distribution factor of the gas turbine. , These are the electrical power and thermal power output by the gas turbine, respectively. This refers to the electricity purchased by the virtual power plant from the electricity market. It is the carbon emission allocation factor that the virtual power plant purchases from the electricity market. For gas turbine The amount, Capture for carbon capture and storage devices Energy consumption This represents the carbon emission coefficient corresponding to the virtual power plant's purchase of electricity in the electricity market.

[0157] Each VPP participates in energy market transactions as an independent entity by integrating energy equipment, aiming to maximize its own interests. The objective function for VPP revenue (virtual power plant revenue function) is as follows:

[0158] ; (twenty four)

[0159] ; (25)

[0160] ; (26)

[0161] ;(27)

[0162] ; (28)

[0163] ; (29)

[0164] ; (30)

[0165] ; (31)

[0166] ; (32)

[0167] in, To find the maximum value function, It is a virtual power plant Revenue from selling electrical and thermal power to users It is a virtual power plant With virtual power plants Energy trading revenue between them It is the cost of electricity purchased by the virtual power plant from the grid. It is the cost for the virtual power plant to purchase natural gas from the natural gas pipeline network. It is the cost of purchasing heat power through a heating network. This refers to the cost for the virtual power plant to purchase hydrogen from the hydrogen network. It is the cost of various devices in the virtual power plant. It is the fee paid to the distributed network when virtual power plants trade energy. It is the carbon cost of a virtual power plant; and It is a virtual power plant at all times The electrical and thermal power sold to users; and These are the energy prices for electrical power and thermal power, respectively. and It is a virtual power plant With virtual power plants The price of electrical and thermal power transactions between them; and It refers to the electrical and thermal power traded between virtual power plants; and It is the price at which a virtual power plant buys and sells electricity from the grid; and It is a virtual power plant Electricity purchased and sold from the power grid; It is the price at which the virtual power plant purchases natural gas from the natural gas pipeline network; This is the amount of natural gas that the virtual power plant purchases from the natural gas pipeline network; It is the price at which the virtual power plant purchases hydrogen from the hydrogen network; It is a virtual power plant The amount of hydrogen purchased from the hydrogen network; and It refers to the price of buying and selling heat power through a heat network; and It is a virtual power plant Heat power purchased and sold from the heating network; These are the cost coefficients of various devices in a virtual power plant; and It is the price coefficient for electricity and heat paid to the distributed network; , These represent the natural gas and hydrogen input quantities for the gas turbine, respectively. This refers to the thermal power output of the gas turbine. This refers to the number of virtual power plants traded. It refers to a time period.

[0168] As an agent for users, the LA can trade energy with the VPP and generate revenue through participation in demand response. The LA's objective is to maximize overall benefits, as defined by the following objective function:

[0169] ; (33)

[0170] ; (34)

[0171] in, It's about virtual power plants. The user's energy use utility function; It is the benefit of user participation in demand response; It is the cost for users to purchase energy from virtual power plants. , It is the demand response price. This is a load that can be reduced.

[0172] Considering the benefits that industrial and commercial users gain from energy consumption, a quadratic function is used to establish the user's energy use utility function model. For users participating in demand response, it is impractical to consider only the revenue after demand rationing (DR) and energy purchase costs as the utility function, because users also gain benefits from the energy they consume, especially industrial and commercial users. Therefore, a quadratic function is used to represent the user's energy use utility function:

[0173] ; (35)

[0174] in, , , , Indicating regarding virtual power plants The weighting coefficient of energy use utility , Virtual power plants At any moment The conventional electrical and thermal loads.

[0175] In S40, a non-cooperative game model is constructed based on the revenue functions of virtual power plants and load aggregators, as well as the strategies of virtual power plants and load aggregators. The C-ADMM algorithm is then used to iteratively solve the non-cooperative game model under the conditions of energy balance and transaction constraints to achieve Nash equilibrium.

[0176] Energy balance and trading constraints include: electricity balance constraints, heat and power balance constraints, natural gas balance constraints, hydrogen balance constraints, and energy trading constraints.

[0177] Establish power balance constraints for the internal power input and output of the VPP to ensure power balance constraints:

[0178] (36)

[0179] in, For the output of photovoltaic generators; For the output of wind turbines; The electrical output of the gas turbine; For the discharge output of the energy storage system;

[0180] Thermal power balance constraint:

[0181] ; (37)

[0182] Natural gas balance constraints:

[0183] ; (38)

[0184] Hydrogen balance constraints:

[0185] ; (39)

[0186] Energy trading constraints are imposed on the volume and price of electricity and heat transactions between VPPs to ensure transaction feasibility and suppress arbitrage behavior (energy trading constraints):

[0187] (40)

[0188] in, and These represent the maximum trading volumes of electrical power and thermal power, respectively. Furthermore, neglecting energy transfer losses, the energy trading of VPP at time t should satisfy the following energy balance:

[0189] ; (41)

[0190] To encourage multi-VPP participation in energy trading and curb arbitrage, energy trading prices should be limited to a reasonable market price range.

[0191] ; (42)

[0192] As rational entities, VPPs actively participate in energy market transactions with the goal of maximizing their own profits. On the other hand, LAs manage the load demands of various user types, aiming to maximize overall benefits. Therefore, a conflict of interest exists between VPPs and LAs, and all market participants adjust their decisions based on each other's strategies. A game-theoretic relationship exists between VPPs and LAs. Constructing a game-theoretic framework based on the behavioral strategies of each entity helps market participants achieve an equilibrium of interests. The non-cooperative game model of multiple VPPs and LAs is as follows:

[0193] ; (43)

[0194] in, Represents a game theory model; For the set of game participants, For virtual power plants Strategy set, For a set of load aggregators, , These are the revenue functions for the virtual power plant and the load aggregator, respectively.

[0195] The three key elements that a game theory model must have are: participants, strategies, and payoffs.

[0196] Participants: VPP and LA. Each VPP participates in the transaction as an independent entity, while LA participates as a representative of the user side.

[0197] Strategy: VPP's strategy includes the output and trading capabilities of energy devices: LA's strategy includes DR and trading capabilities with VPP: .

[0198] Benefits: The benefits of the participants are calculated using equations (24) and (33).

[0199] The C-ADMM algorithm is employed, where virtual power plants and load aggregators continuously adjust their strategies, and revenue is calculated using the revenue functions of virtual power plants and load aggregators. When participants cease actively adjusting their strategies, the energy trading market achieves Nash equilibrium, with each participant's utility function reaching its optimal solution. The Nash equilibrium of the non-cooperative game is represented as follows:

[0200] ; (44)

[0201] Where * represents the optimal strategy.

[0202] This invention employs the C-ADMM algorithm to solve the distributed subproblems of a non-cooperative game model, achieving higher computational efficiency. It updates the dual variable through the average of the variables, achieving consensus in energy trading negotiations without requiring auxiliary variables. All market participants must reach a consensus on energy to determine the energy trading volume and price. Based on the C-ADMM algorithm, the energy trading problem is transformed into two subproblems, and market participants obtain the optimal solution to the global optimization problem by solving their local optimization problems.

[0203] (1) Sub-problem 1: Updating trading power. The VPP adjusts its internal energy output and trading power based on the current trading price, thereby updating its power decision:

[0204] ; (45)

[0205] in, Update the objective function for trading power. , Virtual power plants With virtual power plants Electricity and heat transactions between them at any time The corresponding dual variable, and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and These are the energy prices for electrical power and thermal power, respectively. For penalty parameters, The overall operating cost of a virtual power plant. This refers to the number of virtual power plants traded. It refers to a time period.

[0206] (2) Sub-problem 2: Updating transaction prices. According to

[0207] Sub-problem 1: Power decision update, VPP adjusts trading price:

[0208] ; (46)

[0209] ; (47)

[0210] in, , These are the updated dual variables for electricity and heat trading, respectively.

[0211] The residual value determines the convergence of the C-ADMM algorithm. As market participants update and iterate their optimization models, the residual value... Smaller than image The algorithm converges at this threshold. The convergence condition is as follows:

[0212] ; (48)

[0213] In this invention, by constructing a multi-entity energy trading framework consisting of multiple VPPs and LAs, and integrating photovoltaic power generation, gas turbines, energy storage systems, carbon capture and storage devices, and demand response mechanisms in a unified model, the coordinated operation of multiple energy sources such as electricity, heat, natural gas, and hydrogen is realized. This can effectively improve the resource utilization efficiency and new energy absorption capacity of the multi-energy system, while enhancing the flexibility of users to participate in energy trading and demand response.

[0214] Meanwhile, by introducing a hydrogen co-firing mechanism into the gas turbine and combining it with a tiered carbon trading pricing model and a CCS model, carbon emission costs are directly incorporated into the VPP revenue function, achieving dynamic game optimization of multiple VPPs under low-carbon constraints. This not only reduces natural gas consumption and system carbon emission levels but also improves the economic efficiency and low-carbon operation capabilities in the multi-entity energy trading process.

[0215] Furthermore, by establishing a non-cooperative game model between multiple VPPs and LAs, and using the C-ADMM algorithm to solve the profit maximization problem of each game player in a distributed manner, the energy trading process is decomposed into two sub-problems: power update and price update. This enables each market player to achieve local optimization and global Nash equilibrium convergence under limited information interaction conditions, thereby effectively improving the solution efficiency, distributed collaboration capability, and privacy protection capability of the large-scale multi-player energy trading system.

[0216] On the other hand, the present invention provides a non-cooperative game optimization device for multiple virtual power plants and load aggregators. Figure 3 This is a structural block diagram of a non-cooperative game optimization device for multiple virtual power plants and load aggregators according to an embodiment of the present invention. (Reference) Figure 3 The non-cooperative game optimization device for multiple virtual power plants and load aggregators includes: an energy trading system construction module, an internal energy unit modeling module, a revenue function construction module, and a Nash equilibrium calculation module.

[0217] The energy trading system building module is used to: establish an energy trading system consisting of multiple virtual power plants and load aggregators.

[0218] The internal energy unit modeling module is used to model the internal energy units of each virtual power plant in the energy trading system to obtain the state and / or constraint models of the internal energy units.

[0219] The revenue function construction module is used to: adopt a tiered carbon pricing mechanism to incorporate carbon emission costs into the revenue function of virtual power plants, and combine the gas turbine, hydrogen co-firing, and electricity purchase, as well as the state and / or constraint models of the internal energy units, to calculate the carbon cost of each virtual power plant and construct the revenue function of the virtual power plant based on the carbon cost; and to establish the revenue function of the load aggregator with the goal of maximizing comprehensive benefits by obtaining revenue through demand response.

[0220] The Nash equilibrium calculation module is used to: construct a non-cooperative game model based on the revenue functions of virtual power plants and load aggregators, as well as the strategies of virtual power plants and load aggregators, and then use the C-ADMM algorithm to iteratively solve the non-cooperative game model under the conditions of satisfying energy balance and transaction constraints to achieve Nash equilibrium.

[0221] For further details regarding the non-cooperative game optimization device for multiple virtual power plants and load aggregators, please refer to the previous description of the non-cooperative game optimization method for multiple virtual power plants and load aggregators, which will not be repeated here.

[0222] In the case of implementing the functions of the integrated modules described above in hardware, this embodiment of the invention provides another structure for the electronic device involved in the above embodiments. Figure 4 This is a schematic diagram of a non-cooperative game optimization device for multiple virtual power plants and load aggregators according to an embodiment of the present invention. (Reference) Figure 4 The electronic device includes: at least one processor; and at least one memory. The at least one memory is coupled to the at least one processor and stores instructions for execution by the at least one processor, which, when executed by the at least one processor, implement the method described above.

[0223] A processor can be a set of logic blocks, modules, and circuits that implement or execute the various exemplary logic blocks, modules, and circuits described in connection with embodiments of the present invention. The processor can be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in connection with embodiments of the present invention. A processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a digital signal processor and a microprocessor, etc.

[0224] The memory may be read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0225] In one implementation, the memory can exist independently of the processor. The memory can be connected to the processor via a bus and used to store instructions or program code. When the processor calls and executes the instructions or program code stored in the memory, it can implement the methods provided in the embodiments of the present invention. In another implementation, the memory can also be integrated with the processor.

[0226] On the other hand, the present invention also provides a computer-readable storage medium (e.g., a non-transitory computer-readable storage medium) storing computer program instructions that, when executed on a computer, cause the computer to perform the method as described in any of the above embodiments.

[0227] Exemplary examples show that the aforementioned computer-readable storage media may include, but are not limited to: magnetic storage devices (e.g., hard disks, floppy disks, or magnetic tapes), optical discs (e.g., compact disks (CDs), digital versatile disks (DVDs), etc.), smart cards, and flash memory devices (e.g., erasable programmable read-only memory (EPROMs), cards, sticks, or key drives, etc.). The various computer-readable storage media described in this invention may represent one or more devices and / or other machine-readable storage media for storing information. The term "machine-readable storage medium" may include, but is not limited to, wireless channels and various other media capable of storing, containing, and / or carrying instructions and / or data.

[0228] This invention provides a computer program that, when run on a computer, causes the computer to perform the method of any of the above embodiments.

[0229] This invention provides a computer program product containing instructions that, when run on a computer, cause the computer to perform the method of any of the above embodiments.

[0230] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A non-cooperative game optimization method for multiple virtual power plants and load aggregators, characterized in that, The method includes: Establish an energy trading system consisting of multiple virtual power plants and load aggregators; The internal energy units of each virtual power plant in the energy trading system are modeled to obtain the state and / or constraint models of the internal energy units. A tiered carbon pricing mechanism is adopted, incorporating carbon emission costs into the revenue function of virtual power plants. Combined with the gas turbine, hydrogen co-firing, electricity purchase, and state and / or constraint models of the internal energy units, the carbon cost of each virtual power plant is calculated, and a revenue function of the virtual power plant is constructed based on the carbon cost. In addition, revenue is obtained through demand response, and a revenue function of load aggregator is established with the goal of maximizing comprehensive benefits. A non-cooperative game model is constructed based on the revenue function of the virtual power plant, the revenue function of the load aggregator, the strategy of the virtual power plant, and the strategy of the load aggregator. The C-ADMM algorithm is then used to iteratively solve the non-cooperative game model under the conditions of energy balance and transaction constraints to achieve Nash equilibrium.

2. The method according to claim 1, characterized in that, The internal energy unit includes at least: a photovoltaic generator, a wind turbine, a gas turbine, an electrical energy storage system, and a carbon capture and storage device. Specifically, the internal energy units of each virtual power plant in the energy trading system are modeled to obtain the state and / or constraint models of the internal energy units, including: For photovoltaic generators and wind turbines, output constraint models are established respectively, wherein the actual output of the photovoltaic generator does not exceed the predicted output of photovoltaic power generation, and the actual output of the wind turbine does not exceed the predicted output of wind power generation. For gas turbines, an output power and constraint model is established under the condition of co-firing natural gas and hydrogen: ； ； ； ； ； ； in, This represents the gas-to-electricity heat energy conversion coefficient of a gas turbine. For the energy conversion efficiency of gas turbines, , The gas turbine at time The amount of natural gas and hydrogen imported. , These are the electrical power and thermal power output by the gas turbine, respectively. For the start-up and shutdown variables of the gas turbine, This is the maximum intake air volume of the gas turbine. This represents the ramping limit of a gas turbine. , These are the minimum and maximum values ​​of the heat-to-electricity ratio of a gas turbine, respectively. This represents the maximum hydrogen blending ratio for the gas turbine. Carbon dioxide produced by gas turbines The amount, This is the conversion factor between natural gas and carbon dioxide per unit volume; For carbon capture and storage devices, establish a model constrained by fixed energy consumption and carbon capture efficiency: ； ； ； ； in, For carbon capture and storage devices at all times Energy consumption For the start-up and shutdown variables of carbon capture and storage devices, For the fixed energy consumption of carbon capture and storage devices, , Carbon captured by carbon capture and storage devices Energy consumption and capture per unit mass Energy consumption To capture The efficiency coefficient, For maximum energy consumption, The ramp-up limit for carbon capture and storage devices; For energy storage systems, a charge / discharge state and state of charge constraint model is established: ； ； ； ； ； ； in, For the energy storage system at all times The state of charge, , These are the charging power and discharging power of the energy storage system, respectively. The energy loss rate of the electric energy storage system. , These refer to the charging efficiency and discharging efficiency of the energy storage system, respectively. , These represent the maximum charging power and the maximum discharging power of the energy storage system, respectively. , These are the 0 / 1 variables for charging and discharging, respectively. , These represent the maximum and minimum values ​​of the state of charge of the energy storage system, respectively. This refers to the interval time. The method further includes modeling user-reducible load to describe load changes following demand response: ； ； ； in, , Virtual power plants At any moment Load before and after participating in demand response; For loads that can be reduced, The proportion of load that can be reduced, , These are conventional electrical load and thermal load, respectively.

3. The method according to claim 1, characterized in that, The carbon cost of each virtual power plant is calculated using the following formula: ； ； ； in, It represents the penalty price for the carbon emissions of a virtual power plant at each stage. This is the actual carbon emissions of the virtual power plant. It is the carbon emission allocation of the virtual power plant. It is the length of the carbon step interval. It is the carbon emission distribution factor of the gas turbine. , These are the electrical power and thermal power output by the gas turbine, respectively. This refers to the electricity purchased by the virtual power plant from the electricity market. It is the carbon emission allocation factor that the virtual power plant purchases from the electricity market. For gas turbine The amount, Capture for carbon capture and storage devices Energy consumption This represents the carbon emission coefficient corresponding to the virtual power plant's purchase of electricity in the electricity market; The revenue function of the virtual power plant is: ； ； ； ； ； ； ； ； ； in, To find the maximum value function, It is a virtual power plant Revenue from selling electrical and thermal power to users It is a virtual power plant With virtual power plants Energy trading revenue between them It is the cost of electricity purchased by the virtual power plant from the grid. It is the cost for the virtual power plant to purchase natural gas from the natural gas pipeline network. It is the cost of purchasing heat power through a heating network. This refers to the cost for the virtual power plant to purchase hydrogen from the hydrogen network. It is the cost of various devices in the virtual power plant. It is the fee paid to the distributed network when virtual power plants trade energy. It is the carbon cost of a virtual power plant; and It is a virtual power plant at all times The electrical and thermal power sold to users; and These are the energy prices for electrical power and thermal power, respectively. and It is a virtual power plant With virtual power plants The price of electrical and thermal power transactions between them; and It refers to the electrical and thermal power traded between virtual power plants; and It is the price at which a virtual power plant buys and sells electricity from the grid; and It is a virtual power plant Electricity purchased and sold from the power grid; It is the price at which the virtual power plant purchases natural gas from the natural gas pipeline network; This is the amount of natural gas that the virtual power plant purchases from the natural gas pipeline network; It is the price at which the virtual power plant purchases hydrogen from the hydrogen network; It is a virtual power plant The amount of hydrogen purchased from the hydrogen network; and It refers to the price of buying and selling heat power through a heat network; and It is a virtual power plant Heat power purchased and sold from the heating network; These are the cost coefficients of various devices in a virtual power plant; and It is the price coefficient for electricity and heat paid to the distributed network; , These represent the natural gas and hydrogen input quantities for the gas turbine, respectively. This refers to the thermal power output of the gas turbine. This refers to the number of virtual power plants traded. It refers to a time period; The revenue function for the load aggregator is: ； ； in, It's about virtual power plants. The user's energy use utility function; It is the benefit of user participation in demand response; It is the cost for users to purchase energy from virtual power plants. , It is the demand response price. This is a load that can be reduced.

4. The method according to claim 3, characterized in that, The user's energy consumption utility function is: ； in, , , , Indicating regarding virtual power plants The weighting coefficient of energy use utility , Virtual power plants At any moment The conventional electrical and thermal loads.

5. The method according to claim 4, characterized in that, The energy balance and trading constraints include: Power balance constraints: ； in, For the output of photovoltaic generators; For the output of wind turbines; The electrical output of the gas turbine; For the discharge output of the energy storage system; Thermal power balance constraint: ； in, The thermal power output of the gas turbine. The natural gas balance constraint: ； in, This refers to the amount of natural gas input to the gas turbine. The hydrogen balance constraint: ； in, This refers to the amount of hydrogen input to the gas turbine. Energy trading constraints: ； ； ； in, , Virtual power plants With virtual power plants The maximum trading volume of electrical and thermal power between them; , Virtual power plants With virtual power plants The volume of electrical and thermal power exchanged between them.

6. The method according to any one of claims 1 to 5, characterized in that, The C-ADMM algorithm is used to iteratively solve the non-cooperative game model under the conditions of energy balance and trading constraints to achieve Nash equilibrium, including: The multiple virtual power plants and the load aggregator are defined as participants in a non-cooperative game, wherein each of the multiple virtual power plants participates in energy trading as an independent whole, and the load aggregator participates in energy trading and demand response management as a user-side agent. The energy trading relationship between the multiple virtual power plants and the load aggregator is constructed as a non-cooperative game model to describe the competition and cooperation among the market participants: ； in, Representing a game theory model, For the set of game participants, For virtual power plants Strategy set, For a set of load aggregators, , These are the revenue functions for virtual power plants and load aggregators, respectively. Define virtual power plant (VPS) strategies and load aggregator strategies, wherein the VPS strategy includes energy equipment output and trading capabilities, and the load aggregator strategy includes demand response and trading capabilities with VPS. ； ； in, , These refer to the natural gas and hydrogen input rates for the gas turbine, respectively. Captured by carbon capture and storage devices quantity, , These refer to the charging power and discharging power of the energy storage system, respectively. For demand response load, , These are respectively electrical power trading volume and thermal power trading volume; Using the C-ADMM algorithm, each virtual power plant and load aggregator continuously adjusts its own strategy, and revenue is calculated using the revenue functions of virtual power plants and load aggregators. When all participants no longer adjust their strategies, the energy trading market reaches Nash equilibrium. ； in, , Virtual power plants The optimal set of strategies for load aggregators under Nash equilibrium.

7. The method according to claim 6, characterized in that, Using the C-ADMM algorithm, each virtual power plant and load aggregator continuously adjusts its own strategy, and revenue is calculated using the revenue functions of virtual power plants and load aggregators. When all participants no longer adjust their strategies, the energy trading market reaches Nash equilibrium, including: The C-ADMM algorithm is used to solve the non-cooperative game model of multiple virtual power plants and load aggregators in a distributed manner. The dual variable is updated by the average value of the variable, and energy trading consensus is achieved without introducing auxiliary variables. The energy trading problem is decomposed into a sub-problem of updating trading power and a sub-problem of updating trading price. The virtual power plant adjusts its internal energy output and trading power based on the current trading price to achieve power decision updates. The objective function for updating the trading power in this power decision update is: ； in, Update the objective function for trading power. , Virtual power plants With virtual power plants Electricity and heat transactions between them at any time The corresponding dual variable, and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and It is a virtual power plant With virtual power plants The electrical and thermal power exchanged between them and These are the energy prices for electrical power and thermal power, respectively. For penalty parameters, The overall operating cost of a virtual power plant. This refers to the number of virtual power plants traded. It refers to a time period; Based on the updated power trading results, the dual variables corresponding to the power trading price and the heat trading price are updated: ； ； in, , These are the updated dual variables for electricity and heat trading, respectively. The convergence of the C-ADMM algorithm is determined by the residual value. When the residual value of market participants is less than a preset threshold after iterative updates, the C-ADMM algorithm is considered to have converged and reached a global Nash equilibrium.

8. A non-cooperative game optimization device for multiple virtual power plants and load aggregators, characterized in that, The device includes: The energy trading system construction module is used to: establish an energy trading system consisting of multiple virtual power plants and load aggregators; The internal energy unit modeling module is used to: model the internal energy units of each virtual power plant in the energy trading system to obtain the state and / or constraint models of the internal energy units; The revenue function construction module is used to: adopt a tiered carbon pricing mechanism to incorporate carbon emission costs into the revenue function of virtual power plants, and combine the state and / or constraint models of gas turbines, hydrogen co-firing and electricity purchase, as well as the internal energy units, to calculate the carbon cost of each virtual power plant and construct the revenue function of virtual power plants based on the carbon cost; and to establish the revenue function of load aggregator with the goal of maximizing comprehensive benefits by obtaining revenue through demand response. The Nash equilibrium calculation module is used to: construct a non-cooperative game model based on the revenue function of the virtual power plant and the revenue function of the load aggregator, the strategy of the virtual power plant and the strategy of the load aggregator, and use the C-ADMM algorithm to iteratively solve the non-cooperative game model under the conditions of satisfying energy balance and transaction constraints to achieve Nash equilibrium.

9. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images