Game decision of load aggregator participating in electric-carbon market and energy storage optimal dispatching method

Abstract

The application provides a game decision and energy storage optimization scheduling method for load aggregator participating in the electricity-carbon market, constructs a non-cooperative game model with the maximum comprehensive income of the load aggregator as the target, wherein the income covers the electricity market income and the carbon emission reduction income; designs a dynamic subsidy price mechanism based on the total response quantity to guide the load aggregator to respond in the key period of the system; unifies the optimization of the flexible load response strategy of the load aggregator and the charging and discharging plan of the self-built energy storage system in the game model, and considers the constraint conditions such as the user satisfaction; and solves the non-cooperative game model by using the improved particle swarm optimization algorithm to obtain the optimal scheduling strategy of each load aggregator. The application realizes the independent optimization and collaborative scheduling of the load aggregator resources, effectively improves the economy of the power system operation, promotes the new energy consumption and reduces the carbon emission of the system.

Description

Technical Field

[0001] This invention relates to the field of electricity market and demand response technology, and in particular to a method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market. Background Technology

[0002] With the deepening of power market reform and the advancement of the "dual carbon" strategic goal, load aggregators, as key players in aggregating distributed and flexible resources on the user side, are receiving widespread attention and praise from the industry for their role in improving the operating efficiency of the power system and promoting the consumption of clean energy.

[0003] Currently, load aggregators face many challenges in participating in market operations: traditional demand response often adopts a fixed subsidy price mechanism, which makes it difficult to accurately reflect the marginal cost of real-time system operation; at the same time, there is a lack of an effective collaborative decision-making framework among multiple load aggregators, and their decentralized response strategies may lead to internal resource consumption, failing to achieve the optimal goals of overall system economy and low carbon emissions.

[0004] To guide load aggregators in optimizing resource allocation, the traditional solution relies on unified dispatching through instructions from distribution operators. However, this approach struggles to adapt to the differentiated operational constraints and interests of various aggregators. With the increasing application of non-cooperative game theory in power system dispatching, its ability to effectively characterize the strategic interactions among multiple stakeholders offers a new approach to solving these problems.

[0005] Designing effective game theory mechanisms and dynamic price signals to incentivize load aggregators to achieve multiple objectives—peak shaving, valley filling, increasing renewable energy absorption, and reducing system carbon emissions—while pursuing their own interests has become a key technical problem that needs to be solved. Summary of the Invention

[0006] The main objective of this invention is to provide a method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the carbon electricity market, thereby solving the problem that visibility meters require manual cleaning, which is inconvenient and easily scratches the lens surface.

[0007] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market, the method comprising: S1. Construct a game-theoretic decision-making model for the electric carbon market that includes multi-load aggregators and energy storage systems; S2. Taking the maximization of the comprehensive revenue of each load aggregator as the objective function, and combining the constraints of the power distribution network, user electricity satisfaction, and energy storage system operation, a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions is constructed. S3. The non-cooperative game optimization model is transformed into an equilibrium problem, and an improved particle swarm optimization algorithm is used to solve it, so as to obtain the optimal load response strategy and energy storage scheduling plan for each load aggregator. S4. Based on the optimal strategy obtained from the solution, guide each load aggregator to adjust the operating status of its flexible load and energy storage system in real time, so as to achieve the goal of economic and low-carbon operation of the power system.

[0008] The preferred scheme also includes a dynamic subsidy price mechanism, which includes response quantity subsidy coefficient constraints and constant quantity subsidy coefficient constraints. The response compensation price provided by the power grid depends on the overall load reduction level of all load aggregators. The final revenue of an individual load aggregator depends not only on its own load reduction amount, but also on the load reduction strategies of other load aggregators, as expressed as: ; in, It is the compensation price for the power grid's response to load aggregators; It represents the total load reduction of each load aggregator at different times; This represents the active power of the distribution network load during time period t; This is the price load factor; For constant terms; parameter and All values ​​are greater than 0, and the data are obtained by fitting electricity price data under different load scenarios using the least squares method. To and Since there are no irrelevant constant loads, the formula can be transformed into: ; in, , meaning the response quantity subsidy coefficient; , meaning constant subsidy coefficient; As the load response increases, the unit price of response compensation will decrease accordingly. Once the response power and compensation price of each load aggregator are determined, the load aggregator can obtain the response revenue provided by the power grid during the time period.

[0009] In the preferred scheme, the demand response constraints include transferable load constraints and interruptible load constraints, specifically expressed as follows: ; ; ; in, This represents the active power of the load at node i during time period t; Let be the transferable load response at node i during time period t; Let be the interruptible load response at node i during time period t; The maximum response coefficient for transferable loads. The maximum response coefficient for interruptible loads; The total number of load nodes participating in demand response. This represents the total number of time periods within the scheduling cycle.

[0010] In the preferred scheme, the operational constraints of the energy storage system include active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints, specifically expressed as follows: ; ; ; ; ; ; ; ; in, This represents the discharge power of the energy stored at node i during time period t; This represents the charging power of the energy stored at node i during time period t; express Nodes within the time period The active power of the energy storage device; This indicates the maximum allowable charging and discharging power of the energy storage system; and These are 0-1 variables, representing the charging and discharging flags for energy storage. express Nodes within the time period The reactive power of energy storage; This represents the maximum value of the energy storage reactive power; This indicates the rated apparent power of the energy storage. express State of charge, which stores energy at all times; Indicates the rated capacity of energy storage; and These represent the charging and discharging efficiency of energy storage, respectively. and These represent the maximum and minimum values ​​of the state of charge, respectively. In the preferred scheme, the user satisfaction constraint is used to ensure that users within the load aggregator's region maintain a reasonable level of electricity satisfaction when responding to load transfer under time-of-use pricing incentives. The specific expression is as follows: ; ; ; in, To improve user satisfaction with electricity usage; The minimum value for user satisfaction with electricity usage; The impact of subsidized electricity prices on user satisfaction with electricity usage in response to user demand; The maximum impact of demand-response electricity price subsidies on user electricity satisfaction; and The minimum and maximum subsidized electricity prices that affect user satisfaction with electricity use. Let be the transferable load response at node i during time period t; D represents the active power of the load at node i during time period t; D is the total number of load nodes participating in demand response.

[0011] In the preferred scheme, the detailed steps in step S2 for constructing a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions include: S21. Determine the composition of the comprehensive revenue of each load aggregator. The objective function is to maximize the comprehensive revenue of each load aggregator. The comprehensive revenue includes demand response revenue, energy storage revenue, and carbon emission reduction revenue. After deducting demand response costs and energy storage operating costs, the objective function is expressed as follows: ; in, It is the demand response compensation cost of the power grid to load aggregators. This indicates the response costs that the load aggregator provides to users for different types of load. This represents the benefits that the power grid receives from charging and discharging through the energy storage system. This indicates the daily operating cost of the energy storage system. This indicates the regional carbon emission reduction benefits after the response; Demand Response Benefits The formula for calculation is: ; in, It is the compensation price for the power grid's response to load aggregators. It represents the total load reduction of each load aggregator at different times. Indicates the unit simulation time step. This represents the total number of time periods within the scheduling cycle. Response compensation price The initial calculation formula is: ; ; ; ; in, This represents the active power of the distribution network load during time period t. This is the price load factor. For constant terms, This represents the active power of the load at node i during time period t. This represents the load reduction at node i during time period t. For interruptible load response quantities, This refers to the transferable load response. parameter and All values ​​are greater than 0, and the data are fitted using the least squares method from electricity price data under different load scenarios. To and Irrelevant load constant quantity, therefore response compensation price It can be converted into: ; in, For response quantity subsidy coefficient, It is a constant subsidy coefficient, and the unit price of response compensation decreases as the load response increases; when the response power of each load aggregator... and compensation price Once determined, the load aggregator can obtain the response revenue provided by the grid during time period t. ; Demand response cost The formula for calculation is: ; in, The response subsidy price for transferable load, The response subsidy price for interruptible loads; Energy storage revenue The formula for calculation is: ; in, This represents the electricity price sold by the power grid during time period t. This represents the discharge power of the energy stored at node i during time period t. This represents the charging power of the energy stored at node i during time period t; Energy storage operating costs The formula for calculation is: ; in, This represents the discount rate for energy storage. Indicates the service life of energy storage. Indicates the configuration of an energy storage set. This represents the investment cost per unit capacity of energy storage. This indicates the energy storage capacity configuration at node i. This represents the investment cost per unit power of energy storage. This indicates the power configuration for energy storage at node i; carbon emission reduction benefits The formula for calculation is: ; ; ; ; in, This indicates the proportion of regional carbon reduction benefits received by load aggregators. For carbon trading prices, Carbon emissions before and after demand response. This represents the carbon emission intensity of node i during time period t. Let i be the actual active power of renewable energy at node i during time period t. This represents the renewable energy carbon emission intensity of node i, set to zero. Let i represent the set of all branches flowing into node i. This represents the net active power of branch ij during time period t, after deducting line losses. This represents the carbon flux density of branch ij during time period t. This represents the active power of line ij during time period t; S22. Set distribution network constraints, which include active and reactive power balance constraints at nodes, voltage variation constraints, line current calculation relationship constraints, and voltage amplitude and line current constraints, respectively expressed as follows: ; ; ; ; ; ; in, It represents the set of all lines in the distribution network. Represents a node Flow to Node The route, , They represent the lines respectively. Resistance and reactance, express Time-of-use energy storage system at the access node Active power at the location, express Time-of-day routes Current magnitude, express Renewable energy at nodes reactive power, express Time period nodes The reactive power of the load, express Time-of-use energy storage at nodes reactive power, express Time-of-day routes reactive power on , They represent Time period nodes , Voltage at that point and These represent the upper and lower limits of the node voltage, respectively. Indicates the line The maximum current; S23. Integrate the objective function with the distribution network constraints, and combine dynamic subsidy price constraints, demand response constraints, user electricity satisfaction constraints, and energy storage system operation constraints to form a complete multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon emissions.

[0012] In the preferred scheme, the daily operating cost of the energy storage system in step S21 is... The calculation method is as follows: ; Where r represents the discount rate for energy storage; y represents the lifespan of the energy storage. This indicates the investment cost per unit capacity of energy storage; This indicates the energy storage capacity configuration at node i; This represents the investment cost per unit power of energy storage; This indicates the power configuration of energy storage at node i; Carbon emissions before and after demand response in carbon reduction benefits The calculation method is as follows: ; ; ; in, , These represent the carbon emissions of node i before and after the demand response in time period t, respectively. This represents the carbon emission intensity of node i during time period t; This represents the carbon flux density of branch ij during time period t; This represents the net active power of branch ij during time period t, after deducting line losses. The renewable energy carbon emission intensity of node i is set to zero; This represents the actual active power of renewable energy at node i during time period t; Indicates the unit simulation time step; This represents the set of all branches flowing into node i.

[0013] In the preferred embodiment, the algorithm steps in step S3 using the improved particle swarm optimization algorithm include: S31. Given the network topology of the distribution network, the load data of each node, the predicted output of renewable energy, and the relevant parameters of electricity price and carbon price, set the parameters of the improved particle swarm algorithm, including the inertia weight and its change rules, the initial value of the learning factor, and initialize the position and velocity of the particle swarm. S32. Solve the objective function of each load aggregator, and pass the demand response power, energy storage configuration and scheduling strategy information in the current solution that satisfies the demand of the load aggregator to the particle swarm, and calculate the fitness value of each particle. S33. Compare the current fitness value of each particle with its individual historical best value and the global historical best value. If the current fitness value is better, update the corresponding individual historical best value or the global historical best value and its corresponding strategy; otherwise, retain the historical best value. At the same time, dynamically adjust the inertia weight and learning factor according to the iteration process. S34. Based on the updated parameters and optimal values, update the position and velocity of the particle swarm, and ensure that the new position is feasible in the solution space; S35. Determine whether the solution meets the convergence condition, i.e., the maximum number of iterations has been reached or the quality of the solution is stable. If it is met, output the optimal result; otherwise, return to step S32 to continue iterating.

[0014] In the preferred scheme, the algorithm steps for obtaining the optimal load response strategy and energy storage scheduling plan for each load aggregator in step S3 include: S36. Based on the optimal results output in step S35, extract the total load reduction of each load aggregator at different time periods. Combined with the dynamic subsidy price formula Determine the response compensation price for each time period. ; S37. Based on demand response constraints, allocate the transferable load response amount of each load aggregator at different time periods. and interruptible load response This forms an optimal load response strategy; wherein the demand response constraints are consistent with those in claim 3. S38. Based on the operating constraints of the energy storage system, determine the charging power of the energy storage system in different time periods within the jurisdiction of each load aggregator. Discharge power and state of charge The optimal energy storage scheduling plan is generated; wherein the energy storage system operation constraints are consistent with those in claim 4. S39. Verify whether the optimal load response strategy and energy storage dispatch plan meet the distribution network constraints and user electricity satisfaction constraints. If they meet the constraints, they are determined as the final optimal load response strategy and energy storage dispatch plan. If they do not meet the constraints, return to step S32 for iterative adjustment. Among them, the power distribution network constraint in step S22 is consistent with the power distribution network constraint, and the user electricity satisfaction constraint is consistent with the above-mentioned user satisfaction constraint.

[0015] In the preferred scheme, step S4, which guides each load aggregator to adjust the operating status of its flexible loads and energy storage systems in real time based on the optimal strategy obtained from the solution, to achieve the goals of economic and low-carbon operation of the power system, includes the following steps: S41. Establish a communication link between the optimal strategy and the load aggregator control terminal to transmit the transferable load response for each time period. Interruptible load response and energy storage charging and discharging power , The instruction is sent to the corresponding control terminal; S42. The load aggregator control terminal, based on the received instructions, guides users to transfer transferable loads to off-peak hours during peak electricity consumption periods, disconnects some interruptible loads, and controls the energy storage system to discharge to supplement the power supply gap in the grid; during off-peak hours, it controls the energy storage system to charge and store electrical energy. S43. Real-time monitoring of node voltages in the distribution network. Line current and renewable energy output If there are situations such as node voltage exceeding limits, line current exceeding limits, or excessive fluctuations in renewable energy output, the amount of flexible load transfer and energy storage charging and discharging power will be dynamically adjusted based on the optimal strategy. S44. Calculate the daily operating data of the power system, including the revenue of each load aggregator, the total carbon emissions of the system and the carbon emission reduction revenue, and assess the degree of achievement of the economic and low-carbon operation goals of the power system. If the preset goals are not achieved, return to step S3 to re-optimize the game model and solution parameters until the economic and low-carbon operation goals of the power system are achieved.

[0016] This invention provides a game-theoretic decision-making and energy storage optimization scheduling method for load aggregators participating in the electricity carbon market. By constructing a non-cooperative game model of multiple load aggregators and introducing a dynamic subsidy pricing mechanism, it effectively improves the economic efficiency and low-carbon nature of power system operation. This method guides load aggregators to proactively reduce load and optimize energy storage scheduling during peak hours, significantly reducing peak load and achieving peak shaving and valley filling. Incorporating carbon emission reduction benefits into the aggregators' decision-making objectives incentivizes them to shift load to periods of high renewable energy generation, thereby increasing renewable energy absorption and reducing overall system carbon emissions. The improved particle swarm optimization algorithm used can efficiently solve complex game equilibria, ensuring rapid generation of scheduling strategies and stable system operation. Attached Figure Description

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments. Figure 1 This is the overall flowchart of the present invention; Figure 2 This is a graph showing the total load change before and after the demand response of this invention; Figure 3 This is a diagram of the load reduction plans for each aggregator in this invention; Figure 4 This is a diagram of the energy storage operation strategy for load aggregator 1 of the present invention; Figure 5 This is a diagram of the energy storage operation strategy for load aggregator 2 scheduling according to the present invention; Figure 6 This is a diagram of the energy storage operation strategy for load aggregator 3 scheduling in this invention. Detailed Implementation

[0018] Example 1 like Figure 1-6 As shown, a game-theoretic decision-making and energy storage optimization scheduling method for load aggregators participating in the electricity carbon market is presented. The method includes: S1. Construct a game-theoretic decision-making model for the electric carbon market that includes multi-load aggregators and energy storage systems; S2. Taking the maximization of the comprehensive revenue of each load aggregator as the objective function, and combining the constraints of the power distribution network, user electricity satisfaction, and energy storage system operation, a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions is constructed. S3. The non-cooperative game optimization model is transformed into an equilibrium problem, and an improved particle swarm optimization algorithm is used to solve it, so as to obtain the optimal load response strategy and energy storage scheduling plan for each load aggregator. S4. Based on the optimal strategy obtained from the solution, guide each load aggregator to adjust the operating status of its flexible load and energy storage system in real time, so as to achieve the goal of economic and low-carbon operation of the power system.

[0019] The preferred scheme also includes a dynamic subsidy price mechanism, which includes response quantity subsidy coefficient constraints and constant quantity subsidy coefficient constraints. The response compensation price provided by the power grid depends on the overall load reduction level of all load aggregators. The final revenue of an individual load aggregator depends not only on its own load reduction amount, but also on the load reduction strategies of other load aggregators, as expressed as: ; in, It is the compensation price for the power grid's response to load aggregators; It represents the total load reduction of each load aggregator at different times; This represents the active power of the distribution network load during time period t; This is the price load factor; For constant terms; parameter and All values ​​are greater than 0, and the data are obtained by fitting electricity price data under different load scenarios using the least squares method. To and Since there are no irrelevant constant loads, the formula can be transformed into: ; in, , meaning the response quantity subsidy coefficient; , meaning constant subsidy coefficient; As the load response increases, the unit price of response compensation will decrease accordingly. Once the response power and compensation price of each load aggregator are determined, the load aggregator can obtain the response revenue provided by the power grid during the time period.

[0020] In the preferred scheme, the demand response constraints include transferable load constraints and interruptible load constraints, specifically expressed as follows: ; ; ; in, This represents the active power of the load at node i during time period t; Let be the transferable load response at node i during time period t; Let be the interruptible load response at node i during time period t; The maximum response coefficient for transferable loads. The maximum response coefficient for interruptible loads; The total number of load nodes participating in demand response. This represents the total number of time periods within the scheduling cycle.

[0021] In the preferred scheme, the operational constraints of the energy storage system include active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints, specifically expressed as follows: ; ; ; ; ; ; ; ; in, This represents the discharge power of the energy stored at node i during time period t; This represents the charging power of the energy stored at node i during time period t; express Nodes within the time period The active power of the energy storage device; This indicates the maximum allowable charging and discharging power of the energy storage system; and These are 0-1 variables, representing the charging and discharging flags for energy storage. express Nodes within the time period The reactive power of energy storage; This represents the maximum value of the energy storage reactive power; This indicates the rated apparent power of the energy storage. express State of charge, which stores energy at all times; Indicates the rated capacity of energy storage; and These represent the charging and discharging efficiency of energy storage, respectively. and These represent the maximum and minimum values ​​of the state of charge, respectively. In the preferred scheme, the user satisfaction constraint is used to ensure that users within the load aggregator's region maintain a reasonable level of electricity satisfaction when responding to load transfer under time-of-use pricing incentives. The specific expression is as follows: ; ; ; in, To improve user satisfaction with electricity usage; The minimum value for user satisfaction with electricity usage; The impact of subsidized electricity prices on user satisfaction with electricity usage in response to user demand; The maximum impact of demand-response electricity price subsidies on user electricity satisfaction; and The minimum and maximum subsidized electricity prices that affect user satisfaction with electricity use. Let be the transferable load response at node i during time period t; D represents the active power of the load at node i during time period t; D is the total number of load nodes participating in demand response.

[0022] In the preferred scheme, the detailed steps in step S2 for constructing a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions include: S21. Determine the composition of the comprehensive revenue of each load aggregator. The objective function is to maximize the comprehensive revenue of each load aggregator. The comprehensive revenue includes demand response revenue, energy storage revenue, and carbon emission reduction revenue. After deducting demand response costs and energy storage operating costs, the objective function is expressed as follows: ; in, It is the demand response compensation cost of the power grid to load aggregators. This indicates the response costs that the load aggregator provides to users for different types of load. This represents the benefits that the power grid receives from charging and discharging through the energy storage system. This indicates the daily operating cost of the energy storage system. This indicates the regional carbon emission reduction benefits after the response; Demand Response Benefits The formula for calculation is: ; in, It is the compensation price for the power grid's response to load aggregators. It represents the total load reduction of each load aggregator at different times. Indicates the unit simulation time step. This represents the total number of time periods within the scheduling cycle. Response compensation price The initial calculation formula is: ; ; ; ; in, This represents the active power of the distribution network load during time period t. This is the price load factor. For constant terms, This represents the active power of the load at node i during time period t. This represents the load reduction at node i during time period t. For interruptible load response quantities, This refers to the transferable load response. parameter and All values ​​are greater than 0, and the data are fitted using the least squares method from electricity price data under different load scenarios. To and Irrelevant load constant quantity, therefore response compensation price It can be converted into: ; in, For response quantity subsidy coefficient, It is a constant subsidy coefficient, and the unit price of response compensation decreases as the load response increases; when the response power of each load aggregator... and compensation price Once determined, the load aggregator can obtain the response revenue provided by the grid during time period t. ; Demand response cost The formula for calculation is: ; in, The response subsidy price for transferable load, The response subsidy price for interruptible loads; Energy storage revenue The formula for calculation is: ; in, This represents the electricity price sold by the power grid during time period t. This represents the discharge power of the energy stored at node i during time period t. This represents the charging power of the energy stored at node i during time period t; Energy storage operating costs The formula for calculation is: ; in, This represents the discount rate for energy storage. Indicates the service life of energy storage. Indicates the configuration of an energy storage set. This represents the investment cost per unit capacity of energy storage. This indicates the energy storage capacity configuration at node i. This represents the investment cost per unit power of energy storage. This indicates the power configuration for energy storage at node i; carbon emission reduction benefits The formula for calculation is: ; ; ; ; in, This indicates the proportion of regional carbon reduction benefits received by load aggregators. For carbon trading prices, Carbon emissions before and after demand response. This represents the carbon emission intensity of node i during time period t. Let i be the actual active power of renewable energy at node i during time period t. This represents the renewable energy carbon emission intensity of node i, set to zero. Let i represent the set of all branches flowing into node i. This represents the net active power of branch ij during time period t, after deducting line losses. This represents the carbon flux density of branch ij during time period t. This represents the active power of line ij during time period t; S22. Set distribution network constraints, which include active and reactive power balance constraints at nodes, voltage variation constraints, line current calculation relationship constraints, and voltage amplitude and line current constraints, respectively expressed as follows: ; ; ; ; ; ; in, It represents the set of all lines in the distribution network. Represents a node Flow to Node The route, , They represent the lines respectively. Resistance and reactance, express Time-of-use energy storage system at the access node Active power at the location, express Time-of-day routes Current magnitude, express Renewable energy at nodes reactive power, express Time period nodes The reactive power of the load, express Time-of-use energy storage at nodes reactive power, express Time-of-day routes reactive power on , They represent Time period nodes , Voltage at that point and These represent the upper and lower limits of the node voltage, respectively. Indicates the line The maximum current; S23. Integrate the objective function with the distribution network constraints, and combine dynamic subsidy price constraints, demand response constraints, user electricity satisfaction constraints, and energy storage system operation constraints to form a complete multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon emissions.

[0023] In the preferred scheme, the daily operating cost of the energy storage system in step S21 is... The calculation method is as follows: ; Where r represents the discount rate for energy storage; y represents the lifespan of the energy storage. This indicates the investment cost per unit capacity of energy storage; This indicates the energy storage capacity configuration at node i; This represents the investment cost per unit power of energy storage; This indicates the power configuration of energy storage at node i; Carbon emissions before and after demand response in carbon reduction benefits The calculation method is as follows: ; ; ; in, , These represent the carbon emissions of node i before and after the demand response in time period t, respectively. This represents the carbon emission intensity of node i during time period t; This represents the carbon flux density of branch ij during time period t; This represents the net active power of branch ij during time period t, after deducting line losses. The renewable energy carbon emission intensity of node i is set to zero; This represents the actual active power of renewable energy at node i during time period t; Indicates the unit simulation time step; This represents the set of all branches flowing into node i.

[0024] In the preferred embodiment, the algorithm steps in step S3 using the improved particle swarm optimization algorithm include: S31. Given the network topology of the distribution network, the load data of each node, the predicted output of renewable energy, and the relevant parameters of electricity price and carbon price, set the parameters of the improved particle swarm algorithm, including the inertia weight and its change rules, the initial value of the learning factor, and initialize the position and velocity of the particle swarm. S32. Solve the objective function of each load aggregator, and pass the demand response power, energy storage configuration and scheduling strategy information in the current solution that satisfies the demand of the load aggregator to the particle swarm, and calculate the fitness value of each particle. S33. Compare the current fitness value of each particle with its individual historical best value and the global historical best value. If the current fitness value is better, update the corresponding individual historical best value or the global historical best value and its corresponding strategy; otherwise, retain the historical best value. At the same time, dynamically adjust the inertia weight and learning factor according to the iteration process. S34. Based on the updated parameters and optimal values, update the position and velocity of the particle swarm, and ensure that the new position is feasible in the solution space; S35. Determine whether the solution meets the convergence condition, i.e., the maximum number of iterations has been reached or the quality of the solution is stable. If it is met, output the optimal result; otherwise, return to step S32 to continue iterating.

[0025] In the preferred scheme, the algorithm steps for obtaining the optimal load response strategy and energy storage scheduling plan for each load aggregator in step S3 include: S36. Based on the optimal results output in step S35, extract the total load reduction of each load aggregator at different time periods. Combined with the dynamic subsidy price formula Determine the response compensation price for each time period. ; S37. Based on demand response constraints, allocate the transferable load response amount of each load aggregator at different time periods. and interruptible load response This forms an optimal load response strategy; wherein the demand response constraints are consistent with those in claim 3. S38. Based on the operating constraints of the energy storage system, determine the charging power of the energy storage system in different time periods within the jurisdiction of each load aggregator. Discharge power and state of charge The optimal energy storage scheduling plan is generated; wherein the energy storage system operation constraints are consistent with those in claim 4. S39. Verify whether the optimal load response strategy and energy storage dispatch plan meet the distribution network constraints and user electricity satisfaction constraints. If they meet the constraints, they are determined as the final optimal load response strategy and energy storage dispatch plan. If they do not meet the constraints, return to step S32 for iterative adjustment. Among them, the power distribution network constraint in step S22 is consistent with the power distribution network constraint, and the user electricity satisfaction constraint is consistent with the above-mentioned user satisfaction constraint.

[0026] In the preferred scheme, step S4, which guides each load aggregator to adjust the operating status of its flexible loads and energy storage systems in real time based on the optimal strategy obtained from the solution, to achieve the goals of economic and low-carbon operation of the power system, includes the following steps: S41. Establish a communication link between the optimal strategy and the load aggregator control terminal to transmit the transferable load response for each time period. Interruptible load response and energy storage charging and discharging power , The instruction is sent to the corresponding control terminal; S42. The load aggregator control terminal, based on the received instructions, guides users to transfer transferable loads to off-peak hours during peak electricity consumption periods, disconnects some interruptible loads, and controls the energy storage system to discharge to supplement the power supply gap in the grid; during off-peak hours, it controls the energy storage system to charge and store electrical energy. S43. Real-time monitoring of node voltages in the distribution network. Line current and renewable energy output If there are situations such as node voltage exceeding limits, line current exceeding limits, or excessive fluctuations in renewable energy output, the amount of flexible load transfer and energy storage charging and discharging power will be dynamically adjusted based on the optimal strategy. S44. Calculate the daily operating data of the power system, including the revenue of each load aggregator, the total carbon emissions of the system and the carbon emission reduction revenue, and assess the degree of achievement of the economic and low-carbon operation goals of the power system. If the preset goals are not achieved, return to step S3 to re-optimize the game model and solution parameters until the economic and low-carbon operation goals of the power system are achieved.

[0027] Example 2 Further explanation in conjunction with Example 1, referring to Figure 1 As an embodiment of the present invention, a method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market is provided, comprising: S1: Construct a game-theoretic decision-making model for the electric carbon market that includes multi-load aggregators and energy storage systems.

[0028] S2: Taking the maximization of the comprehensive revenue of each load aggregator as the objective function, and combining the constraints of the power distribution network, user satisfaction and energy storage system operation, a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions is constructed.

[0029] Load aggregators (LAs) earn revenue by scheduling users' flexible load resources for demand response. They provide response costs to users based on different load types. Furthermore, load aggregators autonomously allocate energy storage capacity and control their operating strategies within their jurisdiction; therefore, the investment costs and related benefits of energy storage must be considered. Their objective function is to maximize response revenue, specifically including demand response revenue, demand response costs, energy storage revenue, energy storage operating costs, and carbon emission reduction benefits, expressed as: ; ; ; ; ; ; ; ; ; ; ; ; ; in, It is the demand response compensation cost of the power grid to load aggregators; This indicates the response costs that the load aggregator provides to users for different types of load; This indicates the benefits the power grid receives from charging and discharging through energy storage systems; This indicates the daily operating cost of the energy storage system; This indicates the regional carbon emission reduction benefits after the response; Indicates the unit simulation time step; express t Time period nodes i The active power of the load; express t Time period nodes i The amount of load reduction includes interruptible load and positive transferable load; It is the compensation price for the power grid's response to load aggregators; It represents the total load reduction of each load aggregator at different times; express t Active power of the distribution network load during a given time period; This is the price load factor; For constant terms; The response subsidy price for transferable loads; The response subsidy price for interruptible loads; This refers to the transferable load response. This refers to the interruptible load response. express t Electricity prices on the grid during specific time periods; This represents the discharge power of the energy stored at node i during time period t; represents the charging power of energy storage at node i during time period t; r and y represent the discount rate and term of energy storage, respectively; This indicates the daily investment cost of energy storage; This indicates the investment cost per unit capacity of energy storage; Indicates the configuration of an energy storage collection; This indicates the energy storage capacity configuration at node i; This represents the investment cost per unit power of energy storage; This indicates the power configuration for energy storage at node i; This indicates the proportion of regional carbon reduction revenue received by the load aggregator, with the other portion going to the user; For carbon trading prices; Carbon emissions before and after demand response; Indicates the inflow node i The set of all branches; express t Time period nodes i carbon emission intensity; For nodes i exist t The actual active power of renewable energy during a given time period; Represents a node i The carbon intensity of renewable energy is set to zero to reflect its clean nature; express t Time-of-day branch i - j The net active power has already deducted the line loss portion. express t Time-of-day branch i - j The carbon flux density.

[0030] parameter and All values ​​are greater than 0, and the data are obtained by fitting electricity price data under different load scenarios using the least squares method. To and Since there are no irrelevant constant loads, the formula can be transformed into: ; in: , meaning the response quantity subsidy coefficient; , meaning constant subsidy coefficient.

[0031] As the load response increases, the unit price of response compensation will decrease accordingly, as the response power of each load aggregator... and compensation price Once confirmed, the load aggregator can obtain Response benefits provided by the power grid during the time period .

[0032] The constraints of the model include power flow constraints, response compensation price constraints, demand response constraints, user satisfaction constraints, and energy storage operation constraints.

[0033] Power flow constraints in a distribution network include active and reactive power balance constraints at nodes, voltage variation patterns and line current calculation relationships, and voltage amplitude and line current constraints, expressed as: ; ; ; ; ; ; in, This represents the set of all lines in a power distribution network; Represents a node Flow to Node The route; , They represent the lines respectively. Resistance and reactance; express Time period nodes Active power of renewable energy sources; express t Time period nodes The active power of the load; express Time-of-use energy storage system at the access node Active power at the location; express Time-of-day routes Active power; express Time-of-day routes Current magnitude; express Renewable energy at nodes reactive power; express Time period nodes The reactive power of the load; express Time-of-use energy storage at nodes reactive power; express Time-of-day routes reactive power; express Time period nodes Voltage at the point; and These represent the upper and lower limits of the node voltage, respectively. Indicates the line The maximum current.

[0034] The response compensation price needs to take into account the constraints on the range of the response quantity subsidy coefficient and the constant subsidy coefficient, expressed as follows: ; ; in, and These are the minimum and maximum values ​​of the response quantity subsidy coefficient; and These are the minimum and maximum values ​​of the constant subsidy coefficient.

[0035] Demand response constraints consider upper and lower limits on the response amounts of transferable and interruptible loads, while ensuring that the total amount of transferable load remains unchanged before and after demand response, expressed as: ; ; ; in, The maximum response coefficient for transferable loads. This is the maximum response coefficient for interruptible loads.

[0036] Electricity satisfaction constraints consider that when users respond to load transfer under time-of-use pricing incentives, their satisfaction remains within a reasonable range, expressed as: ; ; ; in, To improve user satisfaction with electricity usage; The minimum value for user satisfaction with electricity usage; The impact of subsidized electricity prices on user satisfaction with electricity usage in response to user demand; The maximum impact of demand-response electricity price subsidies on user electricity satisfaction; and The minimum and maximum values ​​of the subsidized electricity price that affect user satisfaction with electricity use.

[0037] Energy storage operation constraints specifically consider active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints, expressed as: ; ; ; ; ; ; ; ; in, express Nodes within the time period The active power of the energy storage device; This indicates the maximum allowable charging and discharging power of the energy storage system; and These are 0-1 variables, representing the charging and discharging flags for energy storage. This indicates that the storage device cannot be charged and discharged simultaneously; express Nodes within the time period The reactive power of energy storage; This represents the maximum value of the energy storage reactive power; This indicates the rated apparent power of the energy storage. express State of charge, which stores energy at all times; Indicates the rated capacity of energy storage; and These represent the charging and discharging efficiency of energy storage, respectively. and These represent the maximum and minimum values ​​of the state of charge, respectively. S3: The non-cooperative game optimization model is transformed into an equilibrium problem, and an improved particle swarm optimization algorithm is used to solve it, obtaining the optimal load response strategy and energy storage scheduling plan for each load aggregator.

[0038] Because of the game-like relationship among load aggregators, the model constitutes a multi-agent non-cooperative game, making it difficult to directly obtain the equilibrium solution using traditional solvers. Therefore, this paper utilizes an improved particle swarm optimization algorithm to solve for the optimal response and energy storage configuration strategy within the multi-load aggregator game framework until the system converges to a stable equilibrium. The overall flowchart is as follows: Figure 1 As shown.

[0039] The solution steps for this model algorithm are as follows: 1. Given the network topology of the distribution network and the load data of each node, the predicted output of renewable energy and relevant parameters such as electricity price and carbon price, set the parameters of the improved particle swarm algorithm, mainly including the inertia weight and its change rules, the initial value of the learning factor, etc., and initialize the position and velocity of the particle swarm. 2. Solve the objective function of each load aggregator, and pass the information such as the demand response power, energy storage configuration and scheduling strategy in the current solution that satisfies the demand of the load aggregator to the particle swarm, and calculate the fitness value of each particle. 3. Compare the current fitness value of each particle with its individual historical best value and the global historical best value. If the current fitness value is better, update the corresponding individual historical best value or the global historical best value and its corresponding strategy; otherwise, retain the historical best value. Simultaneously, dynamically adjust the inertia weight and learning factor based on the iteration process. 4. Based on the updated parameters and optimal values, update the position and velocity of the particle swarm, and ensure that the new position is feasible in the solution space. 5. Determine whether the solution meets the convergence condition, i.e., the maximum number of iterations has been reached or the quality of the solution is stable. If it is met, output the optimal result; otherwise, return to step 2 to continue iterating. This method can be simulated and verified using the IEEE-33 node distribution system to obtain load data, predicted output values ​​of renewable energy, and data such as time-of-use electricity prices and carbon prices for each node in the distribution network. The system uses 10 MVA as the capacity benchmark and a rated voltage of 12.66 kV. Photovoltaics are connected at nodes 9 and 33, and wind power is connected at nodes 1, 2, 15, and 29, with a power factor set to 0.9. Energy storage is connected at nodes 4, 11, and 33, with its rated power and capacity configuration determined by the load aggregator. The capacity configuration ranges from 0 to 2 MWh, with a charge / discharge efficiency of 0.90 and an initial state of charge (SOC) of 0.35. There are three load aggregators in the distribution system, each managing three different areas. Each area is equipped with an energy storage system, and the load aggregators control the flexible load resources and the operation of the energy storage system within their respective areas.

[0040] To verify the effectiveness of the proposed energy storage optimization scheduling strategy for load aggregators participating in the electricity carbon market, this study designed two schemes to verify their economic and low-carbon benefits. Scheme 1 is a traditional demand response scheduling model based on a fixed price, without a game mechanism. Scheme 2 is a multi-stakeholder game and energy storage collaborative optimization model, allowing load aggregators to make autonomous decisions on energy storage configuration and scheduling.

[0041] Table 1. Comparison of simulation results for the two cases:

[0042] In terms of economics, Case 1 uses a fixed subsidy mechanism, resulting in a passive response from load aggregators, low resource allocation efficiency, and limited aggregator profits. Case 2, building on Case 1, introduces a dynamic pricing mechanism and energy storage self-dispatch capabilities from a non-cooperative game perspective. Aggregators can respond flexibly by combining local resources, improving demand response quality and profitability, demonstrating the economic benefits of multi-entity collaborative optimization.

[0043] Regarding carbon emissions, Case 1's response period was not precise enough, and some load reductions did not cover peak carbon emission periods, resulting in limited emission reduction effects. Case 3, building on Case 1, introduced self-built energy storage and electricity-carbon synergy optimization targets, shifting the load to peak renewable energy output periods, improving renewable energy absorption rates, reducing total carbon emissions, and significantly increasing carbon emission reduction benefits.

[0044] During the simulation, the total load change curves before and after the demand response are as follows: Figure 2 As shown, under the incentives of electricity price and compensation, the demand response successfully shaving off peaks and filling valleys, narrowing the load curve fluctuation range from 5.5-8MWh to 6-7.5MWh, with a significant smoothing effect.

[0045] The load reduction plans of each aggregator are as follows Figure 3 As shown, the load reduction efforts of the three load aggregators are mainly concentrated during peak electricity consumption periods of 9–13 and 19–21. The total reduction varies with load trends, reaching peak values ​​at 10 and 20, effectively achieving peak shaving. During peak periods, load aggregator 1, with its abundant load resources and higher electricity price, bears the majority of the reduction. However, during periods with medium electricity prices, dynamic subsidies guide load aggregators to participate in response rationally, leading to a significant increase in the contribution of load aggregator 2. This mechanism enables autonomous optimization of load scheduling, avoiding excessive resource allocation.

[0046] Table 2 shows the capacity and power configurations of energy storage within the jurisdiction of each load aggregator. The charging and discharging operation strategies of each energy storage system are as follows: Figures 4 to 6 As shown. The energy storage charging and discharging strategies of the three load aggregators are as follows: Figures 4 to 6As shown, the overall charging and discharging trends of the energy storage systems dispatched by each load aggregator are consistent, but their specific charging and discharging periods and amounts are different. This is because each load aggregator, in a state of mutual competition, dispatches the charging and discharging status of the energy storage according to maximizing its own interests.

[0047] Table 2 Energy Storage Configuration:

[0048] Simulation results show that the dynamic subsidy mechanism effectively incentivizes users to participate in the response, while the game theory mechanism coordinates the interests of multiple parties and improves the flexibility, economy, and low carbon emissions of the system operation.

[0049] Example 3 Focusing on the fields of electricity market and demand response technology, this research aims to address the problems faced by load aggregators when participating in market operations, such as the inability of traditional fixed subsidy pricing mechanisms to reflect real-time marginal costs of the system, the lack of a collaborative decision-making framework among multiple stakeholders leading to resource inefficiencies, and the inability to balance system economics and low carbon emissions.

[0050] This method takes maximizing the comprehensive revenue of each load aggregator as its core objective, comprehensively considers the revenue from the electricity market and the carbon emission reduction revenue, and constructs an electricity-carbon market game decision-making model that includes multiple load aggregators and energy storage systems. Furthermore, it forms a multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon. In the process of model construction, not only are distribution network constraints and user electricity satisfaction constraints incorporated, but the operational constraints of the energy storage system are also clearly defined, including active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints, to ensure the feasibility and rationality of the dispatch strategy.

[0051] To guide load aggregators to respond actively during critical periods of the system, the document designs a dynamic subsidy price mechanism based on the total response volume. Under this mechanism, the response compensation price given by the power grid depends on the overall load reduction level of all load aggregators. The revenue of a single load aggregator is affected by its own reduction volume as well as the reduction strategies of other aggregators. Furthermore, as the load response volume increases, the unit price of response compensation will decrease accordingly. At the same time, the definition and calculation method of the response volume subsidy coefficient and the constant subsidy coefficient are clarified through formula transformation. The parameters are all obtained by fitting the electricity price data under different load scenarios using the least squares method.

[0052] In the model solution phase, the document employs an improved particle swarm optimization algorithm, transforming the non-cooperative game optimization model into an equilibrium problem. The solution process requires initializing parameters such as distribution network topology, node loads, predicted renewable energy output, electricity and carbon prices, and initializing algorithm parameters and particle swarm position and velocity. Then, through steps including calculating particle fitness values, updating individual and global optimum values, dynamically adjusting algorithm parameters, and iteratively optimizing position and velocity, the process continues until convergence conditions are met. Finally, the optimal load response strategy and energy storage dispatch plan for each load aggregator are output. Based on the optimal strategy obtained, load aggregators can be guided in real-time to adjust the operating status of flexible loads and energy storage systems. For example, during peak electricity consumption periods, load transfer can be guided, some interruptible loads can be disconnected, and energy storage discharge can be controlled; during off-peak periods, energy storage charging can be controlled. Simultaneously, distribution network operation data is monitored in real-time, and strategies are dynamically adjusted to ensure the achievement of the power system's economic and low-carbon operation goals. From the perspective of application results, this method can effectively improve the economic efficiency of power system operation. It enhances the revenue of load aggregators through multi-stakeholder game and dynamic subsidy mechanism, while promoting the consumption of new energy, reducing system carbon emissions, achieving peak shaving and valley filling, and smoothing the load curve. Moreover, through simulation verification of the IEEE-33 node distribution system, compared with the traditional fixed-price demand response scheduling model, this method shows significant advantages in terms of aggregator revenue, carbon emission reduction revenue, and total carbon emission control, providing an effective technical path for load aggregator resource autonomous optimization and collaborative scheduling.

[0053] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A game-theoretic decision-making and energy storage optimization scheduling method for load aggregators participating in the electricity carbon market, characterized by: The method includes: S1. Construct a game-theoretic decision-making model for the electric carbon market that includes multi-load aggregators and energy storage systems; It also includes a dynamic subsidy pricing mechanism, which includes response quantity subsidy coefficient constraints and constant quantity subsidy coefficient constraints; S2. Taking the maximization of the comprehensive revenue of each load aggregator as the objective function, and combining the constraints of the power distribution network, user electricity satisfaction, and energy storage system operation, a multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions is constructed. The detailed steps in step S2 for constructing a multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon emissions include: S21. Determine the composition of the comprehensive revenue of each load aggregator, with the objective function being to maximize the comprehensive revenue of each load aggregator. The comprehensive revenue includes demand response revenue, energy storage revenue, and carbon emission reduction revenue, minus demand response costs and energy storage operating costs. S22. Set the distribution network constraints, which include the active and reactive power balance constraints of nodes, voltage change law constraints, line current calculation relationship constraints, and voltage amplitude and line current constraints. The operational constraints of energy storage systems include active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints. A complete multi-agent non-cooperative game optimization model considering the synergistic benefits of electricity and carbon emissions is formed; S3. The non-cooperative game optimization model is transformed into an equilibrium problem, and an improved particle swarm optimization algorithm is used to solve it, so as to obtain the optimal load response strategy and energy storage scheduling plan for each load aggregator. The algorithm steps in step S3, which use the improved particle swarm optimization algorithm to solve the problem, include: S31. Given the network topology of the distribution network, the load data of each node, the predicted output of renewable energy, and the relevant parameters of electricity price and carbon price, set the parameters of the improved particle swarm algorithm, including the inertia weight and its change rules, the initial value of the learning factor, and initialize the position and velocity of the particle swarm. S32. Solve the objective function of each load aggregator, and pass the demand response power, energy storage configuration and scheduling strategy information in the current solution that satisfies the demand of the load aggregator to the particle swarm, and calculate the fitness value of each particle. S33. Compare the current fitness value of each particle with its individual historical best value and the global historical best value. If the current fitness value is better, update the corresponding individual historical best value or the global historical best value and its corresponding strategy; otherwise, retain the historical best value. At the same time, dynamically adjust the inertia weight and learning factor according to the iteration process. S34. Based on the updated parameters and optimal values, update the position and velocity of the particle swarm, and ensure that the new position is feasible in the solution space; S35. Determine whether the solution meets the convergence condition, i.e., the maximum number of iterations has been reached or the quality of the solution is stable. If it is met, output the optimal result; otherwise, return to step S32 to continue iterating. S4. Based on the optimal strategy obtained from the solution, guide each load aggregator to adjust the operating status of its flexible load and energy storage system in real time, so as to achieve the goal of economic and low-carbon operation of the power system.

2. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 1, characterized in that: The response compensation price provided by the power grid depends on the overall load reduction level of all load aggregators. The final revenue of an individual load aggregator depends not only on its own load reduction amount, but also on the load reduction strategies of other load aggregators, as expressed as: ； in, It is the compensation price for the power grid's response to load aggregators; It represents the total load reduction of each load aggregator at different times; This represents the active power of the distribution network load during time period t; This is the price load factor; For constant terms; parameter and All values ​​are greater than 0, and the data are obtained by fitting electricity price data under different load scenarios using the least squares method. To and Since there are no irrelevant constant loads, the formula can be transformed into: ； in, , meaning the response quantity subsidy coefficient; , meaning constant subsidy coefficient; As the load response increases, the unit price of response compensation will decrease accordingly. Once the response power and compensation price of each load aggregator are determined, the load aggregator can obtain the response revenue provided by the power grid during the time period.

3. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 1, characterized in that: demand Response constraints include transferable load constraints and interruptible load constraints, and their specific expressions are as follows: ； ； ； in, This represents the active power of the load at node i during time period t; Let be the transferable load response at node i during time period t; Let be the interruptible load response at node i during time period t; The maximum response coefficient for transferable loads. The maximum response coefficient for interruptible loads; The total number of load nodes participating in demand response. This represents the total number of time periods within the scheduling cycle.

4. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 3, characterized in that: The operational constraints of energy storage systems include active power charging and discharging constraints, reactive power constraints, apparent power constraints, and state of charge constraints, specifically expressed as follows: ； ； ； ； ； ； ； ； in, This represents the discharge power of the energy stored at node i during time period t; This represents the charging power of the energy stored at node i during time period t; express Nodes within the time period The active power of the energy storage device; This indicates the maximum allowable charging and discharging power of the energy storage system; and These are 0-1 variables, representing the charging and discharging flags for energy storage. express Nodes within the time period The reactive power of energy storage; This represents the maximum value of the energy storage reactive power; This indicates the rated apparent power of the energy storage. express State of charge, which stores energy at all times; Indicates the rated capacity of energy storage; and These represent the charging and discharging efficiency of energy storage, respectively. and These represent the maximum and minimum values ​​of the state of charge, respectively.

5. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 4, characterized in that: The user satisfaction constraint is used to ensure that users within the load aggregator's region maintain a reasonable level of electricity satisfaction when responding to transferable loads under time-of-use pricing incentives. The specific expression is as follows: ； ； ； in, To improve user satisfaction with electricity usage; The minimum value for user satisfaction with electricity usage; The impact of subsidized electricity prices on user satisfaction with electricity usage in response to user demand; The maximum impact of demand-response electricity price subsidies on user electricity satisfaction; and The minimum and maximum subsidized electricity prices that affect user satisfaction with electricity use. Let be the transferable load response at node i during time period t; D represents the active power of the load at node i during time period t; D is the total number of load nodes participating in demand response.

6. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 5, characterized in that: The detailed steps in step S2 for constructing a multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon emissions include: S21. Determine the composition of the comprehensive revenue of each load aggregator. The objective function is to maximize the comprehensive revenue of each load aggregator. The comprehensive revenue includes demand response revenue, energy storage revenue, and carbon emission reduction revenue. After deducting demand response costs and energy storage operating costs, the objective function is expressed as follows: ； in, It is the demand response compensation cost of the power grid to load aggregators. This indicates the response costs that the load aggregator provides to users for different types of load. This represents the benefits that the power grid receives from charging and discharging through the energy storage system. This indicates the daily operating cost of the energy storage system. This indicates the regional carbon emission reduction benefits after the response; Demand Response Benefits The formula for calculation is: ； in, It is the compensation price for the power grid's response to load aggregators. It represents the total load reduction of each load aggregator at different times. Indicates the unit simulation time step. This represents the total number of time periods within the scheduling cycle. Response compensation price The initial calculation formula is: ； ； ； ； in, This represents the active power of the distribution network load during time period t. This is the price load factor. For constant terms, This represents the active power of the load at node i during time period t. This represents the load reduction at node i during time period t. For interruptible load response quantities, This refers to the transferable load response. parameter and All values ​​are greater than 0, and the data are fitted using the least squares method from electricity price data under different load scenarios. To and Irrelevant load constant quantity, therefore response compensation price It can be converted into: ； in, For response quantity subsidy coefficient, It is a constant subsidy coefficient, and the unit price of response compensation decreases as the load response increases; when the response power of each load aggregator... and compensation price Once determined, the load aggregator can obtain the response revenue provided by the grid during time period t. ; Demand response cost The formula for calculation is: ； in, The response subsidy price for transferable load, The response subsidy price for interruptible loads; Energy storage revenue The formula for calculation is: ； in, This represents the electricity price sold by the power grid during time period t. This represents the discharge power of the energy stored at node i during time period t. This represents the charging power of the energy stored at node i during time period t; Energy storage operating costs The formula for calculation is: ； in, This represents the discount rate for energy storage. Indicates the service life of energy storage. Indicates the configuration of an energy storage set. This represents the investment cost per unit capacity of energy storage. This indicates the energy storage capacity configuration at node i. This represents the investment cost per unit power of energy storage. This indicates the power configuration for energy storage at node i; carbon emission reduction benefits The formula for calculation is: ； ； ； ； in, This indicates the proportion of regional carbon reduction benefits received by load aggregators. For carbon trading prices, Carbon emissions before and after demand response. This represents the carbon emission intensity of node i during time period t. Let i be the actual active power of renewable energy at node i during time period t. This represents the renewable energy carbon emission intensity of node i, set to zero. Let i represent the set of all branches flowing into node i. This represents the net active power of branch ij during time period t, after deducting line losses. This represents the carbon flux density of branch ij during time period t. This represents the active power of line ij during time period t; S22. Set distribution network constraints, which include active and reactive power balance constraints at nodes, voltage variation constraints, line current calculation relationship constraints, and voltage amplitude and line current constraints, respectively expressed as follows: ； ； ； ； ； ； in, It represents the set of all lines in the distribution network. Represents a node Flow to Node The route, , They represent the lines respectively. Resistance and reactance, express Time-of-use energy storage system at the access node Active power at the location, express Time-of-day routes Current magnitude, express Renewable energy at nodes reactive power, express Time period nodes The reactive power of the load, express Time-of-use energy storage at nodes reactive power, express Time-of-day routes reactive power on , They represent Time period nodes , Voltage at that point and These represent the upper and lower limits of the node voltage, respectively. Indicates the line The maximum current; S23. Integrate the objective function with the distribution network constraints, and combine dynamic subsidy price constraints, demand response constraints, user electricity satisfaction constraints, and energy storage system operation constraints to form a complete multi-agent non-cooperative game optimization model that considers the synergistic benefits of electricity and carbon emissions.

7. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 6, characterized in that: Step S21 Daily operating cost of the energy storage system The calculation method is as follows: ； Where r represents the discount rate for energy storage; y represents the lifespan of the energy storage. This indicates the investment cost per unit capacity of energy storage; This indicates the energy storage capacity configuration at node i; This represents the investment cost per unit power of energy storage; This indicates the power configuration of energy storage at node i; Carbon emissions before and after demand response in carbon reduction benefits The calculation method is as follows: ； ； ； in, , These represent the carbon emissions of node i before and after the demand response in time period t, respectively. This represents the carbon emission intensity of node i during time period t; This represents the carbon flux density of branch ij during time period t; This represents the net active power of branch ij during time period t, after deducting line losses. The renewable energy carbon emission intensity of node i is set to zero; This represents the actual active power of renewable energy at node i during time period t; Indicates the unit simulation time step; This represents the set of all branches flowing into node i.

8. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 1, characterized in that: The algorithm steps for obtaining the optimal load response strategy and energy storage scheduling plan for each load aggregator in step S3 include: S36. Based on the optimal results output in step S35, extract the total load reduction of each load aggregator at different time periods. Combined with the dynamic subsidy price formula Determine the response compensation price for each time period. ; S37. Based on demand response constraints, allocate the transferable load response amount of each load aggregator at different time periods. and interruptible load response This leads to the formation of an optimal load response strategy. S38. Based on the operating constraints of the energy storage system, determine the charging power of the energy storage system in different time periods within the jurisdiction of each load aggregator. Discharge power and state of charge The optimal energy storage scheduling plan is generated; wherein the energy storage system operation constraints are consistent with those in claim 4. S39. Verify whether the optimal load response strategy and energy storage dispatch plan meet the distribution network constraints and user electricity satisfaction constraints. If they meet the constraints, they are determined as the final optimal load response strategy and energy storage dispatch plan. If they do not meet the constraints, return to step S32 for iterative adjustment. The power distribution network constraints are consistent with those in step S22.

9. The method for game-theoretic decision-making and energy storage optimization scheduling of load aggregators participating in the electricity carbon market according to claim 1, characterized in that: Step S4, which guides each load aggregator to adjust the operating status of its flexible loads and energy storage systems in real time based on the optimal strategy obtained from the solution, to achieve the goals of economic and low-carbon operation of the power system, includes the following steps: S41. Establish a communication link between the optimal strategy and the load aggregator control terminal to transmit the transferable load response for each time period. Interruptible load response and energy storage charging and discharging power , The instruction is sent to the corresponding control terminal; S42. The load aggregator control terminal, based on the received instructions, guides users to transfer transferable loads to off-peak hours during peak electricity consumption periods, disconnects some interruptible loads, and controls the energy storage system to discharge to supplement the power supply gap in the grid; during off-peak hours, it controls the energy storage system to charge and store electrical energy. S43. Real-time monitoring of node voltages in the distribution network. Line current and renewable energy output If there are situations such as node voltage exceeding limits, line current exceeding limits, or excessive fluctuations in renewable energy output, the amount of flexible load transfer and energy storage charging and discharging power will be dynamically adjusted based on the optimal strategy. S44. Calculate the daily operating data of the power system, including the revenue of each load aggregator, the total carbon emissions of the system and the carbon emission reduction revenue, and assess the degree of achievement of the economic and low-carbon operation goals of the power system. If the preset goals are not achieved, return to step S3 to re-optimize the game model and solution parameters until the economic and low-carbon operation goals of the power system are achieved.

Images