Power system optimization method based on electricity-carbon collaborative virtual power plant

Abstract

The invention relates to a power system optimization method based on an electricity-carbon collaborative virtual power plant, and the method comprises the steps: determining a total carbon emission control target, considering the carbon emission cost, and constructing a virtual power plant benefit optimization model; solving the virtual power plant benefit optimization model to obtain the output power of each energy supply unit of the virtual power plant; calculating to obtain a carbon emission limit demand increment of the virtual power plant; predicting the total carbon emission demand of the virtual power plant according to the historical carbon emission information and the current load condition of the virtual power plant; solving to obtain an optimal carbon emission limit purchase strategy and the lowest carbon compliance cost at each moment by using a corrected finite difference method; and combining a solving result with the output power of each energy supply unit to obtain an optimal solution of the virtual power plant benefit optimization model, namely an optimal scheduling scheme. According to the method, the cost optimization of the virtual power plant is realized, and the overall carbon emission cost of the power system is reduced on the premise of ensuring the power demand and the total carbon emission control target of the virtual power plant.

Description

Technical Field

[0001] The present invention belongs to the field of power optimization and dispatching, and specifically relates to a power system optimization method based on an electric-carbon collaborative virtual power plant. Background Art

[0002] With the continuous growth of global energy demand and the transformation of energy structure, virtual power plants, as an energy system that integrates multiple distributed energy resources and has autonomous operation and scheduling capabilities, are not only the key to achieving efficient energy utilization and system operation optimization through internal load scheduling optimization, but also optimizing the energy scheduling of each virtual power plant and reducing the overall carbon emission cost while ensuring that electricity demand and total carbon emission control targets are met. This is also a key issue. Summary of the invention

[0003] The purpose of the present invention is to address the above-mentioned problems and provide a power system optimization method based on an electricity-carbon coordinated virtual power plant, construct a cost optimization model for the virtual power plant, and adopt a two-stage solution method. The first stage solves the daily power generation of the virtual power plant, and the second stage solves the optimal carbon emission limit purchase strategy and the lowest carbon emission cost of the virtual power plant, ensuring that the virtual power plant reduces the overall carbon emission cost of the power system while ensuring the power demand and total carbon emission control targets.

[0004] In order to achieve the above object, the technical solution provided by the present invention is: An electric power system optimization method based on an electric-carbon collaborative virtual power plant, wherein the electric power system comprises a power generation module, a user module, a power grid module and a heat network module, wherein the power grid module and the heat network module exchange electricity and heat with the power generation module respectively; the power generation module comprises a plurality of virtual power plants, wherein the virtual power plant comprises an energy supply unit and a carbon emission control module; the carbon emission control module is used to predict and calculate the overall carbon emission demand of the virtual power plant, analyze carbon emission data, and identify carbon emission trends and main sources; the energy supply unit comprises a renewable energy supply unit, a non-renewable energy supply unit, a computing unit and an information transmission unit, wherein the computing unit is used to calculate the control parameters of the energy supply unit and the operating cost and carbon emission cost of the virtual power plant, and the information transmission unit is used to transmit the calculation result data of the computing unit.

[0005] The power system optimization method adopts a two-stage step-by-step solution method to solve the virtual power plant benefit optimization model. In the first stage, the output power of the energy supply unit of the virtual power plant benefit optimization model is solved, and in the second stage, the optimal carbon emission limit purchase strategy and the minimum carbon emission cost of the virtual power plant are solved.

[0006] The power system optimization method comprises the following steps: Step 1: Determine the total carbon emission control target, consider the carbon emission cost, and build a virtual power plant benefit optimization model; Step 2: Solve the virtual power plant benefit optimization model to obtain the output power of each energy supply unit of the virtual power plant; Step 3: According to the output power of the energy supply unit obtained in step 2, the incremental demand for carbon emission limit of the virtual power plant is calculated; Step 4: Based on the historical carbon emission information of the virtual power plant and the current load situation, predict the overall carbon emission demand of the virtual power plant; Step 5: Use the modified finite difference method to solve the optimal carbon emission quota purchase strategy and the minimum cost of carbon compliance at each time; Step 6: Combine the solution of step 5 with the output power of each energy supply unit obtained in step 2 to obtain the optimal solution of the virtual power plant benefit optimization model, that is, the optimal scheduling plan; Step 7: Based on the optimal solution of the virtual power plant benefit optimization model obtained in step 6, guide the operation and dispatch of the power system.

[0007] Preferably, in step 1, the virtual power plant benefit optimization model is: ; (1) In the formula, represents the economic benefits of the virtual power plant, Indicates System operating income of virtual power plants, For the The system operation cost of a virtual power plant, For the is the carbon cost of the virtual power plant; K is the number of virtual power plants, For the The number of virtual power plants.

[0008] Furthermore, the calculation formula for the system operating income of the virtual power plant is: ; (2) In the formula Indicates System operating income of virtual power plants, The revenue from selling electricity to the grid for the virtual power plant, Revenue from selling heat energy from the virtual power plant to the heat network, The revenue from selling energy to users for the virtual power plant, The load response compensation income obtained by the virtual power plant participating in the grid dispatching is , They represent the transmission power between the virtual power plant and the power grid and the heat network respectively. , Represent electricity price and heat price respectively. , They represent the power load and heat load provided to users by the virtual power plant after demand response. Base compensation for interrupted load, represents the interruptible load power; t represents the tth time period; T represents the number of time periods, T=24.

[0009] Furthermore, the calculation formula for the system operation cost of the virtual power plant is: ; (3) In the formula is the consumption cost of natural gas during the dispatch period, The cost required for the operation of the energy supply unit. The cost of dealing with pollution sources, represents the price of natural gas, is the state parameter of the gas turbine, Indicates the lower calorific value of natural gas. is the rated output power of the cogeneration system, Indicates the rated power of the gas boiler system. is the operation and maintenance cost per unit power of the energy supply unit g, Indicates the rated power of the energy supply unit g, g = 1, 2, ... M, represents the output power of the pollution source in the virtual power plant system, Indicates the amount of pollutant a produced per unit output power of the pollution source, a = 1, 2, ... N, represents the basic emission cost per unit of pollutant a, For the The minimum cost of carbon compliance for a virtual power plant of this type, M is the number of energy supply units, and N is the number of pollutant types.

[0010] Preferably, in step 2, the minimum carbon compliance cost of the virtual power plant is not considered for the time being, and only the system operating income and system operating cost of the virtual power plant are considered. According to the different load conditions every day, the non-dominated sorting genetic algorithm II is used to solve the virtual power plant benefit optimization model to obtain the sum of the daily energy supply of the virtual power plant energy supply units in a year. , , Represents the power generation of the virtual power plant on day e.

[0011] Furthermore, in step 3, the incremental demand for carbon emission limit is calculated, and the calculation formula for the incremental demand for carbon emission limit is: ; (4) In the formula is the actual carbon emission intensity, As the carbon emission benchmark, Indicates the year sky; Indicates The daily power generation, Represents the time step.

[0012] Preferably, in step 4, based on historical carbon emission information and current load conditions, a long short-term memory network is used to predict the overall carbon emission demand of all virtual power plants. , establish a carbon emission limit price and The price function is ; (5) in is the superscript index of carbon, and is the cost coefficient of the carbon price.

[0013] Preferably, in step 5, the modified finite difference method is used to solve the optimal carbon emission quota purchase amount to be executed at each carbon trading time. and lowest cost of carbon compliance .

[0014] The specific process of solving the modified finite difference method includes: (1) The optimized space of carbon emission quota is gridded into a two-dimensional grid. , vertical axis , build the index , express The carbon emission quota required at any time, Respectively represent the upper and lower boundaries of the carbon emission limit, is the quantization step size, and Respectively indicate time and carbon emission quota requirements The index unit of (2) Set an initial value and calculate the total carbon emission limit demand based on the predicted value of the total carbon emission demand of the virtual power plant. Set to initial value; (3) Iteratively solve the optimal carbon emission quota purchase amount and the minimum carbon compliance cost required to be executed at each carbon trading moment: Calculate the first The price of a carbon emission allowance per day ; For A virtual power plant is used for loop solution: The carbon cost function is calculated by the following formula: ; (6) In the formula Represents the coordinates in the two-dimensional grid and Corresponding The value of Indicates The purchase amount of carbon emission quotas per day, is the carbon cost control function, Indicates the value space of the purchase amount of carbon emission quotas, ; In the formula Indicates the maximum carbon emission limit demand increment within a day; ; (7) ; (8) in express The initial value of is the risk cost parameter of carbon emission quota demand, express Quantization step size; The value of is obtained by: In the value range Take different values ​​and iterate ,when When it reaches the minimum value, Thus we get .

[0015] Preferably, the carbon emission cost calculation function is: ; (9) In the formula represents the 365th day, represents the penalty function for carbon emissions; The risk cost of delaying the demand for carbon emission limits; ; (10) In the formula Indicates the carbon emission quota demand allowed across periods.

[0016] Furthermore, in order to minimize the carbon emission cost, Find the minimum value , the calculation formula is: ; (11) Taking the derivative of equation (11), ; (12) Satisfy the boundary conditions: ; In the formula represents the penalty function for carbon emissions, .

[0017] Compared with the prior art, the beneficial effects of the present invention include: 1) The present invention integrates dispersed energy resources such as renewable energy and non-renewable energy through a virtual power plant to form a virtual centralized energy system, and uses the intelligent computing unit and information transmission unit of the virtual power plant to achieve Power demand response and optimized operation control of power generation and energy supply can achieve cost optimization control of virtual power plants while ensuring the control targets of power demand and total carbon emissions.

[0018] 2) The present invention constructs a cost optimization model for a virtual power plant and adopts a two-stage solution method. The first stage solves the daily power generation of the virtual power plant, and the second stage solves the optimal carbon emission limit purchase strategy and the lowest carbon emission cost of the virtual power plant, which can effectively reduce the overall carbon emission cost of the power system.

[0019] 3) The present invention utilizes the advantages of the modified finite difference method, such as its wide applicability, ability to handle nonlinear problems, simple and intuitive calculation process, and high flexibility, to solve the optimal carbon emission quota purchase strategy and the lowest cost of carbon compliance at each moment. The modified finite difference method, as a method for numerically solving partial differential equations, is applicable to various types of equations, including those complex equations that describe the operation of virtual power plants and carbon emission quota purchase strategies. Through the modified finite difference method, these problems can be handled more accurately, making it suitable for solving the virtual power plant benefit optimization model. The operation and carbon emission management of virtual power plants often involve nonlinear problems such as the volatility of power generation and the dynamic changes of carbon emission quotas. The modified finite difference method can handle these nonlinear problems and ensure the accuracy and reliability of the solution results. The modified finite difference method discretizes the original problem into a differential format by dividing the continuous solution domain into differential grids and replacing the continuous solution domain with limited grid nodes. This method has a simple and intuitive calculation process, is easy to implement and understand, and is easy to promote and apply in practical applications. The modified finite difference method can be adapted to the adaptive grid method and dynamically adjust the grid or difference format according to the characteristics of the solution. Compared with the traditional finite difference method, it can better handle boundary conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0021] Figure 1 FIG. 4 is a schematic diagram of a power system according to an embodiment of the present invention.

[0022] Figure 2This is a distribution diagram of the optimal carbon emission limit procurement volume of a Class 1 virtual power plant in a year in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] like Figure 1 As shown, a power system optimization method based on an electric-carbon collaborative virtual power plant is provided, wherein the power system comprises a power generation module, a user module, a power grid module and a heat network module, wherein the power grid module and the heat network module exchange electricity and heat with the power generation module respectively; the power generation module comprises four types of virtual power plants, namely Class1, Class2, Class3 and Class4, wherein the virtual power plant comprises an energy supply unit and a carbon emission control module; the carbon emission control module calculates the overall carbon emission demand of the virtual power plant based on the long short-term memory network prediction, analyzes the carbon emission data, and identifies the trend and main source of carbon emission; the energy supply unit comprises a renewable energy supply unit, a non-renewable energy supply unit, a calculation unit and an information transmission unit, wherein the calculation unit is used to calculate the control parameters of the energy supply unit and the operating cost and carbon emission cost of the virtual power plant, and the information transmission unit is used to transmit the calculation result data of the calculation unit. The user module is an electricity consumption unit that sends a demand response to the power generation module. The above modules are connected by transmission lines to transmit electric energy, and at the same time, the modules are connected by communication to transmit information.

[0024] The power system optimization method based on the electric-carbon collaborative virtual power plant includes the following steps: Step 1: Determine the total carbon emission control target, consider the carbon emission cost, and build a virtual power plant benefit optimization model; In the embodiment, the virtual power plant benefit optimization model is: ; (1) In the formula, represents the economic benefits of the virtual power plant, Indicates System operating income of virtual power plants, For the The system operation cost of a virtual power plant, For the is the carbon cost of the virtual power plant; K is the number of virtual power plants, For the The number of virtual power plants.

[0025] The calculation formula for the system operating income of the virtual power plant is: ; (2) In the formula Indicates System operating income of virtual power plants, The revenue from selling electricity to the grid for the virtual power plant, Revenue from selling heat energy from the virtual power plant to the heat network, The revenue from selling energy to users for the virtual power plant, The load response compensation income obtained by the virtual power plant participating in the grid dispatching is , They represent the transmission power between the virtual power plant and the power grid and the heat network respectively. , Represent electricity price and heat price respectively. , They represent the power load and heat load provided to users by the virtual power plant after demand response. Base compensation for interrupted load, represents the interruptible load power; t represents the tth time period; T represents the number of time periods, T=24.

[0026] The calculation formula for the system operation cost of the virtual power plant is: ; (3) In the formula is the consumption cost of natural gas during the dispatch period, The cost required for the operation of the energy supply unit. The cost of dealing with pollution sources, represents the price of natural gas, is the state parameter of the gas turbine, Indicates the lower calorific value of natural gas. is the rated output power of the cogeneration system, Indicates the rated power of the gas boiler system. is the operation and maintenance cost per unit power of the energy supply unit g, Indicates the rated power of the energy supply unit g, g = 1, 2, ... M, represents the output power of the pollution source in the virtual power plant system, Indicates the amount of pollutant a produced per unit output power of the pollution source, a = 1, 2, ... N, represents the basic emission cost per unit of pollutant a, For the The minimum cost of carbon compliance for a virtual power plant of this type, M is the number of energy supply units, and N is the number of pollutant types.

[0027] Step 2: Solve the virtual power plant benefit optimization model to obtain the output power of each energy supply unit of the virtual power plant; In the embodiment, the minimum carbon compliance cost of the virtual power plant is not considered for the time being. Only the system operating income and system operating cost of the virtual power plant are considered. According to the different load conditions every day, the non-dominated sorting genetic algorithm II is used to solve the virtual power plant benefit optimization model to obtain the sum of the daily energy supply of the virtual power plant energy supply units in a year. , , Represents the power generation of the virtual power plant on day e.

[0028] Step 3: According to the output power of the energy supply unit obtained in step 2, the incremental demand for carbon emission limit of the virtual power plant is calculated; Calculate the incremental demand for carbon emission limits. The calculation formula for the incremental demand for carbon emission limits is: ; (4) In the formula is the actual carbon emission intensity, As the carbon emission benchmark, Indicates the year sky; Indicates The daily power generation, Represents the time step.

[0029] Step 4: Based on the historical carbon emission information of the virtual power plant and the current load situation, predict the overall carbon emission demand of the virtual power plant; Based on historical carbon emission information and current load conditions, the long short-term memory network is used to predict the overall carbon emission demand of all virtual power plants. , establish a carbon emission limit price and The price function is ; (5) in is the superscript index of carbon, and is the cost coefficient of the carbon price.

[0030] Step 5: Use the modified finite difference method to solve the optimal carbon emission quota purchase strategy and the minimum cost of carbon compliance at each time; In the embodiment, the modified finite difference method is used to solve the optimal carbon emission quota purchase amount required to be executed at each carbon trading time. and lowest cost of carbon compliance ; The specific process of solving the modified finite difference method includes: (1) The optimized space of carbon emission quota is gridded into a two-dimensional grid. , vertical axis , build the index , express The carbon emission quota required at any time, Respectively represent the upper and lower boundaries of the carbon emission limit, is the quantization step size, and Respectively indicate time and carbon emission quota requirements The index unit of (2) Set an initial value and calculate the total carbon emission limit demand based on the predicted value of the total carbon emission demand of the virtual power plant. Set to initial value; (3) Iteratively solve the optimal carbon emission quota purchase amount and the minimum carbon compliance cost required to be executed at each carbon trading moment: Calculate the first The price of a carbon emission allowance per day ; For A virtual power plant is used for loop solution: The carbon cost function is calculated by the following formula: ; (6) In the formula Represents the coordinates in the two-dimensional grid and Corresponding The value of Indicates The purchase amount of carbon emission quotas per day, is the carbon cost control function, Indicates the value space of the purchase amount of carbon emission quotas, ; In the formula Indicates the maximum carbon emission limit demand increment within a day; ; (7) ; (8) in express The initial value of is the risk cost parameter of carbon emission quota demand, express Quantization step size; The value of is obtained by: In the value range Take different values ​​and iterate ,when When it reaches the minimum value, Thus we get .

[0031] The carbon emission cost calculation function is: ; (9) In the formula represents the 365th day, represents the penalty function for carbon emissions; The risk cost of delaying the demand for carbon emission limits; ; (10) In the formula Indicates the carbon emission quota demand allowed across periods.

[0032] In order to minimize the carbon emission cost, Find the minimum value , the calculation formula is: ; (11) Taking the derivative of equation (11), ; (12) Satisfy the boundary conditions: ; In the formula represents the penalty function for carbon emissions, .

[0033] Step 6: Combine the solution of step 5 with the output power of each energy supply unit obtained in step 2 to obtain the optimal solution of the virtual power plant benefit optimization model, that is, the optimal scheduling plan; Step 7: Based on the optimal solution of the virtual power plant benefit optimization model obtained in step 6, guide the operation and dispatch of the power system.

[0034] In the embodiment, the modified finite difference method can be flexibly adjusted and optimized according to the specific situation and needs of the virtual power plant. For example, the division of the difference grid and the selection of the difference formula can be adjusted according to different power generation data, carbon emission quota purchase strategies and other conditions to obtain a more accurate solution result.

[0035] In the embodiment, Class 1, Class 2, Class 3, and Class 4 virtual power plants all use a cogeneration system for power and heat supply, as shown in Table 1.

[0036] Table 1 Parameters of four types of virtual power plants

[0037] It can be seen that the four types of virtual power plants have the same units, but different types of virtual power plants have different power generation characteristics, installed capacity and working conditions. Taking Class 1 virtual power plant as an example, The value is 1, the step length The value is 0.5, the lower limit of the carbon emission limit The value is 0, the upper limit of the carbon emission limit is 130, and the optimal carbon emission quota purchase amount of Class 1 virtual power plant is obtained as follows: Figure 2 shown. Figure 2 In the graph, the X-axis represents 365 days in a year, the Y-axis represents the carbon emission demand in kilotons, and the Z-axis represents the purchase volume of carbon emission quotas in kilotons. Figure 2 On the right is the color bar of the purchase amount of carbon emission quotas, and different carbon emission quota purchase amounts correspond to different colors. Figure 2 In the heat map shown, the corresponding optimal carbon emission limit purchase amount can be found according to the color bar on the right.

[0038] In the embodiment, the user module does not contain any power generation unit and cannot be used for power generation. Each user module can be regarded as a unit participating in power trading, but its power generation is limited to a negative value, then it becomes a one-way power purchasing unit. The power grid module and the heat network module contain power generation units, which can also be power purchasing units and can purchase electricity from virtual power plants. The present invention only addresses the carbon emission problem of virtual power plants. Other modules do not consider carbon emission issues. It is mainly based on the electricity demand in the market, and optimizes the resource scheduling of energy supply units on the premise of meeting the electricity demand, thereby improving resource utilization and enhancing environmental friendliness.

[0039] The system of power system optimization method based on electric-carbon collaborative virtual power plant includes: Virtual power plant cost optimization modeling module, used to build a virtual power plant benefit optimization model; The virtual power plant power generation solution module is used to solve the virtual power plant's daily power generation according to the virtual power plant benefit optimization model; The optimal carbon emission quota purchase strategy module is used to solve the optimal carbon emission quota purchase amount and the minimum carbon compliance cost that the virtual power plant needs to execute at each carbon trading moment based on the virtual power plant benefit optimization model and the daily power generation of the virtual power plant output by the virtual power plant power generation solution module; The optimal dispatching scheme module obtains the optimal dispatching scheme for various virtual power plants based on the daily power generation of the virtual power plant output by the virtual power plant power generation solution module, the optimal carbon emission quota purchase volume and the minimum carbon compliance cost output by the optimal carbon emission quota purchase strategy module; The dispatch instruction sending module issues dispatch instructions to the operators of the energy supply units of each virtual power plant according to the optimal dispatch plan output by the optimal dispatch plan module.

[0040] The present invention takes into account the carbon emission problem on the basis of the optimal economic benefit optimization problem of conventional virtual power plants, and mainly solves the carbon cost problem. On the premise of ensuring that the electricity demand and the total carbon emission control targets are met, the energy scheduling of the power supply units of each virtual power plant is optimized. Based on this, the optimal carbon emission limit purchase strategy that needs to be executed at each moment is given, and the lowest carbon compliance cost can also be obtained. Under the premise of ensuring power supply demand, carbon emission demand and responding to policy requirements, the economic benefits of virtual power plants can be optimized to the greatest extent.

Claims

1. A power system optimization method based on electric-carbon collaborative virtual power plant, characterized in that: The power system includes a power generation module, a user module, a power grid module and a heat network module. The power generation module includes a plurality of virtual power plants of different types. The virtual power plant includes an energy supply unit and a carbon emission control module. The carbon emission control module is used to predict and calculate the overall carbon emission demand of the virtual power plant, analyze carbon emission data, and identify carbon emission trends; the energy supply units include renewable energy supply units and non-renewable energy supply units; The power system optimization method adopts a two-stage step-by-step solution method to solve the virtual power plant benefit optimization model. In the first stage, the output power of the energy supply unit of the virtual power plant benefit optimization model is solved, and in the second stage, the optimal carbon emission limit purchase strategy and the minimum carbon emission cost of the virtual power plant are solved. The power system optimization method comprises the following steps: Step 1: Determine the total carbon emission control target, consider the carbon emission cost, and build a virtual power plant benefit optimization model; Step 2: Solve the virtual power plant benefit optimization model to obtain the output power of each energy supply unit of the virtual power plant; Step 3: Calculate the incremental carbon emission quota demand of the virtual power plant based on the sum of the output powers of the energy supply units obtained in step 2; Step 4: Based on the historical carbon emission information of the virtual power plant and the current load situation, predict the overall carbon emission demand of the virtual power plant; Step 5: Use the modified finite difference method to solve the optimal carbon emission quota purchase strategy and the minimum cost of carbon compliance at each time; Step 6: Combine the solution of step 5 with the output power of each energy supply unit obtained in step 2 to obtain the optimal solution of the virtual power plant benefit optimization model, that is, the optimal scheduling plan; Step 7: Based on the optimal solution of the virtual power plant benefit optimization model obtained in step 6, guide the operation and dispatch of the power system.

2. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 1 is characterized in that: In step 1, the virtual power plant benefit optimization model is: ；（1） In the formula, represents the economic benefits of the virtual power plant, Indicates System operating income of virtual power plants, For the The system operation cost of a virtual power plant, For the The carbon cost of virtual power plants of this type; K is the number of virtual power plants, For the The number of virtual power plants.

3. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 2 is characterized in that: In step 1, the calculation formula for the system operating income of the virtual power plant is: ；（2） In the formula Indicates System operating income of virtual power plants, The revenue from selling electricity to the grid for the virtual power plant, Revenue from selling heat energy from the virtual power plant to the heat network, The revenue from selling energy to users for the virtual power plant, The load response compensation income obtained by the virtual power plant participating in the grid dispatching is , They represent the transmission power between the virtual power plant and the power grid and the heat network respectively. , Represent electricity price and heat price respectively. , They represent the power load and heat load provided to users by the virtual power plant after demand response. Base compensation for interrupted load, represents the interruptible load power; t represents the tth time period; T represents the number of time periods.

4. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 3 is characterized in that: In step 1, the calculation formula for the system operation cost of the virtual power plant is: ；（3） In the formula is the consumption cost of natural gas during the dispatch period, The cost required for the operation of the energy supply unit. The cost of dealing with pollution sources, represents the price of natural gas, is the state parameter of the gas turbine, Indicates the lower calorific value of natural gas. is the rated output power of the cogeneration system, Indicates the rated power of the gas boiler system. is the operation and maintenance cost per unit power of the energy supply unit g, Indicates the rated power of the energy supply unit g, g = 1, 2, ... M, represents the output power of the pollution source in the virtual power plant system, Indicates the amount of pollutant a generated per unit output power of the pollution source, a = 1, 2, ... N, represents the basic emission cost per unit of pollutant a, For the The minimum cost of carbon compliance for a virtual power plant, M is the number of energy supply units, and N is the number of pollutant types.

5. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 4 is characterized in that: In step 2, the minimum carbon compliance cost of the virtual power plant is not considered for the time being. Only the system operating income and system operating cost of the virtual power plant are considered. The non-dominated sorting genetic algorithm II is used to solve the virtual power plant benefit optimization model to obtain the sum of the daily energy supply of the virtual power plant energy supply units in a year. , , Represents the power generation of the virtual power plant on day e.

6. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 5 is characterized in that: In step 3, the incremental demand for carbon emission limit is calculated, and the calculation formula for the incremental demand for carbon emission limit is: ；（4） In the formula is the actual carbon emission intensity, As the carbon emission benchmark, Indicates the year sky; Indicates The daily power generation, Represents the time step.

7. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 6 is characterized in that: In step 4, based on historical carbon emission information and current load conditions, the long short-term memory network is used to predict the overall carbon emission demand of all virtual power plants. , establish a carbon emission limit price and The price function is ；（5） in is the superscript index of carbon, and is the cost coefficient of the carbon price.

8. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 7 is characterized in that: In step 5, the modified finite difference method is used to solve the optimal carbon emission quota purchase amount required to be executed at each carbon trading time. and lowest cost of carbon compliance ; The specific process of solving the modified finite difference method includes: (1) The optimized space of carbon emission quota is gridded into a two-dimensional grid. , vertical axis , build the index , express The carbon emission quota required at any time, Respectively represent the upper and lower boundaries of the carbon emission limit, is the quantization step size, and Respectively indicate time and carbon emission quota requirements The index unit of (2) Set an initial value and calculate the total carbon emission limit demand based on the predicted value of the total carbon emission demand of the virtual power plant. Set to initial value; (3) Iteratively solve the optimal carbon emission quota purchase amount and the minimum carbon compliance cost at each carbon trading moment: Calculate the first The price of a carbon emission allowance per day ; For A virtual power plant is used for loop solution: The carbon cost function is calculated by the following formula: ；（6） In the formula Represents the coordinates in the two-dimensional grid and Corresponding The value of Indicates The purchase amount of carbon emission quotas per day, is the carbon cost control function, Indicates the value space of the purchase amount of carbon emission quotas, ； In the formula Indicates the maximum carbon emission limit demand increment within a day; ；（7） ；（8） in express The initial value of is the risk cost parameter of carbon emission quota demand, express Quantization step size; The value of is obtained by: In the value range Take different values ​​and iterate ,when When it reaches the minimum value, Thus we get .

9. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 7 or 8 is characterized in that: The carbon emission cost calculation function is: ； （9） In the formula represents the 365th day, represents the penalty function for carbon emissions; The risk cost of delaying the demand for carbon emission limits; ； （10） In the formula Indicates the carbon emission quota demand allowed across periods.

10. The power system optimization method based on electric-carbon collaborative virtual power plant according to claim 9 is characterized in that: for Find the minimum value , the calculation formula is: ；（11） Taking the derivative of equation (11), ；（12） Satisfy the boundary conditions: ； In the formula represents the penalty function for carbon emissions, .

Images