Virtual power plant optimal scheduling method and device based on electricity-carbon transaction, and medium

Abstract

The invention provides a virtual power plant optimal scheduling method and device based on electricity-carbon transaction, and a medium. The method specifically comprises the following steps: constructing an optimal scheduling framework of a multi-virtual power plant alliance, and establishing a refined mathematical model of a single virtual power plant; a step transaction mechanism of the carbon market is established, and accurate linearization of the carbon emission cost is carried out by using SOS2; establishing a minimization model of the total operation cost of the multi-virtual power plant alliance containing the carbon emission cost; and solving the collaborative optimization problem of the multi-virtual power plant alliance by adopting an alternating direction multiplier method. According to the virtual power plant optimal scheduling method, a step transaction mechanism of a carbon market is established, accurate linearization of carbon emission cost is carried out by using an SOS2 method, scheduling analysis under power-carbon coupling is carried out on a multi-virtual power plant alliance by using an alternating multiplier method, and a total operation cost minimization objective function containing the carbon emission cost is established; therefore, coupling of carbon price and electricity price is achieved, effective management of virtual power plant energy is achieved, and the effect of peak load shifting is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system optimization and scheduling, and specifically relates to a virtual power plant optimization and scheduling method, device and medium for electricity-carbon trading. Background Art

[0002] A virtual power plant (VPP) is a system that uses advanced information and communication technologies and software systems. Optimized scheduling of a virtual power plant refers to the process of achieving coordinated and optimized resource operation by integrating distributed energy, energy storage systems, controllable loads, and other resources using mathematical models and intelligent algorithms.

[0003] As independent entities participating in market transactions, VPPs face challenges such as limited scheduling flexibility and poor reliability. Scholars have proposed strategies for the joint optimization of multiple VPPs, known as VPP alliances. This multi-party collaborative operation enables resource complementarity, significantly improving VPP scheduling flexibility and transaction reliability.

[0004] Existing research on the optimal scheduling of virtual power plants mainly aims to maximize the benefits of participating in the electricity market or minimize operating costs, and only focuses on the supply, demand and price of electricity commodities, ignoring the costs or benefits of carbon emissions, and focusing on carbon emission trading in industries such as industry and transportation, with a low degree of coupling with electricity production and consumption; some improvements have been added to the calculation of carbon emission costs, and most of them use a fixed carbon price, which cannot reflect the punitive increase in excess emissions and is difficult to incentivize low-carbon operation, resulting in scheduling strategies that do not meet the "dual carbon" goals. Summary of the Invention

[0005] In response to the aforementioned problems in the prior art, the present invention aims to provide a method and apparatus for optimizing the scheduling of virtual power plants based on electricity-carbon trading. By establishing a tiered trading mechanism for the carbon market and utilizing the SOS2 method to accurately linearize the cost of carbon emissions, the alternating multiplier method is used to analyze the scheduling of multiple virtual power plant alliances under electricity-carbon coupling. This establishes an objective function for minimizing the total operating cost, including the carbon emission cost, and achieves the coupling of carbon and electricity prices, effectively managing the energy of virtual power plants and achieving peak shaving and valley filling.

[0006] The technical solution adopted by the present invention to solve its technical problem is: In a first aspect, the present invention provides a virtual power plant optimization scheduling method based on electricity-carbon trading, which includes the following steps: Step S1: constructing an optimized scheduling framework for a multi-virtual power plant alliance, and establishing a refined mathematical model of a single virtual power plant based on the optimized scheduling framework; Step S2: Establish a ladder trading mechanism for the carbon market and use the second type of special ordered set technology to accurately linearize the carbon emission cost; Step S3: establishing a model for minimizing the total operating cost of the multi-virtual power plant alliance including the carbon emission cost based on the tiered trading mechanism; Step S4: solving the collaborative optimization problem of the multi-virtual power plant alliance using an alternating direction multiplier method; Step S5: Introducing cooperative game theory to solve the profit distribution of each virtual power plant in the multi-virtual power plant alliance, and using Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair profit distribution plan; The optimization scheduling framework includes transactions in the electricity market and the carbon market, and cooperative games among multiple virtual power plants.

[0007] Preferably, the step S2 further includes: Step S21: Divide the excess carbon emissions into multiple levels, with the carbon price of each level increasing with the number of levels; Step S22: using the second type of special ordered set technology to convert the nonlinear function of the carbon emission cost into a linear constraint; The multiple steps include carbon-free cost, fixed carbon price, stepped carbon price and piecewise linear continuous carbon price.

[0008] Preferably, the step S22 further includes: Step S221: defining an inflection point sequence describing the relationship between cumulative carbon costs and cumulative excess carbon emissions; Step S222: introducing non-negative weight variables and applying the second type of special ordered set constraints; Step S223: Calculate the net excess carbon emissions and the corresponding total carbon cost through convex combination.

[0009] Preferably, the objective function of the total operating cost minimization model of the multi-virtual power plant alliance in step S3 is:

[0010] Where, is the total operating cost of the multi-virtual power plant alliance.

[0011] Preferably, the step S4 further includes: Step S41: decompose the collaborative optimization problem into multiple sub-problems, and optimize the decision variables of each virtual power plant respectively; Step S42: alternately updating the decision variables to gradually approach the optimal solution; Step S43: coupling the decision variables through the augmented Lagrangian function, and alternately updating the original variables and the dual variables; Step S44: dynamically adjust the penalty factor of the augmented Lagrangian function to accelerate convergence.

[0012] Preferably, the step of quantifying the marginal contribution of each virtual power plant using the Shapley value in step S5 further includes: Step S51: traverse all virtual power plant alliances combinations, calculating the characteristic function value of each of the combinations; Step S52: determining the marginal contribution of each virtual power plant based on the characteristic function value; Step S53: Determine the Shapley value of each virtual power plant based on the weighted sum of the marginal contributions, and distribute the total revenue in the multi-virtual power plant alliance according to the proportion of the Shapley value.

[0013] Preferably, the refined mathematical model of a single virtual power plant includes a distributed photovoltaic output model, a gas turbine operation mathematical model, an energy storage system operation mathematical model, and an interaction model between the virtual power plant and the main grid.

[0014] Preferably, the carbon emission calculation formula of the gas turbine for establishing the mathematical model of the gas turbine operation is as follows:

[0015] in, is the carbon emission intensity of MT power generation (kg / kWh or t / kWh), It is the fixed carbon emissions during MT operation. In a second aspect, the present invention provides a virtual power plant optimization scheduling device based on electricity-carbon trading. Specifically, it includes: A model building module is used to construct an optimized scheduling framework for multiple virtual power plants and establish a refined mathematical model of a single virtual power plant based on the optimized scheduling framework; Cost linearization module, used to establish a ladder trading mechanism for the carbon market, using the second type of special ordered set technology to accurately linearize the cost of carbon emissions; An operating cost module, configured to establish a minimization model for the total operating cost of the multi-virtual power plant alliance including the carbon emission cost based on the tiered trading mechanism; a collaborative optimization module, configured to solve the collaborative optimization problem of the multi-virtual power plant alliance by adopting an alternating direction multiplier method; a benefit distribution module, configured to introduce cooperative game theory to solve the benefit distribution of each virtual power plant in the multi-virtual power plant alliance, and use Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair benefit distribution plan; The optimization scheduling framework includes transactions in the electricity market and the carbon market, and cooperative games among multiple virtual power plants.

[0016] In a third aspect, the present invention provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, it implements the virtual power plant optimization scheduling method based on electricity-carbon trading as described in any one of the first aspects.

[0017] The present invention provides a method for optimizing virtual power plant scheduling based on electricity-carbon trading, specifically comprising the following steps: constructing an optimized scheduling framework for a multi-virtual power plant alliance and establishing a refined mathematical model for a single virtual power plant; establishing a step-by-step trading mechanism for the carbon market and utilizing a two-class special ordered set technique to precisely linearize carbon emission costs; establishing a model for minimizing the total operating cost of the multi-virtual power plant alliance, including carbon emission costs; and employing an alternating direction multiplier method to solve the collaborative optimization problem of the multi-virtual power plant alliance. The virtual power plant optimization scheduling method of the present invention establishes a step-by-step trading mechanism for the carbon market and utilizes the SOS2 method to precisely linearize carbon emission costs. The alternating direction multiplier method is then used to analyze the scheduling of the multi-virtual power plant alliance under electricity-carbon coupling, establishing an objective function for minimizing the total operating cost, including carbon emission costs. This method achieves coupling of carbon and electricity prices, effectively manages the energy consumption of virtual power plants, and achieves peak shaving and valley filling. By deeply integrating "electricity value" with "carbon value," this method leverages the electricity market's role in allocating energy resources while leveraging price signals from the carbon market to promote low-carbon transformation in the power industry, thereby fostering the "dual carbon" goal and promoting market innovation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below with reference to the accompanying drawings and examples.

[0019] Figure 1 1 is a flow chart of a virtual power plant optimization scheduling method based on electricity-carbon trading according to embodiment 1 of the present invention; Figure 2 2. It is a schematic diagram of an optimized scheduling framework of a multi-virtual power plant alliance according to embodiment 1 of the present invention; Figure 3 is a schematic flow chart of the step-by-step method in step S2 of Example 1 of the present invention; Figure 4 1 is a flow chart of the linear constraint method in step S2 of embodiment 1 of the present invention; Figure 5 is a schematic flow chart of the step-by-step method in step S4 of Example 1 of the present invention; Figure 6 is a schematic flow chart of the step-by-step method in step S5 of Example 1 of the present invention; Figure 7 This is a module diagram of a virtual power plant optimization scheduling device based on electricity-carbon trading according to Example 2 of the present invention; Figure 81 is a schematic diagram of submodules of the cost linear module 12 of embodiment 2 of the present invention; Figure 9 It is a schematic diagram of the submodules of the collaborative optimization module 14 of Example 2 of the present invention. DETAILED DESCRIPTION

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] Example 1 This embodiment provides a virtual power plant optimization scheduling method based on electricity carbon trading, such as Figure 1 As shown, the specific steps include: Step S1: construct an optimized scheduling framework for a multi-virtual power plant alliance, and establish a refined mathematical model of a single virtual power plant based on the optimized scheduling framework.

[0022] In this embodiment, if Figure 2 As shown in the figure, the optimized dispatch framework includes transactions in the electricity and carbon markets, as well as cooperative game-playing among multiple virtual power plants. The electricity and carbon markets operate according to their optimal strategies and upload transaction information to the electricity-carbon market service platform. Based on transaction information and the operating status of the upstream power grid, information such as energy prices and carbon quota prices is generated to guide the optimized operation of multiple virtual power plants participating in the electricity and carbon markets. Virtual power plant operators report electricity trading volumes and prices. Through data interaction and iterative calculations, the final energy trading results are determined, and corresponding profit distribution plans are formulated. The final transaction results are then transmitted to each virtual power plant via a communication network. After receiving this information, each virtual power plant reports its electricity and carbon quota trading needs through the electricity-carbon market and reaches a consensus on the transaction, fully leveraging the complementary advantages of cross-regional resources.

[0023] In this embodiment, the refined mathematical model of a single virtual power plant includes a distributed photovoltaic output model, a gas turbine operation mathematical model, an energy storage system operation mathematical model, and a virtual power plant and main grid power interaction model.

[0024] As an optional embodiment, the objective function formula of the distributed photovoltaic output model is as follows: ; Where, for The maximum available power of the photovoltaic system during the period, is the comprehensive conversion efficiency of the photovoltaic array, is the total area of ​​the photovoltaic array, for Solar radiation intensity during the period (kW / m²), is the power temperature coefficient of the solar panel, for The actual operating temperature of the solar panel during the period, is the reference temperature; In this embodiment, the distributed photovoltaic output model requires some constraints during calculation. Each constraint is expressed as a function: The photovoltaic dispatch output constraint function is expressed as: ; Where, for The actual power of photovoltaic power that is connected to the grid or used internally by the VPP during the period; Due to the uncertainty of photovoltaic prediction, the PV output in the future can be more accurately predicted through weather forecasts and statistical models to obtain the predicted output curve. ; Therefore, the amount of abandoned light It can be expressed as: .

[0025] As an optional embodiment, the following constraints are required when calculating the mathematical model for the operation of the energy storage system. Each constraint is expressed as a function: The output constraint can be expressed as: ; In the formula for Output power of MT during the time period, for The running status of the MT in the period, where 0 is shutdown and 1 is startup, is a binary decision variable. and They are the minimum and maximum technical output of MT respectively.

[0026] The climbing constraint can be expressed as: ; ; in, and They are the maximum up-climbing rate and down-climbing rate of MT respectively.

[0027] The fuel cost can be expressed as: ; in, is the fuel cost coefficient of MT, Can represent fixed operating costs or no-load consumption costs.

[0028] The start-stop cost can be expressed as: ; ; In the formula and are the single startup and shutdown costs, respectively. In MILP, this is usually represented accurately by introducing additional binary variables.

[0029] In this embodiment, the emission of the micro gas engine is generally proportional to its fuel consumption. Thus, the emission can be expressed as a function of its output power, which is usually simplified to a linear relationship. The specific expression is as follows: ; in, is the carbon emission intensity of MT power generation (kg / kWh or t / kWh), It is the fixed carbon emissions during MT operation.

[0030] As an optional embodiment, the objective function formula of the mathematical model for the energy storage system operation is: ; In the formula for The state of charge of the energy storage system at the end of the period (usually expressed as a decimal between 0 and 1, or a percentage), and They are The charging and discharging power of the time-slot energy storage system, and are the charging and discharging efficiencies of the energy storage system, is the rated capacity of the energy storage system (kWh), is the length of the scheduling time interval (h).

[0031] The SOC constraint formula is: ; In the formula and are the minimum and maximum states of charge allowed by the energy storage system.

[0032] The charge and discharge constraint formula is: ; ; ; in and are the maximum charging power and maximum discharging power of the energy storage system respectively. and are binary variables, representing The charging and discharging status of the time period (0 for no, 1 for yes). The third formula in the charge and discharge constraint formula is used to ensure that the energy storage system cannot be charged and discharged at the same time in the same time period.

[0033] The initial and final SOC simulation setting formulas are: ; In order to ensure the sustainability of the scheduling cycle, The operating cost of energy storage is primarily considered in terms of cycle life losses. In short-term optimization, this is sometimes simplified to charge and discharge efficiency losses, or indirectly considered by limiting the number of daily charge and discharge cycles or total throughput.

[0034] The formula for including cost in the objective function is: ; In the formula It is the aging cost per unit charge and discharge capacity.

[0035] As an optional embodiment, the calculation of the interaction model between the virtual power plant and the main grid requires some constraints, and each constraint is expressed as a function: The power constraint formula for purchasing and selling electricity is: ; ; In the formula and are the maximum power of electricity purchased from and sold to the main grid by the VPP respectively. and are binary variables, representing The electricity purchasing and selling status of the time period.

[0036] The mutually exclusive constraint formula for electricity purchase and sale is: ; The cost / benefit formula for electricity purchase and sales is: ; in, It is the time period The price of electricity purchased by VPP from the main grid ($ / kWh), It is the time period The price of electricity sold by the VPP to the main grid ($ / kWh), is the length of the scheduling period in hours.

[0037] The formula for the net exchange power with the main grid is: ; In the formula Positive values ​​indicate net electricity purchases, and negative values ​​indicate net electricity sales: The formula for carbon emissions related to electricity purchase is: ; in for The average carbon emission factor of the power grid during the period (kg / kWh or t / kWh).

[0038] Step S2: Establish a ladder trading mechanism for the carbon market and use the Special Ordered Sets of Type 2 (SOS2) technology to accurately linearize the carbon emission cost.

[0039] As an optional embodiment, Figure 3 As shown, step S2 also includes: Step S21: Divide the excess carbon emissions into multiple tiers, with the carbon price of each tier increasing with the number of tiers. In this embodiment, the multiple tiers include carbon-free cost, fixed carbon price, stepped carbon price, and piecewise linear continuous carbon price.

[0040] Step S22: using SOS2 to convert the nonlinear function of the carbon emission cost into a linear constraint.

[0041] In this embodiment, if Figure 4 As shown, step S22 also includes: Step S221: Define an inflection point describing the relationship between cumulative carbon costs and cumulative excess carbon emissions sequence, where . Among them, the definition of the inflection point coordinates is: The 0th inflection point (origin) is .

[0042] For the subsequent Inflection point ( ), its coordinate calculation formula is as follows: ; ; Where, Before filling The cumulative excess emissions during each stage, Indicates the corresponding cumulative carbon cost at this time.

[0043] Step S222: Introduce non-negative weight variables , and impose two types of special ordered set constraints; Step S223, calculate the net excess carbon emissions through convex combination and the corresponding total carbon cost ; Among them, the SOS2 linearization constraint formula is: ; ; ; Where, SOS2 constraints are imposed on this set of weight variables:

[0044] In the formula, the SOS2 constraint stipulates that: in the set In , there can be at most two variables with non-zero values, and if there are two non-zero values, they must be adjacent. This clever constraint ensures that the point Must fall between two adjacent inflection points and This perfectly reproduces the piecewise linear cost function relationship without introducing additional binary variables for each step.

[0045] Step S3: Based on the ladder trading mechanism, a model for minimizing the total operating cost of a multi-virtual power plant alliance including carbon emission costs is established.

[0046] In this embodiment, the total operating cost of the internal optimization scheduling model of a single VPP in step S3 is the sum of the operating costs of grid power exchange, microturbine (MT) power generation, and energy storage system (ESS). Its objective function is expressed as: ; Where, is VPPi in cycle Total operating costs within.

[0047] The objective function of the model for minimizing the total operating cost of a multi-virtual power plant alliance is:

[0048] Where, It is the total operating cost of the VPP alliance.

[0049] Step S4: Using the Alternating Direction Method of Multipliers (ADMM) method to solve the collaborative optimization problem of the multi-virtual power plant alliance.

[0050] In this embodiment, if Figure 5 As shown, step S4 also includes: Step S41: decompose the collaborative optimization problem into multiple sub-problems, and optimize the decision variables of each virtual power plant respectively; The ADMM algorithm decomposes the collaborative optimization problem into a constrained optimization problem in the form of multiple sub-problems: ; In the formula is the optimization variable, and is a convex function. It is a linear constraint that describes the coupling relationship between variables.

[0051] Step S42: alternately updating the decision variables to gradually approach the optimal solution; The iterative process of ADMM usually includes the following three steps (in iterations): 1. -Minimize step; fix and , solve about Sub-problems: ; This sub-problem usually only involves Related objective function and constraints.

[0052] 2. -Minimize step; fix and , solve about Sub-problems: ; This sub-problem usually only involves Related objective function and constraints.

[0053] 3. Dual variable update step; use and Update the dual variables : ; These three steps are performed alternately until the preset convergence criterion is met, that is, the primal residual and the dual residual are small enough.

[0054] Step S43: Couple the decision variables through the augmented Lagrangian function, and alternately update the original variables and the dual variables; wherein the augmented Lagrangian function is: ; in, is the Lagrange multiplier (dual variable), is the augmentation penalty parameter.

[0055] Step S44: dynamically adjust the penalty factor of the augmented Lagrangian function to accelerate convergence.

[0056] In this embodiment, the collaborative optimization problem is decomposed into multiple sub-problems through the ADMM algorithm, and the variables are updated alternately to gradually approach the optimal solution of the collaborative optimization problem. Then, the stability and convergence of the ADMM algorithm are improved by introducing the augmented Lagrangian function.

[0057] Step S5: Introduce cooperative game theory to solve the profit distribution of each virtual power plant in the multi-virtual power plant alliance, and use the Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair profit distribution plan.

[0058] As an optional embodiment, Figure 6 As shown, the Shapley value in step S5 quantifies the marginal contribution of each virtual power plant, including the following steps: step S51, traverse all virtual power plant alliances combinations, and calculate the characteristic function value of each combination; step S52, determine the marginal contribution of each virtual power plant based on the characteristic function value; step S53, determine the Shapley value of each virtual power plant based on the weighted sum of marginal contributions, and distribute the total revenue in the multi-virtual power plant alliance according to the Shapley value ratio.

[0059] In this embodiment, there are two ways to calculate the Shapley value. The first way is to calculate all possible combinations of multiple VPP alliances. The characteristic function values ​​of combinations (including empty combinations) ; Each time for a different VPP alliance Then, for each participating VPP , traverse all the Alliance , calculate its marginal contribution Finally, according to the formula (which is an existing theory and will not be repeated here), the weighted sum is used to obtain the respective Shapley values, and the total benefits in the multi-virtual power plant alliance are distributed according to the Shapley value ratio. The second calculation method: by considering all The possible permutations of participating VPPs to join the Grand Alliance are calculated. , participate in VPP The marginal contribution is Participate in VPP The Shapley value of The average marginal contribution under the various arrangements.

[0060] The calculation formula is: .

[0061] This embodiment provides a virtual power plant optimization scheduling method based on electricity-carbon trading, specifically including: The virtual power plant optimization scheduling method of this embodiment specifically establishes a tiered trading mechanism for the carbon market and uses the SOS2 method to accurately linearize the carbon emission cost. The alternating multiplier method is used to perform scheduling analysis under electricity-carbon coupling on a multi-virtual power plant alliance. An objective function is established to minimize the total operating cost including the carbon emission cost, thereby achieving the coupling of carbon price and electricity price, realizing effective energy management of the virtual power plant, and achieving the effect of peak shaving and valley filling.

[0062] Example 2 This embodiment provides a virtual power plant optimization scheduling device 10 based on electricity-carbon trading, such as Figure 7 As shown, it specifically includes: a model building module 11, a cost linear module 12, an operating cost module 13, a collaborative optimization module 14 and a benefit distribution module 15.

[0063] The model building module 11 is used to build an optimized scheduling framework for multiple virtual power plants and establish a refined mathematical model of a single virtual power plant based on the optimized scheduling framework.

[0064] In this embodiment, if Figure 2 As shown in the figure, the optimized dispatch framework includes transactions in the electricity and carbon markets, as well as cooperative game-playing among multiple virtual power plants. The electricity and carbon markets operate according to their optimal strategies and upload transaction information to the electricity-carbon market service platform. Based on transaction information and the operating status of the upstream power grid, information such as energy prices and carbon quota prices is generated to guide the optimized operation of multiple virtual power plants participating in the electricity and carbon markets. Virtual power plant operators report electricity trading volumes and prices. Through data interaction and iterative calculations, the final energy trading results are determined, and corresponding profit distribution plans are formulated. The final transaction results are then transmitted to each virtual power plant via a communication network. After receiving this information, each virtual power plant reports its electricity and carbon quota trading needs through the electricity-carbon market and reaches a consensus on the transaction, fully leveraging the complementary advantages of cross-regional resources.

[0065] In this embodiment, the refined mathematical model of a single virtual power plant includes a distributed photovoltaic output model, a gas turbine operation mathematical model, an energy storage system operation mathematical model, and a virtual power plant and main grid power interaction model.

[0066] As an optional embodiment, the objective function formula of the distributed photovoltaic output model is as follows: ; Where, for The maximum available power of the photovoltaic system during the period, is the comprehensive conversion efficiency of the photovoltaic array, is the total area of ​​the photovoltaic array, for Solar radiation intensity during the period (kW / m²), is the power temperature coefficient of the solar panel, for The actual operating temperature of the solar panel during the period, is the reference temperature; In this embodiment, the distributed photovoltaic output model requires some constraints during calculation. Each constraint is expressed as a function: The photovoltaic dispatch output constraint function is expressed as: ; Where, for The actual power of photovoltaic power that is connected to the grid or used internally by the VPP during the period; Due to the uncertainty of photovoltaic prediction, the PV output in the future can be more accurately predicted through weather forecasts and statistical models to obtain the predicted output curve. ; Therefore, the amount of abandoned light It can be expressed as: .

[0067] As an optional embodiment, the following constraints are required when calculating the mathematical model for the operation of the energy storage system. Each constraint is expressed as a function: The output constraint can be expressed as: ; In the formula for Output power of MT during the time period, for The running status of the MT in the period, where 0 is shutdown and 1 is startup, is a binary decision variable. and They are the minimum and maximum technical output of MT respectively.

[0068] The climbing constraint can be expressed as: ; ; in, and They are the maximum up-climbing rate and down-climbing rate of MT respectively.

[0069] The fuel cost can be expressed as: ; in, is the fuel cost coefficient of MT, Can represent fixed operating costs or no-load consumption costs.

[0070] The start-stop cost can be expressed as: ; ; In the formula and are the single startup and shutdown costs, respectively. In MILP, this is usually represented accurately by introducing additional binary variables.

[0071] In this embodiment, the emission of the micro gas engine is generally proportional to its fuel consumption. Thus, the emission can be expressed as a function of its output power, which is usually simplified to a linear relationship. The specific expression is as follows: ; in, is the carbon emission intensity of MT power generation (kg / kWh or t / kWh), It is the fixed carbon emissions during MT operation.

[0072] As an optional embodiment, the objective function formula of the mathematical model for the energy storage system operation is: ; In the formula for The state of charge of the energy storage system at the end of the period (usually expressed as a decimal between 0 and 1, or a percentage), and They are The charging and discharging power of the time-slot energy storage system, and are the charging and discharging efficiencies of the energy storage system, is the rated capacity of the energy storage system (kWh), is the length of the scheduling time interval (h).

[0073] The SOC constraint formula is: ; In the formula and are the minimum and maximum states of charge allowed by the energy storage system.

[0074] The charge and discharge constraint formula is: ; ; ; in and are the maximum charging power and maximum discharging power of the energy storage system respectively. and are binary variables, representing The charging and discharging status of the time period (0 for no, 1 for yes). The third formula in the charge and discharge constraint formula is used to ensure that the energy storage system cannot be charged and discharged at the same time in the same time period.

[0075] The initial and final SOC simulation setting formulas are: ; In order to ensure the sustainability of the scheduling cycle, The operating cost of energy storage is primarily considered in terms of cycle life losses. In short-term optimization, this is sometimes simplified to charge and discharge efficiency losses, or indirectly considered by limiting the number of daily charge and discharge cycles or total throughput.

[0076] The formula for including cost in the objective function is: ; In the formula It is the aging cost per unit charge and discharge capacity.

[0077] As an optional embodiment, the calculation of the interaction model between the virtual power plant and the main grid requires some constraints, and each constraint is expressed as a function: The power constraint formula for purchasing and selling electricity is: ; ; In the formula and are the maximum power of electricity purchased from and sold to the main grid by the VPP respectively. and are binary variables, representing The electricity purchasing and selling status of the time period.

[0078] The mutually exclusive constraint formula for electricity purchase and sale is: ; The cost / benefit formula for electricity purchase and sales is: ; in, It is the time period The price of electricity purchased by VPP from the main grid ($ / kWh), It is the time period The price of electricity sold by the VPP to the main grid ($ / kWh), is the length of the scheduling period in hours.

[0079] The formula for the net exchange power with the main grid is: ; In the formula Positive values ​​indicate net electricity purchases, and negative values ​​indicate net electricity sales: The formula for carbon emissions related to electricity purchase is: ; in for The average carbon emission factor of the power grid during the period (kg / kWh or t / kWh).

[0080] The cost linearization module 12 is used to establish a ladder trading mechanism in the carbon market and use the second type of special ordered set technology to accurately linearize the cost of carbon emissions.

[0081] As an optional embodiment, Figure 5 As shown, the cost linear module 12 further includes: The multi-tiered submodule 121 is configured to divide excess carbon emissions into multiple tiers, with the carbon price of each tier increasing with the number of tiers. In this embodiment, the multiple tiers include a carbon-free cost, a fixed carbon price, a stepped carbon price, and a piecewise linear continuous carbon price.

[0082] The linear constraint submodule 122 is configured to transform the nonlinear function of the carbon emission cost into a linear constraint using SOS2.

[0083] In this embodiment, if Figure 8 As shown, the linear constraint submodule also includes: First, define the inflection point that describes the relationship between cumulative carbon costs and cumulative excess carbon emissions sequence, where . Among them, the definition of the inflection point coordinates is: The 0th inflection point (origin) is .

[0084] For the subsequent Inflection point ( ), its coordinate calculation formula is as follows: ; ; Where, Before filling The cumulative excess emissions during each stage, Indicates the corresponding cumulative carbon cost at this time.

[0085] Secondly, we introduce non-negative weight variables , and impose two types of special ordered set constraints; finally, the net excess carbon emissions are calculated through convex combination and the corresponding total carbon cost ; Among them, the SOS2 linearization constraint formula is: ; ; ; Where, SOS2 constraints are imposed on this set of weight variables:

[0086] In the formula, the SOS2 constraint stipulates that: in the set In , there can be at most two variables with non-zero values, and if there are two non-zero values, they must be adjacent. This clever constraint ensures that the point Must fall between two adjacent inflection points and This perfectly reproduces the piecewise linear cost function relationship without introducing additional binary variables for each step.

[0087] The operating cost module 13 is used to establish a minimization model of the total operating cost of a multi-virtual power plant alliance including carbon emission costs based on a ladder trading mechanism.

[0088] In this embodiment, the total operating cost of the internal optimization scheduling model of a single VPP in the operating cost module 13 is the sum of the operating costs of grid power exchange, microturbine (MT) power generation, and energy storage system (ESS). Its objective function is expressed as: ; Where, is VPPi in cycle Total operating costs within.

[0089] The objective function of the model for minimizing the total operating cost of a multi-virtual power plant alliance is:

[0090] Where, It is the total operating cost of the VPP alliance.

[0091] The collaborative optimization module 14 is used to solve the collaborative optimization problem of multiple virtual power plant alliances by using the alternating direction multiplier method.

[0092] In this embodiment, if Figure 9 As shown, the collaborative optimization module 14 also includes: The problem decomposition submodule 141 is used to decompose the collaborative optimization problem into multiple sub-problems and optimize the decision variables of each virtual power plant respectively; In this embodiment, the ADMM algorithm decomposes the collaborative optimization problem into a constrained optimization problem in the form of multiple sub-problems: ; In the formula is the optimization variable, and is a convex function. It is a linear constraint that describes the coupling relationship between variables.

[0093] A variable updating submodule 142 is configured to alternately update the decision variables to gradually approach the optimal solution; The iterative process of ADMM usually includes the following three steps (in iterations): 1. -Minimize step; fix and , solve about Sub-problems: ; This sub-problem usually only involves Related objective function and constraints.

[0094] 2. -Minimize step; fix and , solve about Sub-problems: ; This sub-problem usually only involves Related objective function and constraints.

[0095] 3. Dual variable update step; use and Update the dual variables : ; These three steps are performed alternately until the preset convergence criterion is met, that is, the primal residual and the dual residual are small enough.

[0096] The function coupling submodule 143 is used to couple the decision variables through the augmented Lagrangian function and alternately update the original variables and the dual variables; wherein the augmented Lagrangian function is: ; in, is the Lagrange multiplier (dual variable), is the augmentation penalty parameter.

[0097] The convergence stabilization submodule 144 is used to dynamically adjust the penalty factor of the augmented Lagrangian function to accelerate convergence.

[0098] In this embodiment, the collaborative optimization problem is decomposed into multiple sub-problems through the ADMM algorithm, and the variables are updated alternately to gradually approach the optimal solution of the collaborative optimization problem. Then, the stability and convergence of the ADMM algorithm are improved by introducing the augmented Lagrangian function.

[0099] The benefit distribution module 15 is used to introduce cooperative game theory to solve the benefit distribution of each virtual power plant in the multi-virtual power plant alliance, and use the Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair benefit distribution plan.

[0100] In this embodiment, there are two ways to calculate the Shapley value. The first way is: the profit distribution module 15 calculates all possible combinations in the multi-VPP alliance. The characteristic function values ​​of combinations (including empty combinations) ; Each time for a different VPP alliance Then, for each participating VPP , traverse all the Alliance , calculate its marginal contribution Finally, according to the formula (which is an existing theory and will not be repeated here), the weighted sum is used to obtain the respective Shapley values, and the total benefits in the multi-virtual power plant alliance are distributed according to the Shapley value ratio. The second calculation method: by considering all The possible permutations of participating VPPs to join the Grand Alliance are calculated. , participate in VPP The marginal contribution is Participate in VPP The Shapley value of The average marginal contribution under the various arrangements.

[0101] The calculation formula is: .

[0102] This embodiment provides a virtual power plant optimization and scheduling device based on electricity-carbon trading. This virtual power plant optimization and scheduling device is based on the virtual power plant optimization and scheduling method based on electricity-carbon trading in Example 1. Through a stepped carbon trading mechanism, SOS2 linearization, and Shapley value allocation, this virtual power plant optimization and scheduling device achieves economic low-carbon scheduling and fair benefit distribution across multiple VPP alliances, providing an effective solution for power system optimization under the "dual carbon" goal.

[0103] Example 3 This embodiment relates to a computer-readable storage medium storing a computer program and installed in a kotatsu. When the computer program is executed by a processor, the method for optimizing virtual power plant scheduling based on electricity-carbon trading described in the first embodiment is implemented.

[0104] That is, those skilled in the art will understand that all or part of the steps in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a program. The program is stored in a storage medium and includes a number of instructions for causing a device (which may be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes: a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., various media that can store program code.

[0105] With the above-described preferred embodiments of the present invention as inspiration, and with reference to the above description, relevant personnel may make various changes and modifications without departing from the scope of the present invention. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A virtual power plant optimization scheduling method based on electricity-carbon trading, characterized in that: The following steps are involved: Step S1: constructing an optimized scheduling framework for a multi-virtual power plant alliance, and establishing a refined mathematical model of a single virtual power plant based on the optimized scheduling framework; Step S2: Establish a ladder trading mechanism for the carbon market and use the second type of special ordered set technology to accurately linearize the carbon emission cost; Step S3: establishing a model for minimizing the total operating cost of the multi-virtual power plant alliance including the carbon emission cost based on the tiered trading mechanism; Step S4: solving the collaborative optimization problem of the multi-virtual power plant alliance using an alternating direction multiplier method; Step S5: Introducing cooperative game theory to solve the profit distribution of each virtual power plant in the multi-virtual power plant alliance, and using Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair profit distribution plan; The optimization scheduling framework includes transactions in the electricity market and the carbon market, and cooperative games among multiple virtual power plants.

2. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 1 is characterized in that: The step S2 further includes: Step S21: Divide the excess carbon emissions into multiple levels, with the carbon price of each level increasing with the number of levels; Step S22: using the second type of special ordered set technology to convert the nonlinear function of the carbon emission cost into a linear constraint; The multiple steps include carbon-free cost, fixed carbon price, stepped carbon price and piecewise linear continuous carbon price.

3. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 2 is characterized in that: The step S22 further includes: Step S221: defining an inflection point sequence describing the relationship between cumulative carbon costs and cumulative excess carbon emissions; Step S222: introducing non-negative weight variables and applying the second type of special ordered set constraints; Step S223: Calculate the net excess carbon emissions and the corresponding total carbon cost through convex combination.

4. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 1 is characterized in that: The objective function of the total operating cost minimization model of the multi-virtual power plant alliance in step S3 is: ； Where, is the total operating cost of the multi-virtual power plant alliance.

5. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 1 is characterized in that: The step S4 further includes: Step S41: decompose the collaborative optimization problem into multiple sub-problems, and optimize the decision variables of each virtual power plant respectively; Step S42: alternately updating the decision variables to gradually approach the optimal solution; Step S43: coupling the decision variables through the augmented Lagrangian function, and alternately updating the original variables and the dual variables; Step S44: dynamically adjust the penalty factor of the augmented Lagrangian function to accelerate convergence.

6. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 1 is characterized in that: The step of quantifying the marginal contribution of each virtual power plant by the Shapley value in step S5 further includes: Step S51: traverse all virtual power plant alliances combinations, calculating the characteristic function value of each of the combinations; Step S52: determining the marginal contribution of each virtual power plant based on the characteristic function value; Step S53: Determine the Shapley value of each virtual power plant based on the weighted sum of the marginal contributions, and distribute the total revenue in the multi-virtual power plant alliance according to the proportion of the Shapley value.

7. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 1 is characterized in that: The refined mathematical model of a single virtual power plant includes a distributed photovoltaic output model, a gas turbine operation mathematical model, an energy storage system operation mathematical model, and an interaction model between the virtual power plant and the main grid electricity.

8. The virtual power plant optimization scheduling method based on electricity-carbon trading according to claim 7 is characterized in that: The carbon emission calculation formula of the gas turbine for establishing the mathematical model of gas turbine operation is as follows: ； in, is the carbon emission intensity of MT power generation (kg / kWh or t / kWh), It is the fixed carbon emissions during MT operation.

9. A virtual power plant optimization scheduling device based on electricity-carbon trading, characterized in that: Specifically include: A model building module is used to construct an optimized scheduling framework for multiple virtual power plants and establish a refined mathematical model of a single virtual power plant based on the optimized scheduling framework; Cost linearization module, used to establish a ladder trading mechanism for the carbon market, using the second type of special ordered set technology to accurately linearize the cost of carbon emissions; An operating cost module, configured to establish a minimization model for the total operating cost of the multi-virtual power plant alliance including the carbon emission cost based on the tiered trading mechanism; a collaborative optimization module, configured to solve the collaborative optimization problem of the multi-virtual power plant alliance by adopting an alternating direction multiplier method; a benefit distribution module, configured to introduce cooperative game theory to solve the benefit distribution of each virtual power plant in the multi-virtual power plant alliance, and use Shapley value to quantify the marginal contribution of each virtual power plant to provide a fair benefit distribution plan; The optimization scheduling framework includes transactions in the electricity market and the carbon market, and cooperative games among multiple virtual power plants.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the virtual power plant optimization scheduling method based on electricity-carbon trading according to any one of claims 1 to 8 is implemented.

Images