Day-ahead spot market clearing optimization method considering carbon transaction and virtual power plant dimension reduction

Abstract

The invention belongs to the technical field of power market and low-carbon clearing, and discloses a day-ahead spot market clearing optimization method considering carbon transaction and virtual power plant dimensionality reduction, and the method comprises the steps: firstly constructing a virtual power plant aggregation equivalent dimensionality reduction model, and representing adjustment potential and protecting distributed resource privacy through extracting key operation boundary parameters; secondly, establishing a unit carbon emission quota accounting and carbon transaction cost calculation model based on a regional power grid carbon emission factor; fusing the above models, and establishing a multi-flexibility resource two-stage robust optimization clearing model containing thermal power generating unit start and stop, load reduction and virtual power plant adjustment capability, in the first stage, making a basic scheme with the minimum pre-clearing cost, and in the second stage, performing plan adjustment with the minimum regulation and control risk cost for an extreme scene; finally, the model is converted into a mixed integer linear programming problem, and Camp is adopted; and a CG algorithm is used for efficiently solving.

Description

Technical Field

[0001] This invention belongs to the field of power system operation and control technology, specifically involving a method for optimizing the low-carbon clearing of the day-ahead spot market of the power system by considering carbon trading mechanisms and using virtual power plant dimensionality reduction models to handle uncertainties. It is particularly suitable for new power systems with a high proportion of renewable energy access. Background Technology

[0002] To proactively address climate change, China is accelerating the construction of a new power system dominated by new energy sources. As a core component of power system reform, the electricity spot market's efficient and fair clearing mechanism is crucial for optimizing resource allocation, promoting the integration of new energy sources, and reducing overall system carbon emissions. Simultaneously, the establishment and operation of the national carbon emissions trading market explicitly introduces carbon costs into the power production process, becoming a key economic lever for guiding power structure optimization and promoting emission reduction. With the explosive growth of distributed energy resources (such as rooftop solar, small-scale wind power, energy storage, and adjustable loads), the core challenge lies in how to efficiently and cost-effectively aggregate massive distributed resources (especially low-carbon / zero-carbon resources) to participate in a competitive spot market, and fully reflect the impact of carbon trading costs on market clearing prices and resource dispatch priorities. Virtual power plants (VPPs) are an important technological vehicle for addressing this challenge.

[0003] However, current electricity spot market clearing models face a dual challenge in accommodating large-scale VPP participation and effectively reflecting carbon trading mechanisms. On the one hand, VPPs contain a vast number of diverse, spatiotemporally complex, and highly uncertain distributed resources with varying output and demand characteristics. Directly and meticulously characterizing the physical constraints and uncertainties of all resources within each VPP in a system-level market clearing model would lead to a dramatic expansion of the optimization problem's dimensionality ("curse of dimensionality"), making the model overly complex and computationally time-consuming, failing to meet the real-time requirements of spot market clearing. On the other hand, existing clearing models, when considering VPPs, often fail to effectively distinguish and quantify the carbon cost differences between different types of resources (especially fossil fuel units and renewable energy) and their impact on market equilibrium. Existing methods typically oversimplify VPPs (e.g., treating them as a single equivalent unit or load), neglecting their internal flexible adjustment capabilities and uncertainties, and failing to accurately reflect their potential low-carbon value (e.g., aggregated renewable energy can replace high-carbon electricity) and their economic competitiveness in a carbon-constrained market environment. This could lead to market clearing results that fail to optimally guide the consumption of low-carbon resources, suppress high-carbon power generation, or even distort carbon cost transmission signals, affecting the achievement of emission reduction targets and market efficiency.

[0004] To address these challenges, there is an urgent need to develop market clearing methods that can simultaneously and efficiently handle the complexity of VPPs and accurately reflect carbon trading costs. The core idea is twofold: First, through advanced dimensionality reduction modeling techniques, key operational characteristics of VPPs relative to external systems (grid and market) (such as aggregated power range, ramp-up capability, cost characteristics, and uncertainty boundaries) are extracted to construct a concise equivalent model. Second, the carbon emission attributes or low-carbon contributions of the VPP's internal resources must be clearly embedded into this equivalent model, enabling it to respond accurately to carbon price signals in the clearing model, just like other traditional generating units in the system.

[0005] Therefore, this invention focuses on researching a market clearing method for the electricity spot market that takes into account virtual power plant dimensionality reduction models and carbon trading costs. The aim is to develop a market clearing framework that significantly improves computational efficiency and solution feasibility while accurately quantifying the impact of VPP participation on system carbon emissions and carbon costs. This method is expected to optimize market resource allocation, incentivize the development of low-carbon VPPs, and more effectively achieve carbon cost transmission, thereby supporting the safe, economical, and low-carbon operation of new power systems. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention proposes a day-ahead spot market clearing optimization method that considers carbon trading mechanisms and virtual power plant dimensionality reduction models. Based on robust optimization theory, and coupling carbon trading costs with the aggregation characteristics of virtual power plants, a novel two-stage day-ahead robust clearing model for the power system, incorporating multiple types of flexible resources, is constructed. By defining the objective function and constraints and integrating them into a complete matrix expression, this method can effectively cope with extreme scenarios of wind power, photovoltaic output, and load demand, reducing carbon emissions and operating costs while ensuring stable system operation.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] An optimization method for day-ahead spot market clearing that takes into account carbon trading and virtual power plant dimensionality reduction models is characterized by the following steps:

[0009] Step 1: Construct a multi-market entity model of the power system including virtual power plants, and use a dimensionality reduction method to perform equivalent modeling of the virtual power plants;

[0010] Step 2: Based on the emission source characteristics of each market entity and the standard carbon emission factors issued by the National Energy Administration, establish a carbon emission calculation model on the power supply side, couple it to form a unit carbon emission quota allocation model, and construct a system-level carbon trading cost model accordingly.

[0011] Step 3: Construct a robust two-stage clearing optimization model for day-ahead spot trading that takes into account virtual power plant dimensionality reduction and carbon trading;

[0012] Step 4: Organize the objective function and constraints of the two-stage robust optimization clearing model of the power system that takes into account carbon trading and virtual power plant dimensionality reduction model into matrix form for description, and solve it to realize the low-carbon economic operation of the new power system.

[0013] This technical solution is further optimized by the day-ahead spot market clearing optimization method that takes into account carbon trading and virtual power plant dimensionality reduction models. The objective function in the pre-clearing stage is to minimize the operating cost of the day-ahead clearing scheme, and the optimization decision variables include the start-up and shutdown plans of thermal power units. Power generation plan Upward rotation of standby capacity scheme and downward rotation spare capacity scheme It also includes load reduction plans for Category A loads that can be reduced. , Compensation unit price for Category A load reduction during time period Category B reserve capacity that can reduce load , Compensation unit price for load reduction in Category B time period Clearing plan for loads that can be shifted Compensation price for transferable loads and system carbon emissions Charging and discharging plans for energy storage in virtual power plants and Charge and discharge capacity and Load reduction plans ;in addition For 0-1 variables, express Traditional energy generators during the period The medium-sized generating unit has been started and is in operation. express Traditional energy generators during the period The middle unit was shut down; for Traditional energy generators during the period The generating capacity of the medium-sized generating unit; and for Traditional energy generators during the period The reserve capacity for both upward and downward travel of the generating units; The compensation price for loads that can be moved during time period t; and For 0-1 variables, =1 and =1 indicates that the energy storage in virtual power plant x is charging and discharging during time period t. =0 and =0 indicates that the energy storage in the virtual power plant x during time period t is neither charged nor discharged; This represents the charging and discharging power of the virtual power plant x during time period t; This indicates the amount of load reduction that can be achieved in virtual power plant x during time period t; the objective of the formal clearing phase adjustment plan is to minimize regulation risk costs, and specific adjustments include the power adjustment plan for thermal power units. Category B load reduction plans wind curtailment Wasted light Involuntary load shedding by load-side users .

[0014] This technical solution is further optimized, and step 1 specifically includes:

[0015] Step 1.1: Establish a cost model for traditional energy power generators:

[0016] The operating costs of traditional energy generators are shown in the following formula:

[0017]

[0018] In the formula, , , and They are conventional units Operating costs, fuel consumption costs, start-up and shutdown costs, and standby costs.

[0019] The price quoted by traditional energy power generators for each segment is:

[0020]

[0021] In the formula: This represents the price quoted by traditional energy generator k in the nth segment; This indicates the power generation capacity of the traditional energy generator k; This represents the initial power generation capacity of traditional energy generator k in segment n; This represents the marginal cost of generating electricity for traditional energy generator k; The power generation capacity of traditional energy generator k during time period t; K is the number of all traditional energy generators. The total number of segments in the volume and price curves declared by traditional energy power generators.

[0022] Step 1.2: Establish a load reduction model:

[0023] For load shedding, load aggregators receive the adjusted compensation price from the grid, assess the response potential of load users, and then resubmit the maximum load shedding amount to be provided for the next 4 hours. The electricity spot market trading center formulates a plan based on the demand for specific time periods within an adjustable range. After reaching an agreement, the load shedding responds to grid dispatch at the precise time. Therefore, the load shedding model can be expressed as follows:

[0024]

[0025]

[0026]

[0027] In the formula: for The amount of load reduction that can be implemented in period A; The elasticity coefficient for Category A load reduction and market participation is... ; The supplementary price per unit power reduction for Category A load reduction capacity; for The maximum load reduction allowed for Category A time period; and These represent the compensation prices at which Class A users who can reduce load can just obtain the benefit and when the response volume is maximized, respectively. The maximum elasticity coefficient for load reduction; For load compensation sensitivity, The smaller the value, the more sensitive users are to the compensation price, and the greater the impact of price changes on users. for The cost of reducing load during certain periods; This refers to the day-ahead spot market clearing interval. Reduceable loads are divided into two categories, A and B. These refer to loads that can be reduced in usage without affecting daily needs; the interruption time and duration of power usage can be shortened or increased. Category A loads are similar to shiftable loads, with a relatively slow response time. A pre-planned dispatch schedule is required; ad-hoc decisions are not allowed, and the established schedule must be strictly followed without arbitrary changes. Category B loads have rapid adjustment capabilities and can quickly respond to grid dispatch demands. Therefore, this article develops a reduction plan for them during the day-ahead spot market clearing. The dispatch methods for Category A and Category B reduceable loads are basically the same; the dispatch method for Category B reduceable loads will not be detailed here.

[0028] Step 1.3: Establish a transferable load model:

[0029] When the compensation price published by the electricity spot market trading center is low, the benefits gained by users from changing their electricity usage time are insufficient to compensate for the inconvenience and losses caused by shifting electricity consumption. Consequently, users' enthusiasm for participating in the market will be very low, making it difficult to meet the needs of the electricity spot market trading center. Based on the principles of consumer psychology, the shiftable load model can be expressed as:

[0030]

[0031] In the formula: The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; The compensation price per unit power for the shiftable load.

[0032] Step 1.4: Establish a virtual power plant model:

[0033] Virtual power plants integrate distributed energy resources (such as solar and wind power), energy storage systems, and controllable loads (such as electric vehicles and adjustable industrial and commercial equipment) to form flexible and controllable capabilities similar to traditional power plants, enabling them to participate in day-ahead electricity spot market transactions. A typical virtual power plant aggregates distributed resources such as solar power, load shedding, and energy storage. Therefore, virtual power plant x declares the predicted power output of the aforementioned solar power plant in time period t. Unit charge / discharge power of independent energy storage and Quota capacity Compensation price for load reduction declaration Related parameters participate in the clearing of the electricity spot market. Therefore, the operating cost of the virtual power plant... Represented as:

[0034]

[0035] In the formula: For the energy storage discharge revenue in virtual power plant x, The cost of energy storage charging in a virtual power plant x The battery aging cost in the energy storage of the virtual power plant x. This can reduce load costs in virtual power plants x.

[0036] Step 1.5: Dimensionally reduced representation of the marginal cost of the virtual power plant:

[0037] Precise initial parameter space definition: Establish a multi-period optimization model of the virtual power plant and define the initial parameter space; based on the cost of the virtual power plant... A mathematical model is constructed with minimization as the objective, considering distributed energy sources such as distributed photovoltaics, energy storage, and load shedding, as well as the power flow constraints within the virtual power plant. The interaction between the virtual power plant and the main grid is parameterized and placed on the right-hand side of the constraints. Simultaneously, all operational constraints of the virtual power plant are systematically reconstructed into matrix inequalities. This process ultimately yields a standard multi-parameter linear programming problem whose objective function aims to minimize the operating cost of the virtual power plant.

[0038]

[0039]

[0040]

[0041] In the formula: All of these are related to the objective function and constraints of the virtual power plant. This is a vector of cost coefficients. The decision variable vector represents the clearing plan for distributed resources such as photovoltaic power plants, energy storage, and load shedding. This is the coefficient matrix of the decision variables in the constraints; The matrix represents the constant terms on the right-hand side of the constraint conditions; The coefficient matrix of the parameter vector; This is a parameter vector representing the electricity traded between the virtual power plant and the main grid at the common connection point.

[0042] Subsequently, the feasible region of interactive power quantities of the virtual power plant at the common junction point is characterized based on convex set projection theory. Then, based on multi-parameter programming theory, the critical region is characterized by identifying effective and ineffective constraints. Finally, the marginal cost of the virtual power plant at the common junction point with respect to interactive power quantities is analytically represented based on strong duality theory. The partial derivative of the virtual power plant cost function with respect to parameters within each critical region represents the marginal cost in that region, which can be directly characterized by Lagrange multipliers. Therefore, the dimension-reduced marginal cost of the virtual power plant... It can be characterized as:

[0043]

[0044] In the formula: Let be the net output power of the virtual power plant x at time t, which can be positive or negative; This is the i-th dual solution to the dual problem of the above virtual power plant operating cost function; Let be the 0th-order term of the i-th solution to the dual problem; The i-th critical domain after dimensionality reduction of the virtual power plant.

[0045] This technical solution is further optimized, and step 2 specifically includes:

[0046] Step 2.1: Establish a real-time carbon emission calculation method for the power system:

[0047] Traditional energy power generators medium-sized coal-fired power generating units Carbon emission intensity over time period It can be dynamically calculated based on the fuel traceability model, as shown in the following formula:

[0048]

[0049] In the formula: For traditional energy power generators Carbon oxidation rate during coal combustion in coal-fired power generation units; Carbon emission factor for the type of coal burned in coal-fired power units; For traditional energy power generators Coal loss per kilowatt-hour produced by a coal-fired power unit; defining the unit's coal loss per kilowatt-hour produced. The carbon emission intensity column vector for the time period is Then its first The elements are Wherein, the carbon oxidation rate Based on dynamic adjustments to boiler operating conditions, coal type carbon emission factors It can be determined through laboratory calorific value testing.

[0050] Once the carbon emission intensity of the generating units is calculated, the carbon emissions of each unit can be calculated based on its power generation. Cumulative carbon emissions over a period of time The calculation formula is

[0051]

[0052] Step 2.2: Establish a method for allocating carbon emission allowances for the power system:

[0053] The baseline method first calculates the carbon emission factor based on the relationship between energy consumption and greenhouse gas emissions by the National Energy Administration, and then allocates resources based on the carbon emission factor. This method provides overall control and is more conducive to incentivizing various resources in the power grid to participate in emission reduction.

[0054] This method uses a baseline approach to allocate free carbon emission allowances. The free carbon emission allowance for units in the time-of-use system is

[0055]

[0056] In the formula: This is the system's free quota; The free allowance per unit of power is determined by the average emission factor of Anhui Province as calculated by the state. tCO2 / MWh;

[0057] Step 2.3: Establish a method for calculating the carbon trading costs of the power system:

[0058] This method employs a tiered pricing strategy to set carbon trading prices. This strategy ensures a reasonable gradient relationship between CO2 trading prices and carbon emissions, thereby more effectively guiding enterprises to reduce carbon emissions and achieve sustainable development. The carbon trading cost model is as follows;

[0059]

[0060] In the formula: The carbon trading price is tiered. The market benchmark price; The price growth rate is tiered. This represents the length of the price tier range.

[0061]

[0062] In the formula: For the system The cost of carbon trading during a given period.

[0063] This technical solution is further optimized by establishing a two-stage robust optimization model for the day-ahead spot market in step 3. The model includes two stages: pre-clearing and formal clearing, and specifically includes the following steps:

[0064] Step 3.1: Establish the objective function of the two-stage robust optimization model for the day-to-day spot market:

[0065] The objective of the day-ahead robust optimization clearing model, which takes into account carbon trading and virtual power plant dimensionality reduction models, is to minimize the sum of the costs of the two stages, i.e., the operating cost of the pre-clearing stage. and the regulatory risks and costs during the formal clearing phase sum;

[0066]

[0067] in The current liquidation plan; For 0-1 variables, express Traditional energy generators during the period The medium-sized generating unit has been started and is in operation. express Traditional energy generators during the period The middle unit was shut down; for Traditional energy generators during the period The generating capacity of the medium-sized generating unit; and for Traditional energy generators during the period The reserve capacity for both upward and downward travel of the generating units; The unit power compensation price for load reduction during period t for categories A and B; The compensation price for loads that can be moved during time period t; and For 0-1 variables, =1 and =1 indicates that the energy storage in virtual power plant x is charging and discharging during time period t. =0 and =0 indicates that the energy storage in the virtual power plant x during time period t is neither charged nor discharged; This represents the charging and discharging power of the virtual power plant x during time period t; This represents the load reduction that can be reduced in the virtual power plant x during time period t; traditional energy power generators in the formal clearing phase. Power adjustment plan of medium-sized generating units during time period t Category B load reduction plan for period t The amount of wind curtailment during time period t during the formal clearing phase. Wasted light Involuntary load shedding by load-side users ; These are the uncertain parameters for the new energy sources and the load in the problem; The optimization variables during the formal clearing phase; the operating costs during the pre-clearing phase. This includes the various sub-costs of the recently cleared-out plan, namely, unit operating costs. , for The cost of load reduction in Category A time period for Costs of load reduction in Category B time period, and transaction costs of load shifting. Carbon trading costs The cost of virtual power plant x :

[0068]

[0069] In the formula: , This represents the number of virtual power plants.

[0070] Regulatory risk costs during the formal clearing phase Including the cost of wind curtailment The cost of abandoning light and loss of load cost :

[0071]

[0072] In the formula: T is the number of time periods in the entire clearing cycle. As for the clearing interval, there are a total of Each decision-making period.

[0073] Step 3.2: Establish robust optimization model control stage constraints:

[0074] Operational constraints during the pre-clearing phase:

[0075] The constraints of the day-ahead pre-clearing phase of the new power system that takes carbon trading into account include flexible load constraints, power balance constraints, line transmission restrictions, and system reserves.

[0076] The relevant constraints for flexible loads are as follows:

[0077]

[0078]

[0079]

[0080] In the formula: The elasticity coefficient for Category A load reduction and market participation is... ; for Reserve capacity for load reduction during period B; For Category B, the elasticity coefficient for load reduction and market participation is... ; for The maximum load reduction for Category A time period; for The maximum load reduction for Category B time period; The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; To compensate for the unit price.

[0081] The power balance constraints of the power system are:

[0082]

[0083] In the formula: , The number of wind farms; , The number of photovoltaic power plants; , The number of virtual power plants; For wind farm exist Power forecast values ​​for the time period; For photovoltaic power station exist Power forecast values ​​for the time period; The power prediction value of the photovoltaic power station aggregated in the virtual power plant x during time period t; In order to be in Predicted power within the time period; for Forecast values ​​of loads that can be shifted over time; After load shift Power values ​​for a given time period; Let x be the discharge power of the energy stored in the virtual power plant x during time period t; Let x be the charging power of the energy storage in the virtual power plant during time period t; Let x be the load reduction amount that can be reduced in virtual power plant x during time period t.

[0084] The power system line transmission constraints are:

[0085]

[0086] In the formula: , , , and The units Wind farm Photovoltaic power stations Virtual power plant x and load node exist Time period for the line The power transmission allocation factor; branch road The maximum power; For nodes exist The predicted power for the time period, and ; This represents the total number of nodes in the system.

[0087] The standby capacity constraint is:

[0088]

[0089]

[0090] In the formula: and for The system's backup capacity values ​​for different time periods;

[0091] Operational constraints during the formal clearing phase:

[0092] The constraints for the new power system that takes carbon trading into account during the clearing phase recently include unit-related constraints, power balance constraints, and line transmission restrictions.

[0093] Formal clearing phase unit output upper and lower limit constraints

[0094]

[0095] In the formula: for Time-of-use units The amount of power adjustment; and The units Maximum and minimum output values;

[0096] During the formal clearing phase, the unit ramp-up constraints are as follows:

[0097]

[0098] The system power balance constraint during the formal clearing phase is

[0099]

[0100] In the formula: and For the unit Downward and upward climbing rates; for Time-of-use units The power generation capacity; , and These represent the actual wind power and photovoltaic output values ​​under simulated extreme scenarios during the formal clearing phase, respectively. The simulated actual load demand power value for the formal clearing phase includes both flexible and rigid loads. Corresponding to In extreme scenarios during a certain period of time, the first The variable of wind curtailment power in a wind farm; Corresponding to In extreme scenarios during a certain period of time, the first The variable of curtailed solar power from a single photovoltaic power plant; The variable corresponding to the curtailment power of the photovoltaic power station in the x-th virtual power plant under the extreme scenario of time period t; Corresponding to Involuntary load shedding by users on the load side during a given period; This is a reduction plan for Category B loads during time period t.

[0101] The line transmission power constraint during the formal clearing phase is as follows:

[0102]

[0103] In the formula: for Time period nodes The actual load demand power value.

[0104] This technical solution is further optimized. The objective function and constraints of the two-stage robust optimization clearing model between China and Japan described in step 4 are arranged into matrix form, as shown below:

[0105] Step 4.1: Model Solving Algorithm

[0106] Based on the objective function and constraints of the novel two-stage robust optimization clearing model for the day-ahead of a power system considering carbon trading constructed above, it is described in matrix form. The compact form of the proposed model can be expressed as follows:

[0107]

[0108] In the formula: The operating costs of the pre-clearing phase, The regulatory risks and costs during the formal clearing phase, These are the equality constraints for the pre-clearing phase of the model; These are the inequality constraints for the pre-clearing phase of the model. The equality constraints represent the formal clearing phase. The inequality constraints represent the formal clearing phase.

[0109] Because the model constructed uses a min-max-min structure—minimizing the operating cost in the first stage and minimizing the risk cost under adverse scenarios in the second stage—the solution process is relatively complex. Furthermore, the presence of uncertain variables in the model further complicates the solution process. Therefore, the C&CG algorithm is applied to solve the model. This method decomposes the practical problem into a main problem (MP) and a subproblem (SP), and solves these two problems alternately, gradually approaching the optimal solution. The final solution yields the current day-ahead spot market clearing result. This includes the start-up and shutdown status and output of traditional energy generators at various times, the load that can be reduced or shifted, and the resource output in virtual power plants. The mathematical expressions for the main problem MP and the subproblem SP are as follows:

[0110]

[0111]

[0112] In the formula: The auxiliary variables introduced to replace the subproblems are used to directly obtain a temporary solution. Then, the structure of the subproblems is transformed through duality theory, and the max-min problem is transformed into a max single-layer optimization problem.

[0113] Step 4.2: Solution Algorithm Flow

[0114] Based on the above analysis, the overall solution process can be broken down into the following steps.

[0115] Step 1: Initialize the upper bound of the objective function and lower boundary Number of iterations Let the convergence gap between the upper and lower bounds be . , Set to a small positive number;

[0116] Step Two: Let The initial solution obtained Then, substitute the subproblems to solve for the initial extreme scenario. ;

[0117] Step 3: Extreme scenarios Substituting into the main problem, we obtain the optimal solution. Update the Nether equal ;

[0118] Step 4: Apply the optimal solution from Step 3 Substituting into the subproblem, we obtain the optimal solution. ;make Update the upper boundary for and sum;

[0119] Step 5: Judgment If the condition is met, the process ends; otherwise, it will... Return to step three.

[0120] Unlike existing technologies, the main benefits are as follows: Addressing the challenges of effectively promoting energy conservation and emission reduction in thermal power plants and the absorption of new energy sources, enhancing the flexibility of the power system spot market, and resolving the issue of low participation willingness among diverse distributed entities due to the requirement for virtual power plants to submit full models containing key privacy information such as operating parameters of various distributed resources and internal network structure parameters, a robust optimization clearing method for the day-ahead spot market of the power system, incorporating carbon trading mechanisms and virtual power plant dimensionality reduction models, is proposed. Compared to general power spot market clearing optimization methods, this method enables trading centers to better cope with uncertainties on both the source and load sides, improve grid clearing efficiency, and achieve the three-dimensional characteristics of safety, economy, and low carbon emissions in the clearing results. Simultaneously, the efficient solution using the C&CG algorithm significantly improves the solution rate, performance, and model fit to the actual structure, thereby effectively strengthening the system day-ahead clearing model based on robust optimization and incorporating carbon trading mechanisms and virtual power plant dimensionality reduction models. This enables efficient utilization of the power system's flexible resources and promotes the low-carbon economic operation of the power system. Attached Figure Description

[0121] Figure 1 A schematic diagram of the robust optimized clearing model framework for the day-ahead spot market of the power system;

[0122] Figure 2 A robust clearing optimization flowchart for the power system that takes into account carbon trading and virtual power plant dimensionality reduction models was recently developed.

[0123] Figure 3 Flowchart for solving the dimensionality reduction model of a virtual power plant;

[0124] Figure 4 Here is the flowchart for the C&CG algorithm. Detailed Implementation

[0125] To explain in detail the technical content, structural features, objectives, and effects of the technical solution, the following description is provided in conjunction with specific embodiments and accompanying drawings.

[0126] This invention discloses a robust optimization clearing method for the day-ahead spot market of the power system, incorporating carbon trading and a virtual power plant dimensionality reduction model. It addresses the problems in traditional power spot market clearing, such as the difficulty in effectively coordinating energy conservation and emission reduction of thermal power plants with renewable energy consumption, insufficient flexibility in responding to uncertainties on both the source and load sides, and the requirement for virtual power plants to submit a full model containing detailed operating parameters of various distributed resources and network structure parameters, leading to the exposure of critical privacy information and significantly inhibiting the participation of multiple distributed entities. This method deeply integrates the carbon trading cost mechanism into the market clearing model, using economic levers to guide low-carbon operation of thermal power and promote renewable energy consumption. Simultaneously, it innovatively introduces a virtual power plant dimensionality reduction model, requiring only the submission of aggregated equivalent operating characteristic parameters, efficiently characterizing its regulation capabilities while strictly protecting internal privacy. Based on this, the present invention employs a robust optimization method to construct a system day-ahead spot market clearing model that takes into account carbon emissions and various types of flexible resources (including aggregated virtual power plants), and applies the C&CG algorithm to solve it efficiently. This makes the market clearing results robust in dealing with uncertainties such as wind power, photovoltaic output and load forecasting errors, significantly improving the flexibility, dispatch efficiency and system stability of the electricity spot market. This enables the efficient integration and utilization of system flexible resources and promotes the low-carbon economic operation of the power system.

[0127] Please see Figure 1 The diagram shows a robust optimization clearing model framework for the day-ahead spot market of the power system. In the pre-clearing phase, a preliminary dispatching plan is formulated based on the day-ahead wind power output, photovoltaic power output, and load demand forecasts, with the objective of minimizing clearing costs. Optimization decision variables include the start-up and shutdown plans of thermal power units. Power generation plan A series of variables and parameters are designed to meet the basic operational needs of the power grid and reserve margins for handling uncertainties. The formal clearing phase considers the uncertainties in the forecasting of renewable energy power generation and load demand. By simulating possible power and load values ​​in real-world scenarios, the pre-clearing scheme and the formal clearing process are iteratively solved to obtain renewable energy output and load demand data under extreme scenarios. Based on these extreme scenarios, adjustments are made to the pre-clearing plan with the goal of minimizing regulatory risk costs. Specific adjustment variables include the thermal power unit power adjustment plan. Category B load reduction plans These and other variables ultimately enable the electricity spot market trading center to effectively cope with uncertainties on both the source and load sides, and improve the efficiency and stability of power grid dispatch.

[0128] See also Figure 2 The flowchart shown illustrates the day-ahead robust clearing optimization process for the power system, which incorporates carbon trading and a virtual power plant dimensionality reduction model. The process includes the following steps:

[0129] Step 1: Construct a multi-market entity model of the power system including virtual power plants, and use a dimensionality reduction method to perform equivalent modeling of the virtual power plants;

[0130] Step 1.1: Establish a cost model for traditional energy power generators:

[0131] The operating costs of traditional energy generators are shown in the following formula:

[0132]

[0133] In the formula, , , and Traditional energy generators Operating costs, fuel consumption costs, start-up and shutdown costs, and standby costs;

[0134] The price quoted by traditional energy power generators for each segment is:

[0135]

[0136] In the formula: This represents the price quoted by traditional energy generator k in the nth segment; This indicates the power generation capacity of the traditional energy generator k; This represents the initial power generation capacity of traditional energy generator k in segment n; This represents the marginal cost of generating electricity for traditional energy generator k; The power generation capacity of traditional energy generator k during time period t; K is the number of all traditional energy generators. The total number of segments in the volume and price curves declared by traditional energy power generators.

[0137] Step 1.2: Establish a load reduction model:

[0138] For load shedding, load aggregators receive the adjusted compensation price from the grid, assess the response potential of load users, and then resubmit the maximum load shedding they can provide for the next 4 hours. The electricity spot market trading center formulates a plan based on the demand for specific time periods within an adjustable range. After reaching an agreement, the load shedding responds to grid dispatch at the precise time. Therefore, the load shedding model can be represented by the following formula:

[0139]

[0140]

[0141]

[0142] In the formula: for The amount of load reduction that can be implemented in period A; The elasticity coefficient for Category A load reduction and market participation is... ; The supplementary price per unit power reduction for Category A load reduction capacity; for The maximum load reduction allowed for Category A time period; and These represent the compensation prices at which Class A users who can reduce load can just obtain the benefit and when the response volume is maximized, respectively. The maximum elasticity coefficient for load reduction; For load compensation sensitivity, The smaller the value, the more sensitive users are to the compensation price, and the greater the impact of price changes on users. for The cost of reducing load during certain periods; This refers to the day-ahead spot market clearing interval. Reduceable loads are divided into two categories, A and B. These refer to loads that can be reduced in usage without affecting daily needs; their power consumption can be interrupted, and the duration can be shortened or increased. Category A loads are similar to shiftable loads, with a relatively slow response time. A pre-planned dispatch schedule is required; ad-hoc decisions are not allowed, and the established schedule must be strictly followed without arbitrary changes. Category B loads have rapid adjustment capabilities and can quickly respond to grid dispatch demands. Therefore, this paper formulates their reduction plan during the day-ahead spot market clearing. The dispatch methods for Category A and Category B reduceable loads are basically the same; the dispatch method for Category B reduceable loads will not be detailed here.

[0143] Step 1.3: Establish a transferable load model:

[0144] When the compensation price published by the electricity spot market trading center is low, the benefits gained by users from changing their electricity usage time are insufficient to compensate for the inconvenience and losses caused by shifting electricity consumption. This results in low user participation in the market, making it difficult to meet the needs of the electricity spot market trading center. Based on the principles of consumer psychology, the shiftable load model can be expressed as:

[0145]

[0146] In the formula: The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; The compensation price per unit power for the shiftable load.

[0147] Step 1.4: Establish a virtual power plant model:

[0148] Virtual power plants integrate distributed energy resources (such as solar and wind power), energy storage systems, and controllable loads (such as electric vehicles and adjustable industrial and commercial equipment) to form flexible and controllable capabilities similar to traditional power plants, enabling them to participate in day-ahead electricity spot market transactions. A typical virtual power plant aggregates distributed resources such as solar power, load shedding, and energy storage. Therefore, virtual power plants declare the predicted power output of the aforementioned solar power plants in time period t. Unit charge / discharge power of independent energy storage and Quota capacity Compensation price for load reduction declaration Related parameters participate in the clearing of the electricity spot market. Therefore, the operating cost of the virtual power plant... Represented as:

[0149]

[0150] In the formula: For the energy storage discharge revenue in virtual power plant x, The cost of energy storage charging in a virtual power plant x The battery aging cost in the energy storage of the virtual power plant x. This can reduce load costs in virtual power plants x.

[0151] Step 1.5: Dimensionally reduced representation of the marginal cost of the virtual power plant:

[0152] See also Figure 3 The flowchart for solving the virtual power plant dimensionality reduction model is shown below, and the specific steps are as follows:

[0153] Precise initial parameter space definition: Establish a multi-period optimization model of the virtual power plant and define the initial parameter space; based on the cost of the virtual power plant... A mathematical model is constructed with minimization as the objective, considering distributed energy sources such as distributed photovoltaics, energy storage, and load shedding, as well as the power flow constraints within the virtual power plant. The interaction between the virtual power plant and the main grid is parameterized and placed on the right-hand side of the constraints. Simultaneously, all operational constraints of the virtual power plant are systematically reconstructed into matrix inequalities. This process ultimately yields a standard multi-parameter linear programming problem whose objective function aims to minimize the operating cost of the virtual power plant.

[0154]

[0155]

[0156]

[0157] In the formula: All of these are related to the objective function and constraints of the virtual power plant. This is a vector of cost coefficients. The decision variable vector represents the clearing plan for distributed resources such as photovoltaic power plants, energy storage, and load shedding. This is the coefficient matrix of the decision variables in the constraints; The matrix represents the constant terms on the right-hand side of the constraint conditions; The coefficient matrix of the parameter vector; This is a parameter vector representing the electricity traded between the virtual power plant and the main grid at the common connection point.

[0158] Subsequently, the feasible region of interactive power quantities of the virtual power plant at the common junction point is characterized based on convex set projection theory. Then, based on multi-parameter programming theory, the critical region is characterized by identifying effective and ineffective constraints. Finally, the marginal cost of the virtual power plant at the common junction point with respect to interactive power quantities is analytically represented based on strong duality theory. The partial derivative of the virtual power plant cost function with respect to parameters within each critical region represents the marginal cost in that region, which can be directly characterized by Lagrange multipliers. Therefore, the dimension-reduced marginal cost of the virtual power plant... It can be characterized as:

[0159]

[0160] In the formula: Let be the net output power of the virtual power plant x at time t, which can be positive or negative; This is the i-th dual solution to the dual problem of the above virtual power plant operating cost function; Let be the 0th-order term of the i-th solution to the dual problem; The i-th critical domain after dimensionality reduction of the virtual power plant.

[0161] Step 2: Based on the emission source characteristics of each market entity and the standard carbon emission factors issued by the National Energy Administration, establish a carbon emission calculation model on the power supply side, couple it to form a unit carbon emission quota allocation model, and construct a system-level carbon trading cost model accordingly.

[0162] Step 2.1: Establish a real-time carbon emission calculation method for the power system:

[0163] Traditional energy power generators medium-sized coal-fired power generating units Carbon emission intensity over time period It can be dynamically calculated based on the fuel traceability model, as shown in the following formula:

[0164]

[0165] In the formula: For traditional energy power generators Carbon oxidation rate during coal combustion in coal-fired power generation units; Carbon emission factor for the type of coal burned in coal-fired power units; For traditional energy power generators Coal loss per kilowatt-hour produced by a coal-fired power unit; defining the unit's coal loss per kilowatt-hour produced. The carbon emission intensity column vector for the time period is Then its first The elements are Wherein, the carbon oxidation rate Based on dynamic adjustments to boiler operating conditions, coal type carbon emission factors It can be determined through laboratory calorific value testing.

[0166] Once the carbon emission intensity of the generating units is calculated, the carbon emissions of each unit can be calculated based on its power generation. Cumulative carbon emissions over a period of time The calculation formula is

[0167]

[0168] Step 2.2: Establish a method for allocating carbon emission allowances for the power system:

[0169] The baseline method first calculates the carbon emission factor based on the relationship between energy consumption and greenhouse gas emissions by the National Energy Administration, and then allocates resources based on the carbon emission factor. This method provides overall control and is more conducive to incentivizing various resources in the power grid to participate in emission reduction.

[0170] This method uses a baseline approach to allocate free carbon emission allowances. The free carbon emission allowance for units in the time-of-use system is

[0171]

[0172] In the formula: This is the system's free quota; The free allowance per unit of power is determined by the average emission factor of Anhui Province as calculated by the state. tCO2 / MWh;

[0173] Step 2.3: Establish a method for calculating the carbon trading costs of the power system:

[0174] This method employs a tiered pricing strategy to set carbon trading prices. This strategy ensures a reasonable gradient relationship between CO2 trading prices and carbon emissions, thereby more effectively guiding enterprises to reduce carbon emissions and achieve sustainable development. The carbon trading cost model is as follows;

[0175]

[0176] In the formula: The carbon trading price is tiered. The market benchmark price; The price growth rate is tiered. This represents the length of the price tier range.

[0177]

[0178] In the formula: For the system The cost of carbon trading during a given period.

[0179] Step 3: Construct a robust two-stage clearing optimization model for day-ahead spot trading that takes into account virtual power plant dimensionality reduction and carbon trading.

[0180] Step 3.1: Establish the objective function of the two-stage robust clearing optimization model for day-ahead spot prices:

[0181] The objective of the day-ahead robust optimization clearing model, which takes into account carbon trading and virtual power plant dimensionality reduction models, is to minimize the sum of the costs of the two stages, i.e., the operating cost of the pre-clearing stage. and the regulatory risks and costs during the formal clearing phase sum;

[0182]

[0183] in The current liquidation plan; For 0-1 variables, express Traditional energy generators during the period The medium-sized generating unit has been started and is in operation. express Traditional energy generators during the period The middle unit was shut down; for Traditional energy generators during the period The generating capacity of the medium-sized generating unit; and for Traditional energy generators during the period The reserve capacity for both upward and downward travel of the generating units; The unit power compensation price for load reduction during period t for categories A and B; The compensation price for loads that can be moved during time period t; and For 0-1 variables, =1 and =1 indicates that the energy storage in virtual power plant x is charging and discharging during time period t. =0 and =0 indicates that the energy storage in the virtual power plant x during time period t is neither charged nor discharged; This represents the charging and discharging power of the virtual power plant x during time period t; This represents the load reduction that can be reduced in the virtual power plant x during time period t; traditional energy power generators in the formal clearing phase. Power adjustment plan of medium-sized generating units during time period t Category B load reduction plan for period t The amount of wind curtailment during time period t during the formal clearing phase. Wasted light Involuntary load shedding by load-side users ; These are the uncertain parameters for the new energy sources and the load in the problem; The optimization variables during the formal clearing phase; the operating costs during the pre-clearing phase. This includes the various sub-costs of the recently cleared-out plan, namely, unit operating costs. , for The cost of load reduction in Category A time period for Costs of load reduction in Category B time period, and transaction costs of load shifting. Carbon trading costs The cost of virtual power plant x :

[0184]

[0185] In the formula: , This represents the number of virtual power plants.

[0186] Regulatory risk costs during the formal clearing phase Including the cost of wind curtailment The cost of abandoning light and loss of load cost

[0187]

[0188] In the formula: T is the number of time periods in the entire clearing cycle. As for the clearing interval, there are a total of Each decision-making period.

[0189] Step 3.2: Establish the two-stage robust clearing optimization model for the day-ahead spot market, and define the constraints for the control stage:

[0190] Operational constraints during the pre-clearing phase:

[0191] The constraints of the day-ahead pre-clearing phase of the new power system that takes carbon trading into account include flexible load constraints, power balance constraints, line transmission restrictions, and system reserves.

[0192] The relevant constraints for flexible loads are as follows:

[0193]

[0194]

[0195]

[0196] In the formula: The elasticity coefficient for Category A load reduction and market participation is... ; for Reserve capacity for load reduction during period B; For Category B, the elasticity coefficient for load reduction and market participation is... ; for The maximum load reduction for Category A time period; for The maximum load reduction for Category B time period; The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; To compensate for the unit price;

[0197] Power system power balance constraints are

[0198]

[0199] In the formula: , The number of wind farms; , The number of photovoltaic power plants; , The number of virtual power plants; For wind farm exist Power forecast values ​​for the time period; For photovoltaic power station exist Power forecast values ​​for the time period; The power prediction value of the photovoltaic power station aggregated in the virtual power plant x during time period t; In order to be in Predicted power within the time period; for Forecast values ​​of loads that can be shifted over time; After load shift Power values ​​for a given time period; Let x be the discharge power of the energy stored in the virtual power plant x during time period t; Let x be the charging power of the energy storage in the virtual power plant during time period t; Let x be the load reduction amount that can be reduced in virtual power plant x during time period t.

[0200] The power system line transmission constraints are:

[0201]

[0202] In the formula: , , , and The units Wind farm Photovoltaic power stations Virtual power plant x and load node exist Time period for the line The power transmission allocation factor; branch road The maximum power; For nodes exist The predicted power for the time period, and ; This represents the total number of nodes in the system.

[0203] Reserve capacity constraint is

[0204]

[0205]

[0206] In the formula: and for The system's backup capacity values ​​for different time periods;

[0207] Operational constraints during the formal clearing phase:

[0208] The constraints for the new power system that takes carbon trading into account during the clearing phase recently include unit-related constraints, power balance constraints, and line transmission restrictions.

[0209] Formal clearing phase unit output upper and lower limit constraints

[0210]

[0211] In the formula: for Time-of-use units The amount of power adjustment; and The units Maximum and minimum output values;

[0212] During the formal clearing phase, the unit ramp-up constraints are as follows:

[0213]

[0214] The system power balance constraints during the formal clearing phase are:

[0215]

[0216] In the formula: and For the unit Downward and upward climbing rates; for Time-of-use units The power generation capacity; , and These represent the actual wind power and photovoltaic output values ​​under extreme conditions simulated during the formal clearing phase, respectively. The simulated actual load demand power value for the formal clearing phase includes both flexible and rigid loads. Corresponding to In extreme scenarios during a certain period of time, the first The variable of wind curtailment power in a wind farm; Corresponding to In extreme scenarios during a certain period of time, the first The variable of curtailed solar power from a single photovoltaic power plant; The variable corresponding to the curtailment power of the photovoltaic power station in the x-th virtual power plant under the extreme scenario of time period t; for Involuntary load shedding by users on the load side during a given period.

[0217] The line transmission power constraint during the formal clearing phase is as follows:

[0218]

[0219] In the formula: for Time period nodes The actual load demand power value;

[0220] Step 4: Organize the objective function and constraints of the day-ahead spot two-stage robust clearing optimization model that takes into account virtual power plant dimensionality reduction and carbon trading into matrix form for description, and solve it to achieve low-carbon economic operation of the new power system.

[0221] Step 4.1: Model Solving Algorithm

[0222] Based on the objective function and constraints of the novel two-stage robust optimization clearing model for the day-ahead of a power system considering carbon trading constructed above, it is described in matrix form. The compact form of the proposed model can be expressed as follows:

[0223]

[0224] In the formula: The operating costs of the pre-clearing phase, The regulatory risks and costs during the formal clearing phase, These are the equality constraints for the pre-clearing phase of the model; These are the inequality constraints for the pre-clearing phase of the model. The equality constraints represent the formal clearing phase. The inequality constraints represent the formal clearing phase.

[0225] Because the model constructed uses a min-max-min structure—minimizing the operating cost in the first stage and minimizing the risk cost under adverse scenarios in the second stage—the solution process is relatively complex. Furthermore, the presence of uncertain variables in the model further complicates the solution process. Therefore, the C&CG algorithm is applied to solve the model. This method decomposes the practical problem into a main problem (MP) and a subproblem (SP), and solves these two problems alternately, gradually approaching the optimal solution. The final solution yields the current day-ahead spot market clearing result. This includes the start-up and shutdown status and output of traditional energy generators at various times, the load that can be reduced or shifted, and the resource output in virtual power plants. The mathematical expressions for the main problem MP and the subproblem SP are as follows:

[0226]

[0227]

[0228] In the formula: The auxiliary variables introduced to replace the subproblems are used to directly obtain a temporary solution. Then, the structure of the subproblems is transformed through duality theory, and the max-min problem is transformed into a max single-layer optimization problem.

[0229] Step 4.2: Solution Algorithm Flow

[0230] Based on the above analysis, the overall solution process can be broken down into the following steps.

[0231] Step 1: Initialize the upper bound of the objective function and lower boundary Number of iterations Let the convergence gap between the upper and lower bounds be . , Set to a small positive number;

[0232] Step Two: Let The initial solution obtained Then, substitute the subproblems to solve for the initial extreme scenario. ;

[0233] Step 3: Extreme scenarios Substituting into the main problem, we obtain the optimal solution. Update the Nether equal ;

[0234] Step 4: Apply the optimal solution from Step 3 Substituting into the subproblem, we obtain the optimal solution. ;make Update the upper boundary for and sum;

[0235] Step 5: Judgment If the condition is met, the process ends; otherwise, it will... Return to step three.

[0236] This invention addresses the challenges of strong uncertainties on both the power source and load sides during the clearing process in the electricity spot market, as well as the pain points of privacy leaks and low participation willingness due to the requirement for virtual power plants to submit full models. It proposes a robust optimization clearing method for the day-ahead spot market that considers both carbon trading mechanisms and virtual power plant dimensionality reduction models. This method effectively guides energy conservation and emission reduction in thermal power and promotes the consumption of new energy sources by deeply integrating carbon trading costs into the market clearing objective. It innovatively applies virtual power plant dimensionality reduction models to efficiently characterize their regulation capabilities while strictly protecting internal privacy. Based on robust optimization theory, a two-stage clearing framework considering extreme scenarios is constructed. Through iterative solutions for pre-clearing and formal clearing and efficient implementation using the C&CG algorithm, the method ultimately achieves high robustness of the electricity spot market in dealing with uncertainties such as wind power, photovoltaic output, and load forecasting errors. This significantly improves the market clearing flexibility, dispatch efficiency, and system stability, ensuring the safe and stable operation of the new power system under the low-carbon economic goal.

[0237] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Unless otherwise specified, an element defined by the phrase "comprising..." or "including..." does not exclude the presence of additional elements in the process, method, article, or terminal device that includes said element. Additionally, in this document, "greater than," "less than," "exceeding," etc., are understood to exclude the stated number; "above," "below," "within," etc., are understood to include the stated number.

[0238] Although the above embodiments have been described, those skilled in the art, once they understand the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the above descriptions are merely embodiments of the present invention and do not limit the scope of patent protection of the present invention. Any equivalent structural or procedural transformations made using the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A day-ahead spot market clearing optimization method that takes into account carbon trading and virtual power plant dimensionality reduction, characterized in that, The clearing optimization method includes the following steps: Step 1: Construct a multi-market entity model of the power system including virtual power plants, and use a dimensionality reduction method to perform equivalent modeling of the virtual power plants; Step 2: Based on the emission source characteristics of each market entity and the standard carbon emission factors issued by the National Energy Administration, establish a carbon emission calculation model on the power supply side, couple it to form a unit carbon emission quota allocation model, and construct a system-level carbon trading cost model accordingly. Step 3: Construct a robust two-stage clearing optimization model for day-ahead spot trading that takes into account virtual power plant dimensionality reduction and carbon trading; Step 4: Organize the objective function and constraints of the day-ahead spot two-stage robust clearing optimization model that takes into account virtual power plant dimensionality reduction and carbon trading into matrix form for description, and solve it to achieve low-carbon economic operation of the new power system.

2. The day-ahead spot market clearing optimization method considering carbon trading and virtual power plant dimensionality reduction as described in claim 1, characterized in that, Step 1 involves constructing a multi-market entity model of the power system that includes virtual power plants. The specific model is as follows: Step 1.1: Establish a cost model for traditional energy power generators: The operating costs of traditional energy generators are shown in the following formula: In the formula, , , and Traditional energy generators Operating costs, fuel consumption costs, start-up and shutdown costs, and standby costs; The price quoted by traditional energy power generators for each segment is: In the formula: This represents the price quoted by traditional energy generator k in the nth segment; This indicates the power generation capacity of the traditional energy generator k; This represents the initial power generation capacity of traditional energy generator k in segment n; This represents the marginal cost of generating electricity for traditional energy generator k; The power generation capacity of traditional energy generator k during time period t; K is the number of all traditional energy generators. The total number of segments in the volume and price curves declared by traditional energy power generators; Step 1.2: Establish a load reduction model: For load shedding, load aggregators receive the adjusted compensation price from the grid, assess the response potential of load users, and then resubmit the maximum load shedding amount for the next time period. The electricity spot market trading center formulates a plan based on the demand of the specific time period within an adjustable range. After reaching an agreement, the load shedding responds to grid dispatch at the precise time. Therefore, the load shedding model can be expressed as follows: In the formula: for The amount of load reduction can be reduced during certain time periods; The elasticity coefficient for load reduction and market participation. ; The supplementary price per unit power reduction for Category A load reduction capacity; for The maximum load reduction allowed for Category A time period; and These represent the compensation prices at which users whose load can be reduced can just receive the benefit and when the response volume is maximized, respectively. The maximum elasticity coefficient for load reduction; For load compensation sensitivity, The smaller the value, the more sensitive users are to the compensation price, and the greater the impact of price changes on users; for The cost of reducing load during certain periods; This is the time interval for the spot market to clear out. Step 1.3: Establish a transferable load model: When the compensation price published by the electricity spot market trading center is low, the benefits gained by users from changing their electricity usage time are insufficient to compensate for the inconvenience and losses caused by shifting electricity consumption. Consequently, users' enthusiasm for participating in the market will be very low, making it difficult to meet the needs of the electricity spot market trading center. Based on the principles of consumer psychology, the shiftable load model can be expressed as: In the formula: The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; The compensation price per unit power for the transferable load.

3. The day-ahead spot market clearing optimization method considering carbon trading and virtual power plant dimensionality reduction as described in claim 2, characterized in that, Step 1 employs a dimensionality reduction method to perform an equivalent model of the virtual power plant, including: Step 1.4: Establish a virtual power plant model: Virtual power plants participate in day-ahead electricity spot market transactions by integrating distributed energy resources, energy storage systems, and controllable load resources. Virtual power plants aggregate photovoltaic (PV) power, load shedding, and distributed energy storage resources; therefore, virtual power plant x declares the predicted power output of the PV power plant in time period t. Unit charge / discharge power of independent energy storage and Quota capacity Compensation price for load reduction declaration Relevant parameters participate in the clearing of the electricity spot market, therefore the operating cost of the virtual power plant... Represented as: In the formula: For the energy storage discharge revenue in virtual power plant x, The cost of energy storage charging in a virtual power plant x The battery aging cost in the energy storage of the virtual power plant x. This allows for the reduction of load costs in virtual power plants. Step 1.5: Dimensionally reduced representation of the marginal cost of the virtual power plant: (1) Precise definition of initial parameter space: Establish a virtual power plant multi-time period optimization model and define the initial parameter space; (2) Critical domain segmentation and extraction of multiplier constant properties: initial optimization space solution based on convex set projection theory; (3) Multi-period marginal cost representation: Solving the marginal cost function of the virtual power plant based on multi-period linear programming.

4. The day-ahead spot market clearing optimization method considering carbon trading and virtual power plant dimensionality reduction as described in claim 3, characterized in that, The carbon trading cost modeling in step 2 is as follows: Step 2.1: Establish a real-time carbon emission calculation method for the power system: Traditional energy power generators medium-sized coal-fired power generating units Carbon emission intensity over time Based on dynamic calculation using the fuel traceability model, the formula is as follows: In the formula: For traditional energy power generators Carbon oxidation rate during coal combustion in coal-fired power generation units; Carbon emission factor for the type of coal burned in coal-fired power units; For traditional energy power generators Coal loss per kilowatt-hour produced by a coal-fired power unit; defining the unit's coal loss per kilowatt-hour produced. The carbon emission intensity column vector for the time period is Then its first The elements are ; Once the carbon emission intensity of the generating units is calculated, the carbon emissions of each unit can be calculated based on its power generation. Cumulative carbon emissions over a period of time The calculation formula is: Step 2.2: Establish a method for allocating carbon emission allowances for the power system: The baseline method is used to allocate free carbon emission allowances. The free carbon emission allowance for units in the time-of-use system is In the formula: For the system's free quota, This is a free allowance per unit of power, the value of which is determined by the nationally calculated average emission factor. tCO2 / MWh; Step 2.3: Establish a method for calculating the carbon trading costs of the power system: A tiered pricing strategy was adopted to set carbon trading prices, and the carbon trading cost model is as follows; In the formula: For tiered carbon trading prices, As the market benchmark price, For tiered price growth rates, The length of the price tier range; In the formula: For the system The cost of carbon trading during a given period.

5. The day-ahead spot market clearing optimization method considering carbon trading and virtual power plant dimensionality reduction as described in claim 4, characterized in that, The function of the two-stage robust clearing optimization model for the day-to-day spot market in step 3 is as follows: Step 3.1: Establish the objective function of the two-stage robust clearing optimization model for day-ahead spot prices: The objective of the day-ahead spot market clearing optimization model, which takes into account carbon trading and virtual power plant dimensionality reduction models, is to minimize the sum of costs in the two stages, i.e., the operating cost of the pre-clearing stage. and the regulatory risks and costs during the formal clearing phase sum; in The current liquidation plan; For 0-1 variables, express Traditional energy generators during the period The medium-sized generating unit has been started and is in operation. express Traditional energy generators during the period The middle unit was shut down; for Traditional energy generators during the period The generating capacity of the medium-sized generating unit; and for Traditional energy generators during the period The reserve capacity for both upward and downward travel of the generating units; The unit power compensation price for load reduction during period t for categories A and B; The compensation price for loads that can be moved during time period t; and For 0-1 variables, =1 and =1 indicates that the energy storage in virtual power plant x is charging and discharging during time period t. =0 and =0 indicates that the energy storage in the virtual power plant x during time period t is neither charged nor discharged; This represents the charging and discharging power of the virtual power plant x during time period t; This represents the load reduction that can be reduced in the virtual power plant x during time period t; traditional energy power generators in the formal clearing phase. Power adjustment plan of medium-sized generating units during time period t Category B load reduction plan for period t The amount of wind curtailment during time period t during the formal clearing phase. Wasted light Involuntary load shedding by load-side users ; These are the uncertain parameters for the new energy sources and the load in the problem; These are decision variables during the formal clearing phase. Operating costs during the pre-clearing phase This includes the various sub-costs of the recent clearing plan, namely the operating costs of generating units in traditional energy power generators. , for The cost of load reduction in Category A time period for Costs of load reduction in Category B time period, and transaction costs of load shifting. Operating costs of virtual power plants Carbon trading costs ; In the formula: , The number of virtual power plants; Regulatory risk costs during the formal clearing phase Including the cost of wind curtailment The cost of abandoning light and loss of load cost , In the formula: T is the number of time periods in the entire clearing cycle. As for the clearing interval, there are a total of One decision-making period; Step 3.2: Establish the two-stage robust clearing optimization model for the day-ahead spot market, and define the constraints for the control stage: Operational constraints during the pre-clearing phase: The pre-clearing stage constraints of the day-ahead spot market clearing optimization model, which takes into account carbon trading and virtual power plant dimensionality reduction models, include flexible load constraints, power balance constraints, line transmission limitations, and system reserves. The relevant constraints for flexible loads are as follows: In the formula: The elasticity coefficient for Category A load reduction and market participation is... ; for Reserve capacity for load reduction in period B; For Category B, the elasticity coefficient for load reduction and market participation is... ; for The maximum load reduction for Category A time period; for The maximum load reduction for Category B time period; The starting period after the shift in the electricity consumption curve over a certain period of time; and The earliest and latest start time periods after shifting the electricity consumption curve for loads that can be shifted; To compensate for the unit price; The power balance constraints of the power system are: In the formula: , The number of wind farms; , The number of photovoltaic power plants; , The number of virtual power plants; For wind farm exist Power forecast values ​​for the time period; For photovoltaic power station exist Power forecast values ​​for the time period; The power prediction value of the photovoltaic power station aggregated in the virtual power plant x during time period t; In order to be in Predicted power within the time period; for Forecast values ​​of loads that can be shifted over time; After load shift Power values ​​for a given time period; Let x be the discharge power of the energy stored in the virtual power plant x during time period t; Let x be the charging power of the energy storage in the virtual power plant during time period t; Let x be the load reduction amount that can be reduced in virtual power plant x during time period t; The power system line transmission constraints are: In the formula: , , , and The units Wind farm Photovoltaic power stations Virtual power plant x and load node exist Time period for the line The power transmission allocation factor; branch road The maximum power; For nodes exist The predicted power over the time period, and ; This represents the total number of nodes in the system. Reserve capacity constraint is In the formula: and for The system's backup capacity values ​​for different time periods; Operational constraints during the formal clearing phase: The constraints of the formal clearing phase of the day-ahead spot market clearing optimization model, which takes into account carbon trading and virtual power plant dimensionality reduction models, include unit-related constraints, power balance constraints, and line transmission limitations. Upper and lower limits of unit output during the formal clearing phase: In the formula: for Traditional energy power generation units The amount of power adjustment; and The units Maximum and minimum output values; The ramp-up constraints for the units during the formal clearing phase are: The system power balance constraints during the formal clearing phase are: In the formula: and For the unit Downward and upward climbing rates; for Time-of-use units The power generation capacity; , and These represent the actual wind power and solar power output values ​​under extreme scenarios simulated during the formal clearing phase, respectively. The simulated actual load demand power value for the formal clearing phase includes both flexible and rigid loads. Corresponding to In extreme scenarios during a certain period of time, the first The variable of wind curtailment power in a wind farm; Corresponding to In extreme scenarios during a certain period of time, the first The variable of curtailed solar power from a single photovoltaic power plant; The variable corresponding to the curtailment power of the photovoltaic power station in the x-th virtual power plant under the extreme scenario of time period t; for Involuntary load shedding by users on the load side during a given period; The line transmission power constraint during the formal clearing phase is as follows: In the formula: For the post-clearing Time period nodes The actual load demand power value.

6. The day-ahead spot market clearing optimization method considering the carbon trading and virtual power plant dimensionality reduction model as described in claim 4, characterized in that, In step 4, the objective function and constraints of the two-stage robust clearing optimization model for day-ahead spot trading are organized into matrix form, as shown below: Step 4.1: Model Solving Algorithm Based on the objective function and constraints of the day-ahead spot two-stage robust clearing optimization model constructed above, which takes into account the carbon trading and virtual power plant dimensionality reduction model, it is described in matrix form. The compact form of the proposed model is as follows: In the formula: The operating costs of the pre-clearing phase, The regulatory risks and costs during the formal clearing phase, These are the equality constraints for the pre-clearing phase of the model. These are the inequality constraints for the pre-clearing phase of the model. These represent the equality constraints during the formal clearing phase. The inequality constraints represent the formal clearing phase. The C&CG algorithm is applied to solve the model. This method decomposes the actual problem into a main problem (MP) and a subproblem (SP), and solves these two problems alternately to gradually approach the optimal solution, ultimately yielding the current day-to-date spot market clearing result. It includes the start-up and shutdown status and output of traditional energy generators at various times, the results of load reduction and transferability, and the resource output of virtual power plants. The mathematical expressions for the main problem MP and the subproblem SP are as follows: In the formula: The auxiliary variables introduced to replace the subproblems are used to directly obtain a temporary solution. Then, the structure of the subproblems is transformed through duality theory, and the max-min problem is transformed into a max single-layer optimization problem. Step 4.2: Solving the C&CG algorithm flow Step 1: Initialize the upper bound of the objective function and lower boundary Number of iterations Let the convergence gap between the upper and lower bounds be . , Set to a small positive number; Step Two: Let The initial solution obtained Then, substitute the subproblems to solve for the initial extreme scenario. ; Step 3: Extreme scenarios Substituting into the main problem, we obtain the optimal solution. Update the Nether equal ; Step 4: Apply the optimal solution from Step 3 Substituting into the subproblem, we obtain the optimal solution. ;make Update the upper boundary for and sum; Step 5: Judgment If the condition is met, the process ends; otherwise, it will... Return to step three.

Images