Virtual power plant power bidding strategy optimization method and related equipment

Abstract

The application provides a virtual power plant power bidding strategy optimization method and related equipment, the virtual power plant includes a battery energy storage system, a generator set and a photovoltaic component, the method comprises the following steps: obtaining a total cost function of a scene according to a battery state parameter of the battery energy storage system, a generator set parameter, a day-ahead market power bidding parameter, a real-time market power bidding parameter and a carbon emission parameter, and generating a total operation cost function based on the total cost function of the scene and a scene probability; constructing a conditional value at risk function according to a risk value parameter, a preset confidence value, the total cost function of the scene and the scene probability; constructing a day-ahead market power bidding strategy optimization model based on system operation constraints, power market bidding constraints, the total operation cost function and the conditional value at risk function; solving the day-ahead market power bidding strategy optimization model to obtain a day-ahead market power bidding strategy. The application can simultaneously optimize the economy and risk control capability of the power bidding strategy of the virtual power plant.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method for optimizing power bidding strategies in a virtual power plant and related equipment. Background Technology

[0002] With the transformation of the global energy structure and the large-scale integration of renewable energy, power systems face increasing challenges. Traditional power grids encounter numerous difficulties in coping with the volatility of new energy sources and the integration of distributed power sources. Therefore, virtual power plants, as an emerging power resource management model, are gradually becoming important participants in the electricity market. Virtual power plants integrate multiple distributed energy systems (such as wind power, solar power, energy storage devices, and traditional thermal power) into a coordinated and unified virtual power system through intelligent dispatch and optimized resource allocation, achieving cross-regional and cross-time power resource sharing and optimized dispatch. Unlike traditional power systems, virtual power plants possess greater flexibility and adaptability, dynamically adjusting power generation and consumption strategies based on electricity market demand, energy supply conditions, and price changes, thereby improving grid dispatch efficiency and reducing operating costs.

[0003] Virtual power plants face multi-objective decision-making problems when participating in electricity market bidding. They must not only maximize economic benefits but also balance multiple objectives such as risk management, cost control, and carbon emission reduction, making their bidding decisions more complex. In the electricity market, virtual power plants need to balance day-ahead and real-time market considerations, taking into account uncertainties such as price fluctuations, demand forecasting errors, and generator failure probabilities.

[0004] However, the electricity bidding strategies of virtual power plants in the day-ahead market in related technologies often ignore the aforementioned risks and uncertainties, making virtual power plants prone to significant losses when facing fluctuations in the electricity market. Therefore, how to provide an electricity bidding strategy that can balance operating costs and risk control has become an urgent problem to be solved. Summary of the Invention

[0005] The main objective of this application is to propose a method for optimizing the power bidding strategy of a virtual power plant, which aims to simultaneously optimize the economy and risk control capabilities of the power bidding strategy of the virtual power plant.

[0006] To achieve the above objectives, a first aspect of this application proposes a method for optimizing electricity bidding strategies in a virtual power plant. The virtual power plant includes a battery energy storage system, multiple generator sets, and photovoltaic modules. The method includes:

[0007] Based on the battery status parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters, the total cost function of the virtual power plant in each scenario is obtained. Based on the total cost function of the scenario and the scenario probability corresponding to the scenario, the total operating cost function of the virtual power plant is generated. The day-ahead market electricity price and / or the photovoltaic power generation of the photovoltaic modules are different between every two scenarios.

[0008] A conditional value-at-risk function is constructed based on the value-at-risk parameter, the preset confidence value, the total cost function of the scenario, and the corresponding scenario probability.

[0009] Obtain the system operation constraints and electricity market bidding constraints of the virtual power plant, and construct an optimization model for the day-ahead market electricity bidding strategy of the virtual power plant based on the system operation constraints, the electricity market bidding constraints, the total operating cost function, and the conditional risk value function;

[0010] The day-ahead market electricity bidding strategy optimization model is solved to obtain the day-ahead market electricity bidding strategy of the virtual power plant. The day-ahead market electricity bidding strategy includes at least optimizing the day-ahead market electricity bidding parameters.

[0011] Optionally, in some embodiments, constructing the day-ahead market electricity bidding strategy optimization model for the virtual power plant based on the system operating constraints, the electricity market bidding constraints, the total operating cost function, and the conditional value-at-risk function includes:

[0012] A single-objective optimization model for total operating cost is constructed based on the total operating cost function, the system operating constraints, and the electricity market bidding constraints. A single-objective optimization model for conditional risk value is also constructed based on the conditional risk value function, the risk value parameter, the scenario total cost function, the system operating constraints, and the electricity market bidding constraints.

[0013] Obtain the risk tolerance parameter, and construct the conditional risk value constraint based on the risk tolerance parameter and the conditional risk value function;

[0014] Based on the conditional risk value constraint, the system operation constraint, the electricity market bidding constraint, the conditional risk value function, the risk value parameter, the scenario total cost function, and the operating total cost function, an electricity bidding strategy optimization model that balances conditional risk value and operating total cost is constructed.

[0015] Based on the single-objective optimization model of total operating cost, the single-objective optimization model of conditional value at risk, and the power bidding strategy optimization model that balances conditional value at risk and total operating cost, the day-ahead market power bidding strategy optimization model for the virtual power plant is constructed.

[0016] Optionally, in some embodiments, the battery state parameters are calculated from the scenario charging power parameters, scenario discharging power parameters, charging efficiency, discharging efficiency, and battery capacity corresponding to the battery energy storage system. The generator set parameters include generator set power parameters. The step of obtaining the scenario total cost function of the virtual power plant in each scenario based on the battery state parameters of the battery energy storage system, the generator set parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters includes:

[0017] Obtain the battery degradation cost coefficient corresponding to the battery energy storage system, the generator set cost coefficient corresponding to each generator set, and the carbon emission factor corresponding to each generator set;

[0018] The battery loss cost of the battery energy storage system in each scenario is calculated based on the battery degradation cost coefficient, the scenario charging power parameters, the scenario discharging power parameters, the charging efficiency, the discharging efficiency, and the battery capacity.

[0019] The total power generation cost of the multiple generator sets in each scenario is calculated based on the generator set cost coefficient and the generator set power parameters.

[0020] Calculate the carbon emission cost for each scenario based on the carbon emission factor and the carbon emission parameter.

[0021] The total cost function of the virtual power plant in each scenario is calculated based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market electricity bidding parameters, and the real-time market electricity bidding parameters.

[0022] Optionally, in some embodiments, the day-ahead market electricity bidding parameters include day-ahead market electricity purchase price parameters, day-ahead market electricity purchase price parameters, day-ahead market electricity sales price parameters, and day-ahead market electricity sales parameters; the real-time market electricity bidding parameters include real-time market electricity purchase price parameters, real-time market electricity sales price parameters, real-time market electricity purchase price parameters, and real-time market electricity sales parameters; the step of calculating the total cost function of the virtual power plant in each scenario based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market electricity bidding parameters, and the real-time market electricity bidding parameters includes:

[0023] The net cost of the virtual power plant's day-ahead market transactions in each scenario is calculated based on the day-ahead market electricity purchase price parameters, the day-ahead market electricity purchase volume parameters, the day-ahead market electricity sales price parameters, and the day-ahead market electricity sales volume parameters.

[0024] The real-time market purchase price parameter, the real-time market sales price parameter, the real-time market purchase volume parameter, and the real-time market sales volume parameter are used to calculate the real-time market transaction net cost of the virtual power plant in each scenario.

[0025] The total cost function of the virtual power plant in each scenario is calculated based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market transaction net cost, and the real-time market transaction net cost.

[0026] Optionally, in some embodiments, solving the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy for the virtual power plant includes:

[0027] Construct a scenario-based electricity purchase bidding curve model based on the day-ahead market electricity purchase price parameters and the day-ahead market electricity purchase volume parameters;

[0028] A scenario-based electricity bidding curve model is constructed based on the day-ahead market electricity price parameters and the day-ahead market electricity volume parameters.

[0029] The optimization model for the day-ahead market electricity bidding strategy is solved to obtain the optimized day-ahead market electricity purchase price parameters, optimized day-ahead market electricity purchase volume parameters, optimized day-ahead market electricity sales price parameters, and optimized day-ahead market electricity sales volume parameters for each scenario.

[0030] Based on the optimized day-ahead market electricity purchase price parameters, the optimized day-ahead market electricity purchase quantity parameters, and the electricity purchase bidding curve model, the day-ahead market electricity purchase bidding strategy of the virtual power plant is obtained; and based on the optimized day-ahead market electricity sales price parameters, the optimized day-ahead market electricity sales quantity parameters, and the electricity sales bidding curve model, the day-ahead market electricity sales bidding strategy of the virtual power plant is obtained.

[0031] The day-ahead market power bidding strategy of the virtual power plant is determined based on the day-ahead market power purchase bidding strategy and the day-ahead market power sale bidding strategy.

[0032] Optionally, in some embodiments, solving the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy for the virtual power plant includes:

[0033] The single-objective optimization model for total operating cost is solved to obtain the optimal total operating cost. Based on the optimal total operating cost, the single-objective optimization model for conditional value of risk is solved to obtain the conditional value of risk under the constraint of the optimal total operating cost and the market electricity bidding strategy on the first day.

[0034] The conditional risk value single-objective optimization model is solved to obtain the optimal conditional risk value. Based on the optimal conditional risk value, the total operating cost single-objective optimization model is solved to obtain the total operating cost of the virtual power plant under the constraint of the optimal conditional risk value and the market electricity bidding strategy on the second day.

[0035] The risk tolerance range is determined based on the conditional risk value under the optimal total operating cost constraint and the optimal conditional risk value.

[0036] Obtain the preset conditional value at risk step size, minimum conditional value at risk step size, conditional value at risk step size update ratio, and cost convergence threshold;

[0037] Based on the conditional risk value step size, the minimum conditional risk value step size, the conditional risk value step size update ratio, the cost convergence threshold, and the risk tolerance interval, the power bidding strategy optimization model balancing conditional risk value and total operating cost, as well as the conditional risk value single-objective optimization model, are iteratively solved until the number of iterations reaches the preset number of iterations. The day-ahead market power bidding strategy corresponding to the virtual power plant is determined based on the optimized day-ahead market power bidding parameters obtained from each iteration.

[0038] Optionally, in some embodiments, the iterative solution process corresponding to each iteration number includes the following steps:

[0039] The current risk tolerance threshold is determined based on the preconditional risk value, and the total operating cost of the power bidding strategy optimization model, which balances the conditional risk value and the total operating cost, is calculated based on the current risk tolerance threshold. The preconditional risk value is the conditional risk value obtained from the previous solution of the single-objective optimization model for the conditional risk value.

[0040] The conditional risk value single-objective optimization model is solved based on the total operating cost under the current risk tolerance threshold constraint to obtain the conditional risk value under the current risk tolerance threshold and the day-ahead market electricity bidding strategy of the virtual power plant.

[0041] The risk tolerance parameter is updated based on the preceding total operating cost, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio. The preceding total operating cost is the total operating cost obtained from the previous solution of the power bidding strategy optimization model that balances the conditional risk value and the total operating cost.

[0042] Optionally, in some embodiments, the risk value step size update ratio includes a step size increase ratio and a step size decrease ratio. Updating the risk tolerance parameter based on the preceding total operating cost, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio includes:

[0043] When the difference between the total operating cost under the current risk tolerance threshold constraint and the previous total operating cost exceeds the cost convergence threshold, the conditional risk value step size is updated according to the step size increase ratio to obtain the updated conditional risk value step size.

[0044] Alternatively, when the difference between the total operating cost under the current risk tolerance threshold constraint and the total operating cost of the preceding operation does not exceed the cost convergence threshold, the conditional risk value step size is updated according to the step size reduction ratio and the minimum conditional risk value step size to obtain the updated conditional risk value step size.

[0045] The risk tolerance parameter is updated based on the updated conditional value-at-risk step size to obtain the updated risk tolerance parameter.

[0046] To achieve the above objectives, a second aspect of this application provides a power bidding strategy optimization device for a virtual power plant, the device comprising:

[0047] The generation unit is used to obtain the total cost function of the virtual power plant in each scenario based on the battery status parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters, and to generate the total operating cost function of the virtual power plant based on the total cost function of the scenario and the scenario probability corresponding to the scenario. The day-ahead market electricity price and / or the photovoltaic power generation of the photovoltaic module are different between every two scenarios.

[0048] The first construction unit is used to construct a conditional value-at-risk function based on the value-at-risk parameter, the preset confidence value, the total cost function of the scenario, and the corresponding scenario probability.

[0049] The second construction unit is used to obtain the system operation constraints and electricity market bidding constraints of the virtual power plant, and to construct the day-ahead market electricity bidding strategy optimization model of the virtual power plant based on the system operation constraints, the electricity market bidding constraints, the total operating cost function and the conditional risk value function.

[0050] The solution unit is used to solve the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy of the virtual power plant. The day-ahead market electricity bidding strategy includes at least optimizing the day-ahead market electricity bidding parameters.

[0051] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect.

[0052] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0053] This application constructs and solves an optimization model that comprehensively considers the total operating cost of a virtual power plant, as well as the uncertainties caused by wind and solar power generation and day-ahead market electricity prices. This model simultaneously balances the economic viability and risk control capabilities of the virtual power plant's day-ahead market electricity bidding strategy. The application constructs a total operating cost function that integrates multi-dimensional costs and incorporates the differences in photovoltaic output and day-ahead market electricity price fluctuations. It quantifies the uncertainty risk of the electricity market by constructing a conditional value-at-risk model, limiting extreme losses under confidence constraints. This makes the electricity bidding strategy both economical and robust, avoiding significant losses caused by electricity price fluctuations and improving the risk control capabilities of the virtual power plant in the electricity bidding process. Furthermore, the optimization model embeds system operation constraints and electricity market bidding constraints to ensure that the generated day-ahead market electricity bidding strategy is safe to execute and meets market requirements. The final generated day-ahead electricity bidding strategy can effectively perceive market volatility risks while reducing total operating costs, dynamically balancing costs and risks, achieving synergistic optimization of virtual power plant operating costs and operational stability, and enhancing market competitiveness. Attached Figure Description

[0054] Figure 1 This is a flowchart of the power bidding strategy optimization method for virtual power plants provided in the embodiments of this application;

[0055] Figure 2 yes Figure 1 The flowchart of step S101 in the text;

[0056] Figure 3 yes Figure 2 Flowchart of step S205;

[0057] Figure 4 yes Figure 1 The flowchart of step S103 in the process;

[0058] Figure 5 yes Figure 1 The flowchart of step S104 in the process;

[0059] Figure 6 This is a schematic diagram of the structure of the power bidding strategy optimization device for a virtual power plant provided in the embodiments of this application;

[0060] Figure 7 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0063] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0064] With the transformation of the global energy structure and the large-scale integration of renewable energy, power systems face increasing challenges. Traditional power grids encounter numerous difficulties in coping with the volatility of new energy sources and the integration of distributed power sources. Therefore, Virtual Power Plants (VPPs), as an emerging power resource management model, are gradually becoming important participants in the electricity market. VPPs integrate multiple geographically dispersed and diverse distributed energy systems (such as wind power, solar power, energy storage devices, and traditional thermal power) into a coordinated and unified virtual power system through intelligent dispatch and optimized resource allocation, achieving cross-regional and cross-time power resource sharing and optimized dispatch. Unlike traditional power systems, VPPs possess greater flexibility and adaptability, dynamically adjusting power generation and consumption strategies based on electricity market demand, energy supply conditions, and price changes, thereby improving grid dispatch efficiency and reducing operating costs.

[0065] Virtual power plants face multi-objective decision-making problems when participating in electricity market bidding. They must not only maximize economic benefits but also balance multiple objectives such as risk management, cost control, and carbon emission reduction, making their bidding decisions more complex and difficult. In the electricity market, virtual power plants need to balance day-ahead and real-time market considerations, taking into account uncertainties such as price fluctuations, demand forecasting errors, and generator failure probabilities.

[0066] However, existing technologies for electricity bidding strategies targeting virtual power plants model and optimize based on deterministic and predictable parameters. Furthermore, these optimization processes only consider a single economic objective, failing to adequately account for key risk factors (such as price fluctuations and prediction errors) and the interplay between multiple objectives. This results in significantly greater-than-expected losses for electricity bidding strategies developed using these methods when facing the complexities of the real market.

[0067] Based on this, embodiments of this application provide a method and related equipment for optimizing the power bidding strategy of a virtual power plant, aiming to optimize the economy and risk control capabilities of the power bidding strategy of the virtual power plant and achieve multi-objective collaborative decision-making.

[0068] The method and related equipment for optimizing the power bidding strategy of a virtual power plant provided in this application are specifically described through the following embodiments. First, the method for optimizing the power bidding strategy of a virtual power plant in this application embodiment is described.

[0069] The method for optimizing electricity bidding strategies for virtual power plants provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the method for optimizing electricity bidding strategies for virtual power plants, but is not limited to the above forms.

[0070] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0071] Figure 1 This is an optional flowchart of the power bidding strategy optimization method for virtual power plants provided in this embodiment of the disclosure. Figure 1 The method may include, but is not limited to, steps S101 to S104. It is also understood that this embodiment... Figure 1 The order of steps S101 to S104 is not specifically limited, and the order of steps can be adjusted or some steps can be reduced or added according to actual needs.

[0072] In step S101 of some embodiments, based on the battery state parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters, the total cost function of the virtual power plant under each scenario is obtained, and the total operating cost function of the virtual power plant is generated based on the total cost function of the scenario and the scenario probability corresponding to the scenario. Step S101 will be described in detail below.

[0073] The distributed energy system involved in the virtual power plant mentioned in this disclosure includes battery energy storage systems, multiple generator sets, and photovoltaic modules. Based on this, an optimization method for the electricity bidding strategy of the virtual power plant is introduced. It is understandable that in practical applications, the optimization method for the electricity bidding strategy of the virtual power plant can further incorporate more uncertainties to improve the practicality and robustness of the electricity bidding strategy.

[0074] Battery energy storage systems, generator sets, and photovoltaic modules, as independently operating physical devices, possess their own power generation, energy storage, or regulation characteristics (such as the intermittency of photovoltaics and the charging and discharging capabilities of batteries). Virtual power plants aggregate these distributed energy resources into a logically unified virtual entity, endowing it with collaborative response capabilities to participate in the electricity market or grid dispatch.

[0075] These devices collectively influence the operating costs of virtual power plants, while also introducing numerous uncertainties. Battery storage systems incur costs related to battery wear and capacity degradation during operation, while generator sets generate fuel and carbon emissions, and incur maintenance costs when operational problems arise. Therefore, battery storage systems and generator sets impact the operating costs of virtual power plants. Photovoltaic modules or other renewable energy sources can replace high-cost energy sources, reducing coal-fired power fuel costs, carbon emission costs, and market electricity purchase expenditures. However, price fluctuations in photovoltaic module generation, as well as the difference between day-ahead and real-time markets, and demand forecasting errors, are major reasons for the uncertainty in virtual power plants' participation in electricity market bidding decisions. For example, photovoltaic modules generate the most electricity at midday, at which time bidding strategies may favor selling more electricity or replacing other high-cost energy sources with photovoltaic power. Furthermore, the day-ahead market is affected by factors such as electricity price fluctuations, which also contribute to uncertainty.

[0076] Therefore, when constructing an optimization model for the bidding strategy of virtual power plants participating in the day-ahead market, we can first construct a scenario set S based on photovoltaic power generation and day-ahead market electricity prices. The scenario set S includes multiple scenarios s, each scenario s representing a complete set of uncertain parameters covering all time periods t within the entire scheduling cycle (e.g., 24 hours). Each scenario s has a corresponding scenario probability π. S .

[0077] At least one parameter differs between any two scenarios 's'. This could be a difference in day-ahead market electricity prices, photovoltaic (PV) power generation, or both – representing the uncertainty of the combination of day-ahead market electricity prices and PV power generation. This ensures that the generated scenario parameters do not completely overlap. The PV power generation scenario includes a series of PV output curves, with the horizontal axis representing time (1-24h) and the vertical axis representing PV power generation. The day-ahead market electricity price scenario includes a series of price curves, with the horizontal axis representing time (1-24h) and the vertical axis representing the day-ahead market electricity price. Scenario 's' are obtained by clustering historical scenarios, but can also be generated using other clustering methods; no restrictions are placed here.

[0078] In the optimization problem of electricity bidding strategy for virtual power plants, considering the uncertainties of photovoltaic power generation and day-ahead market electricity prices, a two-stage stochastic programming method is adopted to construct the objective function of the total operating cost of the virtual power plant. Before constructing the objective function of the total operating cost of the virtual power plant, the scenario total cost function under each scenario s can be constructed first. The objective function of the total operating cost represents minimizing the weighted expected value of the operating cost corresponding to all possible scenarios.

[0079] In some embodiments, please refer to Figure 2 Based on the battery state parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters, the total cost function of the virtual power plant in each scenario is obtained, including the following steps S201 to S205:

[0080] Step S201: Obtain the battery degradation cost coefficient corresponding to the battery energy storage system, the generator set cost coefficient corresponding to each generator set, and the carbon emission factor corresponding to each generator set.

[0081] Step S202: Calculate the battery loss cost of the battery energy storage system in each scenario based on the battery degradation cost coefficient, scenario charging power parameters, scenario discharging power parameters, charging efficiency, discharging efficiency, and battery capacity.

[0082] Step S203: Calculate the total power generation cost of multiple generator sets in each scenario based on the generator set cost coefficient and generator set power parameters.

[0083] Step S204: Calculate the carbon emission cost for each scenario based on the carbon emission factor and carbon emission parameters;

[0084] Step S205: Calculate the total cost function of the virtual power plant in each scenario based on battery loss cost, total power generation cost of the generator set, carbon emission cost, day-ahead market electricity bidding parameters, and real-time market electricity bidding parameters.

[0085] Specifically, in this embodiment of the disclosure, the total cost function for each scenario s can be constructed based on the battery state parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters. Among these, the battery state parameter SOC... s (t) is the charging power parameter P corresponding to the battery energy storage system in the scenario. B,c,s (t), Scene discharge power parameter P B,d,s (t), charging efficiency η ch Discharge efficiency η dis and battery capacity E BESS The calculation is obtained, and the specific calculation formula is SOC. s (t)=SOC s (t-1)+[η ch P B,c,s (t)-P B,d,s (t) / η diis ] / E BESs Among them, SOC s (t) represents the state of charge level of the battery energy storage system during time period t in scenario s, P B,c,s (t) represents the battery charging power parameter during time period t in scenario s, P B,d,s (t) represents the battery discharge power parameter within time period t in scenario s. Generator parameters include the generator power parameter P. G,s (t), which represents the power generation parameter of the generator set during time period t in scenario s.

[0086] In step S201 of some embodiments, the battery degradation cost coefficients a, b, c, and d corresponding to the battery energy storage system can be obtained first, and the generator set cost coefficient a corresponding to each generator set can be obtained. m b m c m and the carbon emission factor corresponding to each generator set. It is understandable that the charging efficiency η ch Discharge efficiency η dis Battery capacity E BESS Battery degradation cost coefficients a, b, c, d; generator set cost coefficient a m b m c m and the carbon emission factor corresponding to each generator set. The input information required to construct these scenario total cost functions can all be obtained directly from the database.

[0087] In step S202 of some embodiments, the battery loss cost C of the battery energy storage system in scenario s can be calculated first. B,s Specifically, this can be determined based on battery degradation cost coefficients a, b, c, and d, and the scenario charging power parameter P. B,c,s (t), Scene discharge power parameter P B,d,s (t), charging efficiency η ch Discharge efficiency η dis and battery capacity E BESS Calculate the battery loss cost C of the battery energy storage system in each scenario s. B,s The resulting battery loss cost C B,s The specific calculation formula is shown in formula (1) below:

[0088]

[0089] Among them, SOC s (t-1) represents the state of charge level of the battery energy storage system in the previous time period t-1 in scenario s.

[0090] In step S203 of some embodiments, the total power generation cost C of the generator set corresponding to scenario s can be calculated. G,s Specifically, it can be based on the generator set cost coefficient a. m b m c m and generator set power parameters P G,s (t) Calculate the total power generation cost of multiple generator sets in each scenario s, and obtain the total power generation cost C of the generator sets. G,s The specific calculation formula (2) is as follows:

[0091]

[0092] In step S204 of some embodiments, the carbon emission cost under each scenario s can be calculated. Specifically, it can be based on carbon emission factors. and carbon emission parameters Calculate the carbon emission cost for each scenario s The carbon emission cost obtained The specific calculation formula (3) is as follows:

[0093]

[0094] In step S205 of some embodiments, the battery wear cost C can be determined based on the battery wear cost. B,s Total power generation cost of generator set CG,s Carbon emission costs The total cost function C of the virtual power plant in each scenario s is obtained by calculating the current market electricity bidding parameters and the real-time market electricity bidding parameters. s .

[0095] Please refer to Figure 3 In some embodiments, step S205 includes, but is not limited to, the following steps S301 to S303:

[0096] Step S301: Calculate the net cost of the virtual power plant's day-ahead market transactions in each scenario based on the day-ahead market electricity purchase price parameters, day-ahead market electricity purchase volume parameters, day-ahead market electricity sales price parameters, and day-ahead market electricity sales volume parameters.

[0097] Step S302: Calculate the real-time net market transaction cost of the virtual power plant in each scenario based on the real-time market electricity purchase price parameters, real-time market electricity sales price parameters, real-time market electricity purchase parameters, and real-time market electricity sales parameters.

[0098] Step S303: Calculate the total cost function of the virtual power plant in each scenario based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the net cost of day-ahead market transactions, and the net cost of real-time market transactions.

[0099] Specifically, the total operating cost of a virtual power plant is also affected by the net cost of day-ahead market transactions and the net cost of real-time market transactions. The net cost of day-ahead market transactions is the cost incurred in the first-stage bidding decision (the bidding decision made before the disclosure of uncertainty), while the net cost of real-time market transactions is the cost incurred in the second-stage bidding decision (adjusting the bidding decision according to the actual market conditions). Day-ahead market power bidding decisions refer to the centralized trading market held the day before the power delivery date. Participants in day-ahead market power trading (such as power plants, virtual power plants, or users) submit power purchase bid curves or power sales bid curves based on their forecasts of power supply and demand for the next day. The market operator then clears the market and determines the hourly trading volume and price for the next day. Real-time market power bidding decisions refer to the adjustment of bidding strategies on the day of power delivery to balance the deviation between actual power consumption and day-ahead plans (such as forecast errors, unit failures, etc.). Participants adjust their bidding strategies according to real-time supply and demand conditions.

[0100] The net cost of day-ahead market transactions is calculated based on the day-ahead market electricity bidding parameters, which include the day-ahead market electricity purchase price parameters, the day-ahead market electricity purchase volume parameters, the day-ahead market electricity sales price parameters, and the day-ahead market electricity sales volume parameters. The net cost of real-time market transactions is calculated based on the real-time market electricity bidding parameters, which include the real-time market electricity purchase price parameters, the real-time market electricity sales price parameters, the real-time market electricity purchase volume parameters, and the real-time market electricity sales volume parameters.

[0101] In step S301 of some embodiments, the electricity purchase price parameter c is used as a reference. DA,b,s (t), Daily market electricity purchase parameter P DA,b,s The current market electricity sales price parameter c DA,se,s (t) and the day-to-day market electricity sales parameter P DA,se,s Calculate the day-ahead net market transaction cost C of the virtual power plant in each scenario s. DA,s The net cost of the day-to-day market transactions, C DA,s As shown in formula (4) below:

[0102]

[0103] Among them, C DA,s c represents the net cost of day-ahead market transactions in scenario s. DA,b,s (t) represents the day-ahead market electricity price for electricity purchased in the day-ahead market during time period t under scenario s, P DA,b,s c represents the day-ahead market purchase volume within time period t under scenario s. DA,se,s (t) represents the day-ahead market electricity price for electricity sold in the market during time period t under scenario s, P DA,se,s This represents the day-ahead market electricity sales volume within time period t under scenario s. It should be noted that the day-ahead market electricity price c for electricity sold within time period t under scenario s... DA,Se,s (t) can also be expressed as the ratio coefficient λ between the day-ahead market electricity sales price and the day-ahead market electricity purchase price (usually the sales price is lower than the purchase price) and c. DA,b,s The product of (t) is calculated.

[0104] In step S302 of some embodiments, based on the real-time market electricity purchase price parameter c RT,b,s (t), Real-time market electricity price parameter c RT,se,s (t), Real-time market electricity purchase parameters P RT,b,s and real-time market electricity sales parameters P RT,se,s Calculate the real-time net market transaction cost C of the virtual power plant in each scenario s. RT,s The obtained real-time market transaction net cost C RT,s As shown in formula (5) below:

[0105]

[0106] Among them, C RT,s c represents the real-time net cost of market transactions in scenario s. RT,b,s (t) represents the real-time market electricity price for electricity purchased in the real-time market during time period t under scenario s, P RT,b,s c represents the real-time market purchase volume within time period t under scenario s. RT,se,s (t) represents the real-time market electricity price for electricity sold in the real-time market during time period t under scenario s, P DA,se,s This represents the real-time market sales volume within time period t under scenario s.

[0107] In step S303 of some embodiments, the battery loss cost C corresponding to each scenario s is constructed through the aforementioned steps. B,s Total power generation cost of generator set C G,s Carbon emission costs The current market transaction net cost C DA,s and real-time market transaction net cost C RT,s Then, based on each cost sub-item, a scenario total cost function C can be constructed for the virtual power plant in each scenario s. s The total cost function C for the virtual power plant in each scenario s is... s The specific calculation method is shown in the following formula (6):

[0108]

[0109] Furthermore, the total cost function C of the virtual power plant in each scenario s is calculated. s Then, based on the total cost function C of the scenario... s The total operating cost function of the virtual power plant is constructed, and the objective function of the total operating cost represents minimizing the weighted expected value of the operating cost for all possible scenarios. The total operating cost function of the virtual power plant is constructed using the following formula (7):

[0110]

[0111] Among them, F Cost Let π be the total operating cost function of the virtual power plant. S Let s be the scenario probability corresponding to scenario s. The objective of the total operating cost function is to minimize the total operating cost of the virtual power plant in the day-ahead market.

[0112] In step S102 of some embodiments, a conditional value-at-risk function is constructed based on the value-at-risk parameter, the preset confidence value, the total cost function of the scenario, and the corresponding scenario probability. Step S102 will be described in detail below.

[0113] After constructing the total operating cost function of the virtual power plant through step S101 and its sub-steps, the second objective function of this disclosure embodiment can be constructed, namely the objective function corresponding to Conditional Value at Risk (CVaR). The Conditional Value at Risk is used to quantify the financial risks that the virtual power plant needs to bear when bidding in the day-ahead market.

[0114] In its original definition, Conditional Value at Risk (VaR) is a risk measurement technique that measures the tail risk of a portfolio. It is defined as the average expected value of a loss exceeding the value at risk (VaR) in a continuous distribution of a loss random variable at a given confidence level (a given confidence value α∈(0,1), for example, α is 90% or 95%).

[0115] In this embodiment, CVaR (or α-CVaR, representing the conditional value of risk with a confidence level of α) and VaR (or α-VaR, representing the value of risk with a confidence level of α) are used to construct the objective function of the conditional value of risk. Since the calculation of CVaR depends on VaR as a benchmark, that is, VaR needs to be calculated first and then its tail mean needs to be calculated. In this embodiment, VaR is represented by a risk measurement auxiliary variable ξ, and then the calculation expression of CVaR is constructed based on the risk measurement auxiliary variable ξ. VaR is an α quantile of a loss distribution. The loss distribution describes the statistical distribution of the amount of loss and the probability of its occurrence. For example, the total operating cost (considered as loss) of a virtual power plant in a specific scenario within a certain time period may have various values ​​and probabilities. The α quantile refers to the probability that the loss random variable Z is less than or equal to this value (VaR) is exactly α. That is, the probability that the loss random variable has α will not exceed VaR, and the probability that it has 1-α will exceed VaR. The specific calculation formula of α-VaR is shown in the following formula (8):

[0116] α-VaR=inf(c∈R:P r (z≤c)≥α) (8)

[0117] Where c represents the loss threshold, i.e., the VaR value, z represents the loss random variable, α represents the confidence level, and P r (z≤c) represents the probability that the loss random variable z does not exceed the VaR value, P r (z≤ξ)≥α means that the probability that the loss random variable z does not exceed the VaR value is greater than or equal to α. inf(·) means the infimum operation. The minimum loss threshold c that satisfies the above probability conditions can be used to obtain α-VaR.

[0118] Unlike VaR, CVaR is a coherent risk measure with better mathematical properties. CVaR can be effectively incorporated into the optimization problem through linear constraints, while VaR requires nonlinear constraints for modeling. Therefore, in this embodiment, the total operating cost function is used as the loss random variable z. Unlike the case of continuous distribution, the total operating cost of the virtual power plant is discretely distributed across the scenario set, and the scenario total cost function C... s It simulates uncertainty (such as electricity price fluctuations and wind and solar power generation) through a finite set of scenarios s, and each scenario s has a total scenario cost C. s and the corresponding scenario probability π s This means that the total cost is not continuous, but rather represents a finite number of discrete points, each corresponding to a probability π. s .

[0119] In this embodiment of the disclosure, in order to incorporate CVaR into the optimization problem of the total operating cost of the virtual power plant, it can be achieved by minimizing the following objective function:

[0120]

[0121] Where ξ is the Value at Risk parameter (i.e., the auxiliary variable for risk measurement), which means the optimization variable corresponding to the Value at Risk, including multiple candidate VaRs, and α is the preset confidence level.

[0122] The definition of CVaR itself is difficult to handle directly in optimization problems, especially when it is used as an objective or constraint. Therefore, Equation (9) provides an equivalent expression for the calculation of CVaR, and adopts the convex optimization reconstruction method to transform the total cost function C of each scenario s into a single function. s Treat it as a loss function, and then use the scene probability π s We then perform a weighted average of these loss values ​​and, by introducing the risk metric auxiliary variable ξ, transform the nonlinear CVaR from a probabilistic definition into a linear form in an optimization problem. In other words, we transform the problem of finding the minimum CVaR into a problem of minimizing the risk metric auxiliary variable ξ. Equation (9) means finding an optimal solution ξ. * Make Minimum, ξ * The final solution is the Value at Risk α-VaR, which is the minimum loss threshold that may be exceeded at the confidence level α.

[0123] Furthermore, due to the max function in formula (9), i.e. max(0,C) sThe max function (ξ) is non-smooth and difficult to solve in optimization models. By introducing the risk metric auxiliary variable ξ and linear inequality constraints, the max function is equivalently replaced, achieving a linear reconstruction of the maximum operator. This transforms the CVaR optimization problem into an easily solvable linear programming problem. Specifically, max(0,C) is transformed into a linear programming problem. s By performing linear reconstruction of -ξ), the following conditional risk value function and corresponding linear constraints can be obtained, as shown in formulas (10) to (12):

[0124]

[0125]

[0126] Where, ΔC s The function C represents the total cost of a scenario in scenario s. s The portion exceeding ξ, and ΔC s F must be greater than 0. CVaR The conditional risk value function (i.e., the objective function corresponding to the conditional risk value) is represented by Equations (11) and (12). Equations (11) and (12) provide a set of linear rules (constraints), which enable the calculation of the originally complex CVaR risk index using an easily solvable linear model, greatly reducing the computational complexity and enabling efficient processing of CVaR in power bidding strategy optimization models that contain a large number of scenarios and decision variables.

[0127] The method for constructing the objective function corresponding to the conditional value at risk (CVaR) provided in this disclosure achieves linearized modeling and efficient optimization of the risk metric CVaR by minimizing the risk metric auxiliary variable ξ and scaling. The optimized form of the sum of expected excess losses is equivalent to defining the core calculation logic of CVaR (Formula (9)); then, by introducing the scenario auxiliary variable ΔC s Linear inequality constraints (Equations (11) and (12)) are applied to transform the nonlinear max function in Equation (9) into a linearly solvable form. This reconstruction enables the CVaR risk objective to be seamlessly embedded into the multi-objective power bidding strategy optimization model of the virtual power plant, transforming the complex risk control problem into a standard mathematical programming problem, thereby significantly improving the computational efficiency of the model while ensuring the accuracy of risk quantification.

[0128] In step S103 of some embodiments, the system operation constraints and electricity market bidding constraints of the virtual power plant are obtained, and an optimization model for the day-ahead market electricity bidding strategy of the virtual power plant is constructed based on the system operation constraints, electricity market bidding constraints, total operating cost function and conditional risk value function. Step S103 will be described in detail below.

[0129] In this embodiment of the disclosure, in addition to the expression of the objective function of CVaR after linear transformation constructed through the aforementioned steps and the linear constraints of its corresponding inequalities, when constructing the day-ahead market electricity bidding strategy optimization model for the virtual power plant, it is also necessary to consider the physical constraints in the power system (i.e., system operation constraints), such as physical constraints of battery energy storage systems, generator sets, and power balance.

[0130] Specifically, the system operation constraints include the following constraints: battery initial energy state constraints, battery dynamic energy transfer constraints, battery energy boundary constraints, battery charging power limits, battery discharging power limits, daily battery charging total constraints, daily battery discharging total constraints, generator capacity constraints, and system power balance constraints. Among them, the energy balance within the system can be expressed as the sum of battery energy storage system discharge, generator power generation, photovoltaic power generation, day-ahead market purchases, and real-time market purchases should be equal to the sum of battery energy storage system charging, electricity demand, day-ahead market sales, and real-time market sales, which satisfies the law of conservation of energy.

[0131] Furthermore, it is also necessary to consider the electricity market bidding constraints in the electricity bidding process. Specifically, the electricity market bidding constraints include: (1) the monotonicity constraint of the day-ahead market electricity purchase price (the electricity purchase bidding curve must meet the constraint of "the higher the price, the less electricity purchased"); (2) the monotonicity constraint of the day-ahead market electricity sales price (the electricity sales bidding curve must meet the constraint of "the higher the price, the more electricity sold"); (3) active bidding point constraints: when the bidding point is in the "active" state, its bidding increment should be between the minimum and maximum limits (if a bidding point is activated (selected), its electricity increment must be within the minimum and maximum values ​​specified by the market to ensure that the electricity in the bidding segment is dispatchable); (4) bidding point quantity limit: the number of bidding points of each electricity purchase or sales bidding curve at any time does not exceed the maximum value specified by the market (the electricity purchase / sales bidding curve in a single time period can contain the maximum number of valid bidding points specified by the market); (5) mutually exclusive buying and selling logic constraints: electricity purchase quotations and electricity sales quotations cannot be submitted at the same time period.

[0132] After obtaining the system operation constraints and electricity market bidding constraints of the virtual power plant, an optimization model for the day-ahead market electricity bidding strategy of the virtual power plant is constructed based on the system operation constraints, electricity market bidding constraints, the total operating cost function expressed by formula (7), the conditional risk value function expressed by formula (10), and their corresponding linear constraints (formulas (11) and (12)). The day-ahead market electricity bidding strategy optimization model is used to solve the multi-objective bidding strategy optimization problem of minimizing the total operating cost of the virtual power plant and minimizing risk (CVaR). The constructed multi-objective bidding strategy optimization problem can be expressed as the following formula (13):

[0133] min x∈Ω (FCost ,F CVaR (13)

[0134]

[0135] Where Ω is the feasible region of the multi-objective optimization problem, Ω={x|Eqs.(1)-(7),Eqs.(10)-(12),system operation constraints,electricity market bidding constraints}, and the multi-objective bidding strategy optimization function to be solved is: When solving the multi-objective bidding strategy optimization function, it is necessary to satisfy the constraints of formulas (10) to (12), system operation constraints, and electricity market bidding constraints.

[0136] In formula (13), x represents the vector of all decision variables. The decision vector includes the electricity purchase bid curve and the electricity sales bid curve in the hourly day-ahead market for electricity trading, the hourly charging scheduling strategy of the battery energy storage system, the hourly discharging scheduling strategy of the battery energy storage system, the hourly scheduling strategy of the generator set, real-time market transactions, and risk measurement auxiliary variables. Specifically, the electricity purchase bid curve in the hourly day-ahead market for electricity trading includes c DA,b,s (t) and P DA,b,s The hourly day-ahead electricity sales bid curve includes c DA,se,s (t) and P DA,se,S The hourly charging scheduling strategy for the battery energy storage system includes P B,c,s (t), the hourly discharge scheduling strategy of the battery energy storage system includes P B,d,s (t), the hourly scheduling strategy for generator sets includes P G,s (t), real-time market transactions include c RT,b,s (t), c RT,se,s (t), P RT,b,s and P RT,se,s The auxiliary variables for risk measurement include ξ.

[0137] In some embodiments, please refer to Figure 4 An optimization model for the day-ahead market electricity bidding strategy of a virtual power plant is constructed based on system operation constraints, electricity market bidding constraints, total operating cost function, and conditional risk-value function, including but not limited to the following steps S401 to S404:

[0138] Step S401: Construct a single-objective optimization model for total operating cost based on the total operating cost function, system operating constraints, and electricity market bidding constraints; and construct a single-objective optimization model for conditional risk value based on the conditional risk value function, risk value parameters, scenario total cost function, system operating constraints, and electricity market bidding constraints.

[0139] Step S402: Obtain the risk tolerance parameter, and construct the conditional risk value constraint based on the risk tolerance parameter and the conditional risk value function;

[0140] Step S403: Based on conditional risk value constraints, system operation constraints, electricity market bidding constraints, conditional risk value function, risk value parameters, scenario total cost function, and operating total cost function, construct an electricity bidding strategy optimization model that balances conditional risk value and operating total cost.

[0141] Step S404: Construct a day-ahead market electricity bidding strategy optimization model for the virtual power plant based on the single-objective optimization model of total operating cost, the single-objective optimization model of conditional risk value, and the electricity bidding strategy optimization model that balances conditional risk value and total operating cost.

[0142] In step S401 of some embodiments, the single-objective problem SP-Cost is defined as a special case of the multi-objective problem, used to minimize F Cost (Therefore F is ignored) CVaR The specific definition of SP-Cost is shown in formula (14):

[0143] min x F Cost =∑ s π s ·C s (14)

[0144] System operation constraints and electricity market bidding constraints

[0145] SP-Cost is a single-objective optimization model for total operating cost, min x F Cost =∑ s π s ·C s The objective function is defined by system operation constraints and electricity market bidding constraints.

[0146] Meanwhile, the single-objective problem SP-CVaR is defined as a special case of the multi-objective problem, used to minimize F. CVaR (Therefore F is ignored) Cost The specific definition of SP-CVaR is shown in formula (15):

[0147]

[0148] SP-CVaR is a single-objective optimization model for conditional value at risk. The objective function is defined by the following constraints: Equations (11) and (12), system operation constraints, and electricity market bidding constraints.

[0149] This disclosure transforms the abstract concept of "risk minimization" into a solvable linear optimization problem, avoiding the issue of risk optimization models in related technologies relying on subjective weight updates. By defining and solving SP-Cost, the lower bound of the theoretical total operating cost of the virtual power plant is revealed, providing a reference for negotiation space in bidding strategies. Two-terminal solutions intuitively demonstrate the consequences of extreme power bidding strategies, helping power plant operators understand "how much risk they are willing to take in exchange for cost savings," thus improving the transparency of power bidding decisions. Through the synergy of SP-Cost and SP-CVaR, a complete set of power bidding strategies for the virtual power plant, ranging from "aggressive arbitrage" to "absolute hedging," can be generated, supporting data-driven risk preference decisions.

[0150] In step S402 of some embodiments, to resolve the conflict between multiple objectives, the multi-objective power bidding strategy optimization problem (Equation (13)) is transformed into a single-objective problem using the ε-constraint method. The ε-constraint method transforms all objective functions except for one objective function into inequality constraints. In this embodiment, the conditional risk-value function F is selected as... CVaR Write it as an inequality constraint and set F. CVaR The upper bound is ε, and the total operating cost is F. Cost The conditional risk value constraint, which remains in the objective function of the power bidding strategy optimization problem, is shown in the following formula (16):

[0151] F CVaR ≤ε (16)

[0152] Formula (16) represents the conditional risk value constraint, where ε represents the risk tolerance parameter, used to quantify the decision-maker's acceptance of risk, ε is the upper limit constraint value of the conditional risk value, and represents the maximum risk level acceptable to the decision-maker. In the optimization problem of multi-objective power bidding strategy, the conditional risk value function is transformed into F CVaR Transform into constraint F CVaR The approach ≤ε transforms the original problem into a single-objective optimization problem that minimizes total operating cost, ensuring manageable risk while optimizing total operating cost. By adjusting ε, multiple sets of day-ahead market electricity bidding solutions for virtual power plants are generated. Each ε corresponds to a power bidding strategy that balances total operating cost and conditional risk value, achieving a trade-off between total operating cost and conditional risk value. The specific solution steps will be introduced later.

[0153] In step S403 of some embodiments, an optimization model for power bidding strategy that balances conditional risk value and total operating cost can be constructed based on conditional risk value constraints, system operation constraints, power market bidding constraints, conditional risk value function, risk value parameter, scenario total cost function, and operating total cost function. Specifically, after transforming the multi-objective power bidding strategy optimization problem represented by formula (13) using the ε-constraint method, a single-objective problem (ε-SP) based on risk tolerance parameter can be obtained, which can be expressed as:

[0154]

[0155] Based on the reformulated single-objective problem ε-SP, an optimization process is applied to generate a set of power bidding strategies for virtual power plants, balancing two conflicting objectives.

[0156] In step S404 of some embodiments, after constructing the single-objective optimization model of total operating cost, the single-objective optimization model of conditional value of risk, and the power bidding strategy optimization model that balances conditional value of risk and total operating cost through the aforementioned steps, the final day-ahead market power bidding strategy optimization model of the virtual power plant can be obtained. Subsequently, the day-ahead market power bidding strategy optimization model will be solved by solving these three optimization models respectively.

[0157] In step S104 of some embodiments, the day-ahead market electricity bidding strategy optimization model is solved to obtain the day-ahead market electricity bidding strategy of the virtual power plant. Step S104 will be described in detail below.

[0158] In this embodiment, the day-ahead market electricity bidding strategy for virtual power plants includes at least optimizing day-ahead market electricity bidding parameters. Optimizing day-ahead market electricity bidding parameters may include optimizing day-ahead market electricity sales price parameters, optimizing day-ahead market electricity purchase price parameters, optimizing day-ahead market electricity sales parameters, and optimizing day-ahead market electricity purchase parameters. Before detailing the day-ahead market electricity bidding strategy, the method for constructing the electricity bidding decision model for virtual power plants participating in the day-ahead market provided in this disclosure will be introduced first.

[0159] In some embodiments, please refer to Figure 5 The day-ahead market electricity bidding strategy optimization model is solved to obtain the day-ahead market electricity bidding strategy for the virtual power plant, including but not limited to the following steps S501 to S505:

[0160] Step S501: Construct a scenario-based electricity purchase bidding curve model based on the day-ahead market electricity purchase price parameters and the day-ahead market electricity purchase volume parameters;

[0161] Step S502: Construct a scenario-based electricity sales bidding curve model based on the day-ahead market electricity sales price parameters and the day-ahead market electricity sales volume parameters;

[0162] Step S503: Solve the day-ahead market electricity bidding strategy optimization model to obtain the optimized day-ahead market electricity purchase price parameters, optimized day-ahead market electricity purchase volume parameters, optimized day-ahead market electricity sales price parameters, and optimized day-ahead market electricity sales volume parameters for each scenario;

[0163] Step S504: Based on the optimized day-ahead market electricity purchase price parameters, optimized day-ahead market electricity purchase parameters, and electricity purchase bidding curve model, the day-ahead market electricity purchase bidding strategy of the virtual power plant is obtained; and based on the optimized day-ahead market electricity sales price parameters, optimized day-ahead market electricity sales parameters, and electricity sales bidding curve model, the day-ahead market electricity sales bidding strategy of the virtual power plant is obtained.

[0164] Step S505: Determine the day-ahead market power bidding strategy for the virtual power plant based on the day-ahead market power purchase bidding strategy and the day-ahead market power sales bidding strategy.

[0165] Specifically, in this embodiment of the disclosure, the day-ahead market electricity bidding strategy BC is defined as a set of hourly day-ahead market electricity bidding curves obtained through random optimization. The set of electricity bidding curves includes a scenario-based electricity purchase bidding curve model and a scenario-based electricity sales bidding curve model, wherein the electricity purchase bidding curve model is used to purchase electricity, and the electricity sales bidding curve model is used to sell electricity. The specific representation of the set of electricity bidding curves is shown in the following formula (18):

[0166] BC = {BC} b (t),BC se (t)|t∈T} (18)

[0167] Among them, BC b (t) represents the virtual power plant's electricity purchase and bidding curve model in time period t, BC se (t) represents the electricity sales bidding curve model of the virtual power plant in time period t, where T represents the total number of time periods in the virtual power plant's scheduling cycle.

[0168] In step S501 of some embodiments, the electricity purchase price parameter c is used as a reference. DA,b,s (t) and the day-to-day market electricity purchase parameter P DA,b,s Construct a power purchase bidding curve model BC based on scenario s b The expression for (t) is shown in formula (19) below:

[0169] BC b (t)={(P DA,b,s ,c DA,b,s (t))|s∈S∧θ DA,b,s (t)=1} (19)

[0170] Where S represents the scene set, BC b(t) is derived from the point set (P) DA,b,s ,c DA,b,s (t)) constitutes, θ DA,b,s (t) = 1 indicates that under the day-ahead market electricity price scenario s, the amount of electricity purchased at time t increases incrementally in the electricity purchase bidding curve, while 0 indicates the opposite; simultaneously, θ DA,b,s (t) = 1 also indicates that the electricity purchase bidding point at scenario s and time t is selected as a "valid point" of the bidding curve, used to construct a progressively increasing bidding curve, if θ DA,b,s If (t) = 0, then this point does not participate in the formation of the bidding curve. Thus, there is only one electricity purchase bidding curve at each time t, which is composed of multiple "effective price-electricity bidding points". This preserves the probabilistic information of the optimization result and conforms to market rules.

[0171] Specifically, for formula (19), for each time period t, we can first determine the electricity purchase bidding decision (including the day-ahead market electricity purchase parameter P) under each scenario s. DA,b,s And the current market electricity purchase price parameter c DA,b,s (t) is transformed into discrete points on the corresponding electricity purchase bidding curve, then summarized to obtain a candidate point set, and then the point sequence that satisfies the monotonicity of electricity purchase (the amount of electricity bid decreases when the price increases, and the amount of electricity is not increasing with the price) is selected by 0-1 variables. This not only preserves the probabilistic information of the optimization result, but also conforms to market rules.

[0172] In step S502 of some embodiments, according to the day-ahead market electricity price parameter c DA,se,s (t) and the day-to-day market electricity sales parameter P DA,se,s Constructing a power sales bidding curve model BC based on scenario s se The expression for (t) is shown in formula (20) below:

[0173] BC se (t)={(P DA,se,s ,c DA,se,s (t))|s∈S∧θ DA,se,s (t)=1} (20)

[0174] Where S represents the scene set, BC se (t) is derived from the point set (P) DA,se,s ,c DA,se,s (t)) constitutes, θ DA,se,s (t) = 1 indicates that under the day-ahead market electricity price scenario s, the electricity sales bid volume at time t shows an incremental increase in the electricity sales bid curve, while 0 indicates the opposite; simultaneously, θ DA,se,s (t) = 1 also indicates that the electricity sales bidding point at scenario s and time t is selected as a "valid point" of the bidding curve, used to construct a progressively increasing bidding curve, if θ DA,se,sIf (t) = 0, then this point does not participate in the formation of the bidding curve.

[0175] Specifically, for formula (20), for each time period t, we can first determine the electricity sales bidding decision (including the day-ahead market electricity sales parameter P) under each scenario s. DA,se,s And the current market electricity sales price parameter c DA,se,s The points (t) are transformed into discrete points on the corresponding electricity sales bidding curve, then summarized to obtain a candidate point set. Finally, a sequence of points satisfying the monotonicity of electricity sales (price increases with bidding volume, and electricity volume does not decrease with price) is selected using 0-1 variables. Thus, there is only one electricity sales bidding curve at each time t, which is composed of multiple "effective price-electricity bidding points." This preserves the probabilistic information of the optimization result while conforming to market rules.

[0176] In step S503 of some embodiments, by solving the day-ahead market electricity bidding strategy optimization model constructed in the aforementioned steps, the optimized decision variables for each scenario s can be obtained. These specifically include optimized day-ahead market electricity purchase price parameters, optimized day-ahead market electricity purchase quantity parameters, optimized day-ahead market electricity sales price parameters, and optimized day-ahead market electricity sales quantity parameters. Simultaneously with solving for these optimized parameters, the scheduling strategies for related equipment in the power system of the virtual power plant can also be solved. For example, optimized decision variables such as optimized charging power parameters and optimized discharging power parameters of the battery energy storage system, optimized generator set power parameters of the generator set, and optimized carbon emission parameters can be obtained.

[0177] In step S504 of some embodiments, the bidding points that satisfy the monotonicity of electricity purchase are constructed based on the optimized day-ahead market electricity purchase price parameters and the corresponding optimized day-ahead market electricity purchase quantity parameters obtained by solving. Based on the electricity purchase bidding curve model constructed in the aforementioned steps, the day-ahead market electricity purchase bidding strategy of the virtual power plant can be obtained.

[0178] Furthermore, based on the optimized day-ahead market electricity sales price parameters and the corresponding optimized day-ahead market electricity sales parameters obtained from the solution, a bidding point that satisfies the monotonicity of electricity sales is constructed. Based on the electricity sales bidding curve model constructed in the aforementioned steps, the day-ahead market electricity sales bidding strategy of the virtual power plant is obtained.

[0179] In step S505 of some embodiments, the day-ahead market power bidding strategy of the virtual power plant is determined according to the day-ahead market power purchase bidding strategy and the day-ahead market power sales bidding strategy. It should be noted that, as described above, each risk tolerance parameter can correspond to a set of day-ahead market power purchase bidding strategies and day-ahead market power sales bidding strategies. Therefore, the final generated day-ahead market power bidding strategy of the virtual power plant includes a combination of multiple day-ahead market power purchase bidding strategies and day-ahead market power sales bidding strategies.

[0180] In some embodiments, step S104 includes, but is not limited to, the following steps:

[0181] The single-objective optimization model for total operating cost is solved to obtain the optimal total operating cost. Based on the optimal total operating cost, the single-objective optimization model for conditional value of risk is solved to obtain the conditional value of risk under the constraint of the optimal total operating cost and the market electricity bidding strategy on the first day.

[0182] The single-objective optimization model of conditional risk value is solved to obtain the optimal conditional risk value. Based on the optimal conditional risk value, the single-objective optimization model of total operating cost is solved to obtain the total operating cost of the virtual power plant under the constraint of the optimal conditional risk value and the market electricity bidding strategy on the second day.

[0183] The risk tolerance range is determined based on the conditional risk value under the optimal total operating cost constraint and the optimal conditional risk value.

[0184] Obtain the preset conditional value at risk step size, minimum conditional value at risk step size, conditional value at risk step size update ratio, and cost convergence threshold;

[0185] Based on the conditional risk value step size, minimum conditional risk value step size, conditional risk value step size update ratio, cost convergence threshold, and risk tolerance interval, the power bidding strategy optimization model that balances conditional risk value and total operating cost, as well as the conditional risk value single-objective optimization model, are iteratively solved until the number of iterations reaches the preset number of iterations. The day-ahead market power bidding strategy corresponding to the virtual power plant is determined based on the optimized day-ahead market power bidding parameters obtained from each iteration.

[0186] Specifically, in this embodiment of the disclosure, the single-objective optimization model of total operating cost shown in formula (14) is first solved, that is, the single-objective problem SP-Cost is solved, and the decision variables include the battery charging and discharging strategy (scenario charging power parameter P). B,c,s (t), Scene discharge power parameter P B,d,s (t)), generator output (P) G,S (t)), day-ahead market electricity purchase bid curve (P) DA,b,s c DA,b,s (t)), the day-ahead market electricity sales bidding curve (P) DA,se,s c DA,se,s (t)), real-time market trading volume (c) RT,b,s (t), P RT,b,s c RT,se,s (t) and P RT,se,s The constraints include system operation constraints and electricity market bidding constraints.

[0187] Solve the optimization problem SP-Cost, minimizing the total running cost while ignoring risk indicators, to obtain the cost-optimal baseline solution. The expression (21) is as follows:

[0188]

[0189] Furthermore, while maintaining the minimized total operating cost without deterioration, we further minimize the conditional risk value, as shown in expression (22):

[0190] constraint:

[0191] In formula (22), SP-CVaR is a single-objective optimization problem that only optimizes CVaR, as shown in formula (15). Based on the constraints shown in formula (15), additional constraints are added. The constraints. At this point, the virtual power plant's first-day market electricity bidding strategy BC1 and the objective of minimizing CVaR are... The solution will be found by referring to formula (18), where the first day-ahead market electricity bidding strategy BC1 includes the day-ahead market electricity purchase bidding curve. and the recent market electricity sales bidding curve

[0192] Thus, the first non-dominated bid solution can be obtained, used to anchor the low-cost endpoint of the Pareto front. In this step, while ensuring... Under the premise of further minimizing F CVaR In other words, the goal is to find a "more conservative but low-cost" solution that minimizes risk without increasing costs as much as before.

[0193] Thus, by solving this problem, the total cost is minimized, even though the risk may be extremely high. This step, ignoring risk constraints, minimizes the total operating cost by solving for SP-Cost, obtaining the theoretically lowest solution for the total operating cost. This reveals the economic potential of the virtual power plant's power system and provides a lower bound for the total operating cost in subsequent analyses that weigh total operating cost against conditional risk value, assuming a stable market electricity price and a highly risk-tolerant operating environment.

[0194] Furthermore, the single-objective optimization model of conditional value of risk (SP-CVaR) is solved to obtain the optimal conditional value of risk and generate the baseline solution with the lowest risk. Specifically, the optimization problem with CVaR as the sole objective (i.e., SP-CVaR) is solved, ignoring the total operating cost, to obtain the baseline solution with the lowest risk. The problem to be solved is shown in expression (23):

[0195]

[0196] In this step, the total operating cost F is completely ignored. Cost In the case of minimizing risk F CVaR The most conservative, lowest-risk virtual power plant day-ahead market electricity bidding strategy was obtained, but F Cost It can be very large (high cost), where |J| represents the total number of day-ahead market power bidding strategies to be generated. For example, |J| is the number of candidates for the day-ahead market power bidding strategy. For example, when 5 candidate schemes are generated, |J| = 5.

[0197] Based on this risk-optimal solution, a CVaR upper bound constraint is introduced to further minimize the total operating cost. The solution is shown in expression (24):

[0198] constraint:

[0199] This step optimizes the total operating cost objective to ensure further optimization based on minimizing CVaR, but without worsening the risk (not exceeding the previous minimum). The solution generated in this step serves as the high-risk endpoint of the Pareto front. SP-Cost in Equation (24) is a single-objective optimization problem that optimizes only the total operating cost, as shown in Equation (14), with additional constraints added to the conditions shown in Equation (14). The constraints. At this point, the virtual power plant's second-day-ahead market electricity bidding strategy BC... |J| and the goal of minimizing total operating costs The solution will be found, where please refer to formula (18), the market electricity bidding strategy BC the day before the second day. |J| Including the current market power purchase bidding curve and the recent market electricity sales bidding curve

[0200] As described above, the embodiments of this disclosure employ the ε-constraint method, transforming the conditional risk value from the objective function into progressively tightening constraint terms. This ensures that each day-ahead market electricity bidding scheme, while optimizing total operating costs, is controlled by a specific risk tolerance ε. j Where j represents a specific power bidding solution index, that is, the j-th generated power bidding strategy, by enumerating different ε j This resulted in a set of day-ahead market electricity bidding solutions for virtual power plants with different trade-offs between "total operating cost and conditional risk value". Specifically, the process involves first finding two extreme solutions (optimal total operating cost and optimal risk) through the aforementioned steps, and then generating a series of equilibrium solutions by progressively adjusting the ε value between them. The process of obtaining this series of equilibrium solutions will be described in detail below.

[0201] Solve for the conditional value of risk under the optimal total operating cost constraint. and optimal conditional value of risk Then, a risk tolerance range can be determined, where the risk tolerance range is determined by the risk tolerance parameter ε. in It is the starting point of the risk tolerance range and the minimum conditional risk value. It is the end of the risk tolerance range.

[0202] Then you can obtain the preset conditional value-at-risk step size Δε. j and minimum conditional value at risk step size Δε min Among them, the conditional value-at-risk step size Δε is set. j The initial value is Δε j =ε init Then, obtain the conditional value at risk step size update ratio η and the cost convergence threshold τ. cost Furthermore, the ε-SP (Equivalent-to-Conditional-Risk-Value) power bidding strategy optimization model balancing conditional risk value and total operating cost, as well as the SP-CVaR (Conditional-Risk-Value Single-Objective Optimization Model), can be iteratively solved based on the conditional risk value step size, minimum conditional risk value step size, conditional risk value step size update ratio, cost convergence threshold, and risk tolerance interval, until the preset number of iterations is reached. The day-ahead market power bidding strategy corresponding to the virtual power plant is then determined based on the optimized day-ahead market power bidding parameters obtained from each iteration. Here, the iteration number is j, the preset number of iterations is |J|, and each solution yields an ε... j And its corresponding day-ahead market electricity bidding strategy.

[0203] In some embodiments, the iterative solution process corresponding to each iteration number includes, but is not limited to, the following steps:

[0204] The current risk tolerance threshold is determined based on the preconditional risk value, and the total operating cost of the power bidding strategy optimization model under the current risk tolerance threshold constraint is calculated based on the current risk tolerance threshold to solve the balance conditional risk value and total operating cost.

[0205] The conditional risk value is solved based on the total operating cost under the current risk tolerance threshold constraint, resulting in the conditional risk value under the current risk tolerance threshold and the day-ahead market electricity bidding strategy for virtual power plants.

[0206] The risk tolerance parameters are updated based on the total operating cost of the preceding operation, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio.

[0207] In this embodiment, as obtained from the aforementioned steps, when j = 1, it indicates that only the total operating cost is optimized, ignoring risk (CVaR), resulting in the "most cost-effective" electricity bidding strategy. When j = |J|, it indicates that only CVaR is optimized, ignoring the total operating cost, resulting in the "most secure" solution. When j = 2, ..., |J|-1, the solution yields multiple "compromise" solutions by balancing the total operating cost and CVaR. The iterative solution process corresponding to each iteration number is to solve for the electricity bidding strategy corresponding to j = 2, ..., |J|-1.

[0208] Specifically, the current risk tolerance threshold can be determined first based on the preconditional risk value, where the preconditional risk value is the conditional risk value obtained from the previous solution of the single-objective optimization model (SP-CVaR) for conditional risk value, and the current risk tolerance threshold is ε. j For example, the conditional value of risk obtained when the preconditional value of risk is j=1. When solving for the power bidding strategy when j=2, the conditional value of risk obtained when j=1 is used. Determine the current risk tolerance threshold ε2.

[0209] Then, the total operating cost of the power bidding strategy optimization model under the current risk tolerance threshold constraint is solved by balancing the risk value and total operating cost. The specific solution is shown in the following expression (25):

[0210] Constraint: ε = ε j (25)

[0211] in, This represents the total operating cost obtained from the j-th solution. For example, let's illustrate expression (25) with the conditional risk value obtained when the preconditional risk value is j=1. When solving for the power bidding strategy when j=2, the conditional value of risk obtained when j=1 is used. Determine the current risk tolerance threshold ε1. Under the constraints of the ε-SP optimization problem as shown in formula (17), add the current risk tolerance threshold constraint to solve the power bidding strategy optimization model (ε-SP) of the equilibrium condition risk value and total operating cost, that is, solve for F when j=2. Cost The expression is as follows:

[0212] Constraint: ε = ε1

[0213] Furthermore, it can be based on the current risk tolerance threshold ε j Total operating cost under constraints Solving the single-objective optimization model of conditional value at risk (SP-CVaR) yields the conditional value at risk under the current risk tolerance threshold. and the day-ahead market electricity bidding strategy for virtual power plants BC j The specific solution expression (26) is as follows:

[0214] constraint:

[0215] in, This represents the conditional risk value obtained from the j-th solution, and this step maintains the above total operating cost. Without degradation, further minimizing CVaR reduces risk even more, representing the best risk outcome without cost loss. By solving for SP-Cost, the scheduling strategy for power equipment in the virtual power plant and the total scenario cost C can be determined. s , The specific solution method is as follows: According to formulas (11) and (12), multiple candidate values ​​of ξ can be traversed within the preset range of the value-at-risk parameter ξ to calculate C. s -ξ, determines that C s -All ξ values ​​greater than or equal to 0, and then among all ξ values ​​that satisfy the aforementioned requirements, determine the value of C. s The minimum value of ξ is obtained by calculating the value of -ξ using formula (10).

[0216] Furthermore, it can be based on the current risk tolerance threshold ε j The improvement of the total operating cost under constraints compared to the previous total operating cost is adaptively updated based on the risk tolerance threshold and the step size of the conditional risk value. The previous total operating cost is the total operating cost obtained from the previous solution of the power bidding strategy optimization model that balances the conditional risk value and the total operating cost.

[0217] In some embodiments, the value at risk step update ratio includes the step increase ratio η. up and step size reduction ratio η down The risk tolerance parameters are updated based on the total operating cost of the preceding operation, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio, including but not limited to the following steps:

[0218] When the difference between the total operating cost under the current risk tolerance threshold constraint and the previous total operating cost exceeds the cost convergence threshold, the conditional risk value step size is updated according to the step size increase ratio to obtain the updated conditional risk value step size.

[0219] Alternatively, when the difference between the total operating cost under the current risk tolerance threshold constraint and the previous total operating cost does not exceed the cost convergence threshold, the conditional risk value step size is updated according to the step size reduction ratio and the minimum conditional risk value step size to obtain the updated conditional risk value step size.

[0220] The risk tolerance parameter is updated based on the updated conditional value at risk step size to obtain the updated risk tolerance parameter.

[0221] Specifically, in this embodiment of the disclosure, the current risk tolerance threshold ε can be obtained. j Total operating cost under constraints and total operating costs of preceding processes Calculate the absolute value of the difference between the two, and then compare the calculated absolute value of the difference with the cost convergence threshold τ. cost When comparing, When this occurs, it indicates that there is significant room for compromise and the step size needs to be increased. The step size can be increased proportionally by η. up Conditional Value at Risk Step Δε j Perform an update to obtain the updated conditional value-at-risk step size Δε j+1 Update conditional value at risk step size Δε j+1 The expression (27) is shown below:

[0222] Δε j+1 =Δε j ·η up (27)

[0223] Alternatively, the current risk tolerance threshold ε j Total operating cost under constraints Compared with the total cost of preceding operation The difference between them does not exceed the cost convergence threshold τ Cost When, that is, when When this is the case, it indicates that reducing the step size increases the solution density, and the ratio η can be reduced according to the step size. down and the minimum conditional value at risk step size Δε min Conditional Value at Risk Step Δε j Perform an update to obtain the updated conditional value-at-risk step size Δε j+1 Update conditional value at risk step size Δε j+1 The expression (28) is shown below:

[0224] Δε j+1 =max(Δε) j ·η down ,Δε min (28)

[0225] Formula (28) is used to ensure that the updated conditional value at risk step size is not less than the minimum conditional value at risk step size, taking Δε. j ·η down and Δε min The maximum value in.

[0226] Furthermore, the updated conditional value at risk step size Δε can be used as a basis. j+1 Risk tolerance parameter ε j The updated risk tolerance parameter ε is obtained by performing an update. j+1 And the iterative solution for index j, the specific update expression (29) is as follows:

[0227] ε j+1 =ε j +Δε j+1 ,j←j+1 (29)

[0228] In this step, from j = 2 to |J| - 1, the optimization process is achieved by gradually changing F. CVaR By repeatedly solving the optimization problem (ε-SP) with an upper limit constraint ε, a series of day-ahead market electricity bidding strategies with different risk-cost trade-offs are generated. For example, if |J| = 5, then j = [1, 2, 3, 4, 5]. When j is 1, the strategy purely minimizes cost (without regard to risk); when j is 5, the strategy purely minimizes risk (without regard to cost); when j is 2 or 3, the resulting day-ahead market electricity bidding strategy is an intermediate solution that balances risk and cost; when j is 4 or 5, the resulting day-ahead market electricity bidding strategy places greater emphasis on risk control.

[0229] This step, by dynamically adjusting the risk tolerance threshold ε, avoids problems such as uneven solution set distribution and wasted computational resources that exist in traditional uniform ε sampling, and can generate a more compact and representative day-ahead market electricity bidding strategy solution set.

[0230] This application constructs and solves an optimization model that comprehensively considers the total operating cost of a virtual power plant, as well as the uncertainties caused by wind and solar power generation and day-ahead market electricity prices. This model simultaneously balances the economic viability and risk control capabilities of the virtual power plant's day-ahead market electricity bidding strategy. The application constructs a total operating cost function that integrates multi-dimensional costs and incorporates the differences in photovoltaic output and day-ahead market electricity price fluctuations. It quantifies the uncertainty risk of the electricity market by constructing a conditional value-at-risk model, limiting extreme losses under confidence constraints. This makes the electricity bidding strategy both economical and robust, avoiding significant losses caused by electricity price fluctuations and improving the risk control capabilities of the virtual power plant in the electricity bidding process. Furthermore, the optimization model embeds system operation constraints and electricity market bidding constraints to ensure that the generated day-ahead market electricity bidding strategy is safe to execute and meets market requirements. The final generated day-ahead electricity bidding strategy can effectively perceive market volatility risks while reducing total operating costs, dynamically balancing costs and risks, achieving synergistic optimization of virtual power plant operating costs and operational stability, and enhancing market competitiveness.

[0231] Please see Figure 6 This application also provides a device for optimizing the power bidding strategy of a virtual power plant, which can implement the above-mentioned method for optimizing the power bidding strategy of a virtual power plant. The device includes:

[0232] The generation unit 610 is used to obtain the total cost function of the virtual power plant in each scenario based on the battery status parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters. Based on the total cost function of the scenario and the scenario probability corresponding to the scenario, the total operating cost function of the virtual power plant is generated. The day-ahead market electricity price and / or photovoltaic power generation of the photovoltaic module are different between each two scenarios.

[0233] The first construction unit 620 is used to construct a conditional value-at-risk function based on the value-at-risk parameter, the preset confidence value, the total cost function of the scenario, and the corresponding scenario probability.

[0234] The second building unit 630 is used to obtain the system operation constraints and electricity market bidding constraints of the virtual power plant, and to build an optimization model for the day-ahead market electricity bidding strategy of the virtual power plant based on the system operation constraints, electricity market bidding constraints, total operating cost function and conditional risk value function;

[0235] Solver 640 is used to solve the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy of the virtual power plant. The day-ahead market electricity bidding strategy includes at least optimizing the day-ahead market electricity bidding parameters.

[0236] The specific implementation method of the power bidding strategy optimization device for the virtual power plant is basically the same as the specific implementation method of the power bidding strategy optimization method for the virtual power plant described above, and will not be repeated here.

[0237] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned method for optimizing the power bidding strategy of a virtual power plant. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0238] Please see Figure 7 , Figure 7 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0239] The processor 701 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0240] The memory 702 can be implemented as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 702 and is called and executed by the processor 701 to execute the power bidding strategy optimization method for the virtual power plant of this application embodiment.

[0241] The input / output interface 703 is used to implement information input and output;

[0242] The communication interface 704 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0243] Bus 705 transmits information between various components of the device (e.g., processor 701, memory 702, input / output interface 703, and communication interface 704);

[0244] The processor 701, memory 702, input / output interface 703, and communication interface 704 are connected to each other within the device via bus 705.

[0245] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for optimizing the power bidding strategy of a virtual power plant.

[0246] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0247] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0248] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0249] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0250] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0251] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0252] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0253] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0254] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0255] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0256] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A method for optimizing electricity bidding strategies in a virtual power plant, characterized in that, The virtual power plant includes a battery energy storage system, multiple generator sets, and photovoltaic modules. The method includes: Based on the battery status parameters of the battery energy storage system, the generator parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters, the total cost function of the virtual power plant in each scenario is obtained. Based on the total cost function of the scenario and the scenario probability corresponding to the scenario, the total operating cost function of the virtual power plant is generated. The day-ahead market electricity price and / or the photovoltaic power generation of the photovoltaic modules are different between every two scenarios. A conditional value-at-risk function is constructed based on the value-at-risk parameter, the preset confidence value, the total cost function of the scenario, and the corresponding scenario probability. Obtain the system operation constraints and electricity market bidding constraints of the virtual power plant, and construct an optimization model for the day-ahead market electricity bidding strategy of the virtual power plant based on the system operation constraints, the electricity market bidding constraints, the total operating cost function, and the conditional risk value function; Solve the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy of the virtual power plant. The day-ahead market electricity bidding strategy includes at least optimizing the day-ahead market electricity bidding parameters. Based on all the day-ahead market electricity bidding strategies obtained through iterative solutions, a set of day-ahead market electricity bidding strategies corresponding to the virtual power plant is generated. The step of constructing the day-ahead market electricity bidding strategy optimization model for the virtual power plant based on the system operation constraints, the electricity market bidding constraints, the total operating cost function, and the conditional risk-value function includes: A single-objective optimization model for total operating cost is constructed based on the total operating cost function, the system operating constraints, and the electricity market bidding constraints. A single-objective optimization model for conditional risk value is also constructed based on the conditional risk value function, the risk value parameter, the scenario total cost function, the system operating constraints, and the electricity market bidding constraints. Obtain the risk tolerance parameter, and construct the conditional risk value constraint based on the risk tolerance parameter and the conditional risk value function; Based on the conditional risk value constraint, the system operation constraint, the electricity market bidding constraint, the conditional risk value function, the risk value parameter, the scenario total cost function, and the operating total cost function, an electricity bidding strategy optimization model that balances conditional risk value and operating total cost is constructed. Based on the single-objective optimization model of total operating cost, the single-objective optimization model of conditional value at risk, and the power bidding strategy optimization model that balances conditional value at risk and total operating cost, the day-ahead market power bidding strategy optimization model of the virtual power plant is constructed. The step of solving the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy for the virtual power plant includes: The single-objective optimization model for total operating cost is solved to obtain the optimal total operating cost. Based on the optimal total operating cost, the single-objective optimization model for conditional value of risk is solved to obtain the conditional value of risk under the constraint of the optimal total operating cost and the market electricity bidding strategy on the first day. The conditional risk value single-objective optimization model is solved to obtain the optimal conditional risk value. Based on the optimal conditional risk value, the total operating cost single-objective optimization model is solved to obtain the total operating cost of the virtual power plant under the constraint of the optimal conditional risk value and the market electricity bidding strategy on the second day. The risk tolerance range is determined based on the conditional risk value under the optimal total operating cost constraint and the optimal conditional risk value. Obtain the preset conditional value at risk step size, minimum conditional value at risk step size, conditional value at risk step size update ratio, and cost convergence threshold; Based on the conditional risk value step size, the minimum conditional risk value step size, the conditional risk value step size update ratio, the cost convergence threshold, and the risk tolerance interval, the power bidding strategy optimization model balancing conditional risk value and total operating cost, as well as the conditional risk value single-objective optimization model, are iteratively solved until the number of iterations reaches the preset number of iterations. The day-ahead market power bidding strategy corresponding to the virtual power plant is determined based on the optimized day-ahead market power bidding parameters obtained from each iteration.

2. The method according to claim 1, characterized in that, The battery state parameters are calculated from the scenario charging power parameters, scenario discharging power parameters, charging efficiency, discharging efficiency, and battery capacity corresponding to the battery energy storage system. The generator set parameters include generator set power parameters. The total scenario cost function for the virtual power plant in each scenario is obtained based on the battery state parameters of the battery energy storage system, the generator set parameters of each generator set, the day-ahead market electricity bidding parameters, the real-time market electricity bidding parameters, and the carbon emission parameters. This function includes: Obtain the battery degradation cost coefficient corresponding to the battery energy storage system, the generator set cost coefficient corresponding to each generator set, and the carbon emission factor corresponding to each generator set; The battery loss cost of the battery energy storage system in each scenario is calculated based on the battery degradation cost coefficient, the scenario charging power parameters, the scenario discharging power parameters, the charging efficiency, the discharging efficiency, and the battery capacity. The total power generation cost of the multiple generator sets in each scenario is calculated based on the generator set cost coefficient and the generator set power parameters. Calculate the carbon emission cost for each scenario based on the carbon emission factor and the carbon emission parameter. The total cost function of the virtual power plant in each scenario is calculated based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market electricity bidding parameters, and the real-time market electricity bidding parameters.

3. The method according to claim 2, characterized in that, The day-ahead market electricity bidding parameters include day-ahead market electricity purchase price parameters, day-ahead market electricity purchase price parameters, day-ahead market electricity sales price parameters, and day-ahead market electricity sales parameters. The real-time market electricity bidding parameters include real-time market electricity purchase price parameters, real-time market electricity sales price parameters, real-time market electricity purchase price parameters, and real-time market electricity sales parameters. The calculation based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market electricity bidding parameters, and the real-time market electricity bidding parameters yields the scenario total cost function for the virtual power plant in each scenario, including: The net cost of the virtual power plant's day-ahead market transactions in each scenario is calculated based on the day-ahead market electricity purchase price parameters, the day-ahead market electricity purchase volume parameters, the day-ahead market electricity sales price parameters, and the day-ahead market electricity sales volume parameters. The real-time market purchase price parameter, the real-time market sales price parameter, the real-time market purchase volume parameter, and the real-time market sales volume parameter are used to calculate the real-time market transaction net cost of the virtual power plant in each scenario. The total cost function of the virtual power plant in each scenario is calculated based on the battery loss cost, the total power generation cost of the generator set, the carbon emission cost, the day-ahead market transaction net cost, and the real-time market transaction net cost.

4. The method according to claim 3, characterized in that, Solving the day-ahead market electricity bidding strategy optimization model to obtain the day-ahead market electricity bidding strategy for the virtual power plant includes: Construct a scenario-based electricity purchase bidding curve model based on the day-ahead market electricity purchase price parameters and the day-ahead market electricity purchase volume parameters; A scenario-based electricity bidding curve model is constructed based on the day-ahead market electricity price parameters and the day-ahead market electricity volume parameters. The optimization model for the day-ahead market electricity bidding strategy is solved to obtain the optimized day-ahead market electricity purchase price parameters, optimized day-ahead market electricity purchase volume parameters, optimized day-ahead market electricity sales price parameters, and optimized day-ahead market electricity sales volume parameters for each scenario. Based on the optimized day-ahead market electricity purchase price parameters, the optimized day-ahead market electricity purchase quantity parameters, and the electricity purchase bidding curve model, the day-ahead market electricity purchase bidding strategy of the virtual power plant is obtained; and based on the optimized day-ahead market electricity sales price parameters, the optimized day-ahead market electricity sales quantity parameters, and the electricity sales bidding curve model, the day-ahead market electricity sales bidding strategy of the virtual power plant is obtained. The day-ahead market power bidding strategy of the virtual power plant is determined based on the day-ahead market power purchase bidding strategy and the day-ahead market power sale bidding strategy.

5. The method according to claim 1, characterized in that, The iterative solution process for each iteration number includes the following steps: The current risk tolerance threshold is determined based on the preconditional risk value, and the total operating cost of the power bidding strategy optimization model, which balances the conditional risk value and the total operating cost, is calculated based on the current risk tolerance threshold. The preconditional risk value is the conditional risk value obtained from the previous solution of the single-objective optimization model for the conditional risk value. The conditional risk value single-objective optimization model is solved based on the total operating cost under the current risk tolerance threshold constraint to obtain the conditional risk value under the current risk tolerance threshold and the day-ahead market electricity bidding strategy of the virtual power plant. The risk tolerance parameter is updated based on the preceding total operating cost, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio. The preceding total operating cost is the total operating cost obtained from the previous solution of the power bidding strategy optimization model that balances the conditional risk value and the total operating cost.

6. The method according to claim 5, characterized in that, The risk value step size update ratio includes a step size increase ratio and a step size decrease ratio. Updating the risk tolerance parameter based on the preceding total operating cost, the total operating cost under the current risk tolerance threshold constraint, the cost convergence threshold, the conditional risk value step size, the minimum conditional risk value step size, and the conditional risk value step size update ratio includes: When the difference between the total operating cost under the current risk tolerance threshold constraint and the previous total operating cost exceeds the cost convergence threshold, the conditional risk value step size is updated according to the step size increase ratio to obtain the updated conditional risk value step size. Alternatively, when the difference between the total operating cost under the current risk tolerance threshold constraint and the total operating cost of the preceding operation does not exceed the cost convergence threshold, the conditional risk value step size is updated according to the step size reduction ratio and the minimum conditional risk value step size to obtain the updated conditional risk value step size. The risk tolerance parameter is updated based on the updated conditional value-at-risk step size to obtain the updated risk tolerance parameter.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the power bidding strategy optimization method for the virtual power plant according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the power bidding strategy optimization method for the virtual power plant as described in any one of claims 1 to 6.

Images