A multi-virtual power plant cooperative transaction method, system, device and readable medium

Abstract

The present application relates to a kind of multi-virtual power plant's collaborative transaction method, system, equipment and readable medium, construct the multi-energy coupling model containing gas turbine, energy storage device, biomass power generation and new energy vehicle, generate multi-source data set to quantify carbon emission cost and charge-discharge constraint;Based on F figure Copula distribution function, generate wind and light output joint probability distribution model, with scene closeness degree evaluation screening adverse scenario set by random sampling;Design renewable energy-load correlation coefficient and green energy consumption compensation mechanism, correct virtual power plant scheduling objective function;Establish double-layer dynamic transaction mechanism, iteratively optimize upper layer dynamic electricity price and lower layer multi-source transaction electric quantity;Adopt multi-level robust optimization algorithm to deal with wind, light and electric vehicle uncertainty, generate transaction strategy between multiple virtual power plants under adverse scenario, to solve the problem of low-carbon incentive deficiency, multi-source uncertainty response deficiency, etc., realize the economy, low carbon and robustness of multi-virtual power plant collaborative scheduling transaction.

Description

Technical Field

[0001] This invention relates to the field of virtual power plant collaborative trading technology, specifically to a collaborative trading method, system, equipment, and readable medium for multiple virtual power plants. Background Technology

[0002] With the rapid development of the global economy, electricity has become an important energy source due to its clean and efficient nature. To improve electricity utilization efficiency and integrate regional resources, Virtual Power Plants (VPPs) have become a significant research area. Researching the optimal scheduling of VPPs can improve energy efficiency, achieve system economy, and promote low-carbon operation. VPPs can aggregate and optimize the control of distributed resources such as distributed power sources, energy storage, and controllable loads, and participate as a whole in grid dispatching and electricity market transactions. By aggregating and managing distributed power sources, loads, and energy storage facilities within a park, they can improve the overall efficiency of energy utilization and promote the local consumption of renewable energy. Extensive research has been conducted in recent years on the optimal operation of VPPs.

[0003] Existing research on Virtual Power Plants (VPPs) focuses on optimizing scheduling for better economic returns and value. This involves solving for the optimal economic returns through objective functions and operational constraints. Building upon this, research explores optimal interaction strategies between VPPs and the distribution network to enhance the flexibility of the VPPs themselves. Some studies also consider shared energy storage, enabling flexible energy management through energy storage regulation. However, VPP management limited to a single region cannot effectively integrate resources within that region. Therefore, research on Multi-Regional Virtual Power Plants (MVPPs) demonstrates that multi-regional studies can better improve energy efficiency. With the global development of electric vehicles, optimized scheduling of VPPs with electric vehicle integration has also become a research direction.

[0004] Regarding the uncertainty of renewable energy, existing research addresses the uncertainty of wind turbine and photovoltaic output by generating basic wind and solar scenario sets through Latin hypercube sampling and establishing stochastic optimization models using an improved K-means algorithm for scenario reduction. Building upon this, other studies have addressed the uncertainty of VPP reserves and other types of uncertainty, proposing optimal scheduling schemes under uncertainties such as market prices, renewable energy generation, and reserve deployment and deployment. Furthermore, recent research proposes robust optimization scheduling methods for MVPPs (Very Important Power Plants) to address uncertain parameters related to different stakeholder behaviors. For solution methods, data-driven and risk-constrained approaches are common. However, decision theory based on multiple uncertainties offers better application opportunities.

[0005] Regarding virtual power plant (VPP) electricity market trading mechanisms, existing research has focused on electricity pricing mechanisms. Current studies employ multi-agent uncertainty methods and deep learning approaches combined with electricity markets to achieve optimal pricing strategies. Building upon this, some research considers MVPP aggregation risk, investigating VPP electricity trading across different markets to manage its financial risk. To further explore VPP low-carbon economic dispatch from a market mechanism perspective, existing research has proposed VPP optimal dispatch methods considering carbon emission costs, and further developed flexible carbon emission optimization methods. Simultaneously, research has incorporated carbon capture equipment for low-carbon dispatch. In addition, existing research has explored VPP optimal dispatch related to carbon certificates. Furthermore, research has addressed potential privacy and security issues in traditional centralized dispatch processes, proposing a decentralized VPP market trading mechanism based on blockchain technology, ensuring transaction security while achieving economic efficiency. Other research explores a two-way market trading method, considering electricity and natural gas bidirectional flow dispatch, and developing optimization models for dual-energy markets under different scenarios to maximize VPP profits.

[0006] As described in the aforementioned research, existing technologies for traditional virtual power plant dispatching primarily focus on economic optimization within a single region, lacking a coordinated quantification of cross-regional carbon trading costs and green energy absorption capacity. On one hand, the strong randomness of wind and solar renewable energy leads to output prediction biases, and existing scenario generation methods struggle to accurately characterize the impact of extreme scenarios on trading strategies, resulting in insufficient robustness of dispatching schemes. On the other hand, the green energy absorption incentive mechanism and the spatiotemporal matching of load demand have not yet been effectively established, leading to a disconnect between compensation incentives and low-carbon goals. Furthermore, the dynamic electricity price interaction and traded electricity feedback mechanisms among multiple virtual power plants lack iterative optimization capabilities, making it difficult to balance economic benefits and system stability. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a collaborative trading method, system, equipment and readable medium for multiple virtual power plants to realize the collaborative quantification of carbon metering and green energy consumption rewards.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0009] A collaborative trading method for multiple virtual power plants, characterized by comprising the following steps:

[0010] S1: Construct a low-carbon operation framework for a virtual power plant containing a multi-energy coupling model based on the acquired multi-source historical data, and process the multi-source raw data based on the low-carbon operation framework of the virtual power plant to generate a multi-energy coupling data set.

[0011] S2: Perform joint probabilistic modeling and sampling sorting on the acquired historical power output datasets of wind turbines and photovoltaics, evaluate the historical scenario set of the historical power output datasets of wind turbines and photovoltaics, and generate a typical scenario set containing a set of adverse scenarios and a set of ideal scenarios;

[0012] S3: Calculate the renewable energy-load correlation coefficient and compensation amount for the multi-energy coupled data set and the typical scenario set, and generate the green energy consumption compensation amount;

[0013] S4: Based on the multi-energy coupling data set, the typical scenario set and the green energy consumption compensation amount, a multi-entity trading mechanism is established. The multi-entity trading mechanism is used to iteratively optimize the purchase and sale price of electricity from the upper layer and the multi-source trading current from the lower layer, and generate the final optimized dynamic purchase and sale price and trading current feedback signal.

[0014] S5: Based on a multi-level robust optimization solution algorithm, robust scheduling processing is performed on the typical scenario set, the transaction current feedback signal, and the charging and discharging power distribution set of new energy vehicles in the multi-energy coupling data set to generate cross-regional trading strategies and equipment output plans under adverse scenarios. The cross-regional trading strategies are used to guide the collaborative scheduling of multiple virtual power plants.

[0015] To solve the above-mentioned technical problems, the present invention adopts other technical solutions as follows:

[0016] A collaborative trading system for multiple virtual power plants, the system comprising:

[0017] The virtual power plant framework construction module is used to construct a low-carbon operation framework for a virtual power plant containing a multi-energy coupling model based on the acquired multi-source historical data, and to process the multi-source raw data based on the low-carbon operation framework of the virtual power plant to generate a multi-energy coupling data set.

[0018] The multi-source historical data includes historical operating data of gas turbines, charging and discharging records of energy storage devices, biomass fuel parameters, and charging logs of new energy vehicles. The multi-energy coupled data set includes the output power of gas turbines, the charging and discharging constraints of energy storage devices, the power generation and carbon emission reduction of biomass power generation devices, and the charging and discharging power distribution set of new energy vehicles.

[0019] The adverse-ideal scenario generation module is used to perform joint probabilistic modeling and sampling sorting on the acquired historical power output datasets of wind turbines and photovoltaics, evaluate the historical scenario set of the historical power output datasets of wind turbines and photovoltaics, and generate a typical scenario set containing adverse scenario set and ideal scenario set;

[0020] The energy-load correlation module is used to calculate the renewable energy-load correlation coefficient and compensation amount for the multi-energy coupling data set and the typical scenario set, and generate the green energy consumption compensation amount.

[0021] The trading mechanism construction module is used to process the multi-energy coupling data set, the typical scenario set and the green energy consumption compensation amount to establish a multi-entity trading mechanism, gradually iteratively optimize the purchase and sale price of electricity from the upper layer and the multi-source trading current from the lower layer, and generate the final optimized dynamic purchase and sale price of electricity and trading current feedback signal.

[0022] The robust optimization solution module is used to perform robust scheduling processing on the typical scenario set, the transaction current feedback signal and the charging and discharging power distribution set of new energy vehicles in the multi-energy coupling data set based on the multi-level robust optimization solution algorithm, and generate cross-regional trading strategies and equipment output plans under adverse scenarios. The cross-regional trading strategies are used to guide the collaborative scheduling of multiple virtual power plants.

[0023] To solve the above-mentioned technical problems, the present invention adopts other technical solutions as follows:

[0024] An apparatus includes a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the aforementioned collaborative trading method for multiple virtual power plants.

[0025] To solve the above-mentioned technical problems, the present invention adopts other technical solutions as follows:

[0026] A storage medium storing a computer program that, when executed by a processor, implements the aforementioned collaborative trading method for multiple virtual power plants.

[0027] The beneficial effects of this invention are as follows: This invention provides a collaborative trading method, system, equipment, and readable medium for multiple virtual power plants. By constructing a low-carbon operation framework for virtual power plants containing a multi-energy coupling model and processing multi-source data, it comprehensively integrates multi-source information from gas turbines, energy storage, biomass, and new energy vehicles, generating a multi-energy coupling data set, laying a solid data foundation for subsequent optimization processing. Joint probabilistic modeling, sampling sorting, and scenario evaluation are performed on historical data of wind turbine and photovoltaic output, effectively generating a typical scenario set containing both adverse and ideal scenarios, fully considering the uncertainty of renewable energy output, and improving the robustness of the dispatch scheme. The renewable energy-load correlation coefficient and green energy consumption compensation amount are calculated to generate low-carbon incentive parameters, establishing a green energy consumption reward mechanism that matches the spatiotemporal needs of load demand, enhancing the correlation between compensation incentives and low-carbon goals. Based on the multi-energy coupling data set, typical scenario set, and low-carbon incentive parameters, an MVPP multi-entity trading mechanism is constructed to achieve iterative optimization of dynamic electricity price interaction and traded electricity feedback, balancing economic benefits and system stability. Finally, a multi-level robust optimization algorithm is used for robust scheduling to generate cross-regional trading strategies and equipment output plans under adverse scenarios. This effectively guides the collaborative scheduling of multiple virtual power plants, significantly improves the flexibility, reliability, and environmental friendliness of the power system under the background of vehicle-grid integration and the uncertainty of renewable energy, effectively solves the problems existing in traditional virtual power plant scheduling, and realizes the synergistic quantitative optimization of cross-regional carbon trading costs and green energy absorption capacity. Attached Figure Description

[0028] Figure 1 A flowchart illustrating a multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption rewards, provided as an embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram of the structure of the MVPP multi-party transaction mechanism provided in an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram of the structure of the vehicle-to-grid integration uncertainty risk management model provided in an embodiment of the present invention;

[0031] Figure 4 This is a schematic diagram of the process for generating green energy consumption compensation amount provided in an embodiment of the present invention;

[0032] Figure 5 A schematic diagram of a multi-virtual power plant collaborative trading system that combines carbon metering and green energy consumption rewards is provided for another embodiment of the present invention;

[0033] Label Explanation:

[0034] 600. A multi-virtual power plant collaborative trading system that combines carbon metering with green energy consumption incentives;

[0035] 610. VPP framework construction module; 620. Harsh-ideal scenario generation module; 630. Energy-compliance correlation module; 640. Transaction mechanism construction module; 650. Robust optimization solution module. Detailed Implementation

[0036] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0037] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0038] 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 invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0039] In one embodiment, such as Figure 1 As shown, a multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption incentives is provided. This embodiment illustrates the method applied to a terminal, but it is understood that the method can also be applied to a server, or to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0040] S1: Based on the acquired multi-source historical data, construct a virtual power plant (VPP) low-carbon operation framework containing a multi-energy coupling model, and process the multi-source raw data based on the virtual power plant (VPP) low-carbon operation framework to generate a multi-energy coupling data set.

[0041] Specifically, the multi-source historical data includes historical operating data of gas turbines, charging and discharging records of energy storage devices, biomass fuel parameters, and charging logs of new energy vehicles. The multi-energy coupled data set includes the output power of gas turbines, the charging and discharging constraints of energy storage devices, the power generation and carbon emission reduction of biomass power generation devices, and the power distribution set of charging and discharging of new energy vehicles.

[0042] Specifically, the system collects multi-source historical data generated from various equipment and business processes related to the virtual power plant. This includes historical operating data of gas turbines, such as detailed parameters like output power, fuel consumption, operating time, and number of start-stop cycles under different times and operating conditions; charging and discharging records of energy storage devices, such as charging power, discharging power, charging and discharging time, and state of charge (SOC) change curves for battery energy storage, as well as relevant charging and discharging data for supercapacitor energy storage; biomass fuel parameters, including fuel type, calorific value, supply frequency, and unit energy consumption; and new energy vehicle charging logs, containing information such as charging power, charging time, and charging frequency of new energy vehicles connected to charging piles during different time periods.

[0043] After data collection, the system builds a low-carbon operation framework based on the acquired multi-source historical data. This framework comprehensively reflects the conversion, transmission, and utilization of various energy sources within the virtual power plant. It integrates the operational characteristics of different energy units, such as gas turbines, energy storage devices, biomass power generation equipment, and new energy vehicles. Corresponding mathematical models are established to describe the energy flow and interrelationships between these units. Examples include models for the electrical and thermal power conversion efficiency of gas turbines, the residual power relationship of energy storage devices, the power generation calculation model for biomass power generation equipment, and the interaction model between the charging and discharging power of new energy vehicles and the grid load. This forms a multi-energy coupling model, providing the basic infrastructure support for subsequent data processing and analysis.

[0044] Specifically, the system cleans the source historical data, removing outliers, errors, and duplicates to ensure accuracy and reliability. Then, based on the physical relationships and energy conversion laws of each energy unit in the multi-energy coupling model, the cleaned data is transformed and integrated. For example, the output power of the gas turbine is calculated based on its operating data, and its power output range is determined by combining relevant constraints; the charging and discharging records of energy storage devices are analyzed to clarify their charging and discharging constraints, including the upper limit of charging power, the lower limit of discharging power, and the reasonable range of SOC; the power generation capacity and corresponding carbon emission reduction of biomass power generation equipment are calculated based on biomass fuel parameters, with the carbon emission reduction calculated according to relevant standards and formulas for carbon emission reduction resulting from biomass energy replacing traditional fossil fuel power generation; and the charging logs of new energy vehicles are statistically analyzed to generate a charging and discharging power distribution set, reflecting the distribution of charging power of new energy vehicles in different time periods. Finally, these processed data are integrated to form a multi-energy coupling dataset, providing a data foundation for subsequent analysis and optimization.

[0045] S2: Perform joint probabilistic modeling and sampling sorting on the acquired historical power output datasets of wind turbines and photovoltaics, evaluate the historical scene set of the historical power output datasets of wind turbines and photovoltaics, and generate a typical scene set containing a set of adverse scene sets and a set of ideal scene sets.

[0046] Specifically, the system collects long-term historical power output data from wind farms and photovoltaic power plants. This data includes power generation data from wind turbines under different wind speeds, seasons, and time periods, as well as power generation data from photovoltaic panels under different light intensities, temperatures, and seasonal conditions. This ensures the data covers a sufficient range of operating conditions and meteorological variations to comprehensively reflect the randomness and volatility of wind and solar power output. Simultaneously, the system employs a suitable joint probability distribution model to model the historical power output datasets. Considering the potential correlation between wind and solar power output, a multivariate probability distribution model that reflects this relationship, such as a joint probability distribution model based on the Copula function, is selected. Statistical analysis methods are used to determine the model parameters, ensuring they accurately describe the joint probabilistic characteristics of wind and solar power output. Based on this, Monte Carlo sampling and other methods are used to sample the joint probability distribution model, obtaining a large number of wind and solar power output sample scenarios. These sample scenarios are then sorted according to certain rules, such as by power output magnitude and probability of occurrence, to facilitate subsequent scenario evaluation and selection.

[0047] Finally, the system comprehensively evaluates the historical scenario set formed by the historical datasets of wind turbine and photovoltaic power output. On the one hand, the system combines historical meteorological data and power system operation data to analyze the meteorological conditions such as wind speed, solar intensity, and temperature, as well as the grid load characteristics corresponding to each scenario in the historical scenario set. This determines which scenarios are considered severe or ideal. For example, extreme low temperatures limiting wind turbine output and continuous rainy weather causing a significant drop in photovoltaic output are considered severe scenarios, while stable and suitable wind speeds, sufficient sunlight, and stable grid load demand are considered ideal scenarios. On the other hand, based on a certain evaluation index system, such as the frequency of scenario occurrence and the degree of impact on the operating costs and reliability of the virtual power plant, the system quantitatively evaluates and filters various scenarios in the historical scenario set. Ultimately, representative severe and ideal scenarios are selected from the historical scenario set and combined to form a typical scenario set containing both severe and ideal scenario sets. This typical scenario set can, to a certain extent, cover various typical characteristics of wind turbine and photovoltaic power output, providing basic scenario data support for subsequent virtual power plant optimization scheduling and trading strategy formulation.

[0048] S3: Calculate the renewable energy-load correlation coefficient and compensation amount for multi-energy coupled data sets and typical scenario sets, and generate green energy consumption compensation amount.

[0049] Specifically, the system uses a multi-energy coupling data set and a typical scenario set as input to calculate the renewable energy-load correlation coefficient. By analyzing the temporal and spatial correlation between renewable energy output and load demand, it calculates the correlation coefficient using methods such as the Pearson correlation coefficient to quantify the degree of coupling between the two. Simultaneously, based on green energy consumption incentive policies and market mechanisms, and considering factors such as renewable energy generation, load consumption, and carbon emission reduction benefits, the system calculates the green energy consumption compensation amount. This compensation amount aims to incentivize virtual power plants to increase renewable energy consumption and achieve low-carbon operation goals. After the above calculations, the system generates a green energy consumption compensation amount including the renewable energy-load correlation coefficient, providing crucial data support for the subsequent establishment of a trading mechanism.

[0050] S4: Based on the multi-energy coupling data set, typical scenario set and low-carbon incentive parameters, an MVPP multi-entity trading mechanism is established, and the purchase and sale price of electricity from the upper layer and the multi-source trading current from the lower layer are gradually optimized to generate the final optimized dynamic purchase and sale price and trading current feedback signal.

[0051] Specifically, in this mechanism, multiple stakeholders include distributed energy resources within the virtual power plant, such as gas turbines, energy storage devices, biomass power generation equipment, and new energy vehicles, as well as external power grid and market participants. By constructing a multi-objective optimization model encompassing economic efficiency, low carbon emissions, and system stability, and using the purchase and sale price of electricity and multi-source trading current as optimization variables, an iterative optimization algorithm is employed to gradually optimize the purchase and sale price from the upper layer and the multi-source trading current from the lower layer. During the optimization process, factors such as equipment characteristics in the multi-energy coupled data set, uncertainties in typical scenario clusters, and correlation coefficients and compensation amounts in low-carbon incentive parameters are comprehensively considered to continuously adjust the trading strategy, enabling the virtual power plant to achieve a balance between economic benefits and system stability. Finally, optimized dynamic purchase and sale price and trading current feedback signals are generated, providing a basis for the virtual power plant's real-time trading decisions.

[0052] S5: Based on a multi-level robust optimization solution algorithm, robust scheduling processing is performed on the charging and discharging power distribution set of new energy vehicles in typical scenario sets, trading current feedback signals and multi-energy coupling data sets to generate cross-regional trading strategies and equipment output plans under adverse scenarios. The cross-regional trading strategies are used to guide the collaborative scheduling of multiple virtual power plants.

[0053] Specifically, such as Figure 2As shown, the multi-level robust optimization algorithm can effectively cope with multiple uncertainties in the system, including fluctuations in renewable energy output, changes in load demand, and uncertainties in the charging and discharging behavior of new energy vehicles. During robust scheduling, optimized scheduling is performed for system operation under adverse scenarios, generating cross-regional trading strategies and equipment output plans. The cross-regional trading strategy guides the coordinated scheduling among multiple virtual power plants, achieving optimized allocation of energy resources over a wider range and improving system flexibility and reliability. Simultaneously, the equipment output plan clarifies the output arrangements of each distributed energy resource under different scenarios, ensuring that virtual power plants can stably participate in grid dispatch and market transactions while meeting low-carbon operation goals.

[0054] In summary, this invention provides a multi-virtual power plant collaborative trading method that combines carbon metering and green energy consumption rewards. By constructing a low-carbon operation framework for virtual power plants with a multi-energy coupling model and processing multi-source data, it comprehensively integrates information from gas turbines, energy storage, biomass, and new energy vehicles, generating a multi-energy coupling data set, laying a solid data foundation for subsequent optimization. Joint probabilistic modeling, sampling, sorting, and scenario evaluation are performed on historical wind turbine and photovoltaic output data to effectively generate a typical scenario set containing both adverse and ideal scenarios, fully considering the uncertainty of renewable energy output and improving the robustness of the dispatch scheme. The renewable energy-load correlation coefficient and green energy consumption compensation amount are calculated to generate low-carbon incentive parameters, establishing a green energy consumption reward mechanism that matches load demand in time and space, enhancing the correlation between compensation incentives and low-carbon goals. Based on the multi-energy coupling data set, typical scenario set, and low-carbon incentive parameters, an MVPP multi-entity trading mechanism is constructed to achieve iterative optimization of dynamic electricity price interaction and traded electricity feedback, balancing economic benefits and system stability. Finally, a multi-level robust optimization algorithm is used for robust scheduling to generate cross-regional trading strategies and equipment output plans under adverse scenarios. This effectively guides the collaborative scheduling of multiple virtual power plants, significantly improves the flexibility, reliability, and environmental friendliness of the power system under the background of vehicle-grid integration and the uncertainty of renewable energy, effectively solves the problems existing in traditional virtual power plant scheduling, and realizes the synergistic quantitative optimization of cross-regional carbon trading costs and green energy absorption capacity.

[0055] In one embodiment, the Virtual Power Plant (VPP) low-carbon operation framework includes an electrical power to thermal power conversion efficiency model, a surplus power relationship model, and a power generation calculation model. Based on the VPP low-carbon operation framework, multi-source raw data is processed to generate a multi-energy coupled data set, including:

[0056] S11: Based on the electric power to thermal power conversion efficiency model, calculate the gas turbine historical operating data of the multi-source historical data, constrain its upper and lower limits of operating power and ramp rate, and generate the gas turbine output power.

[0057] Specifically, based on the equipment characteristics and operating principles of gas turbines, an efficiency model for the conversion of electrical power to thermal power is established. This model comprehensively considers factors such as the thermal efficiency of the gas turbine, power generation efficiency, and fuel calorific value. Through experimental data or performance curves provided by the manufacturer, the conversion relationship between the electrical power and thermal power of the gas turbine under different operating conditions is determined. The output power of the gas turbine (GT) can be expressed as follows:

[0058] ;

[0059] in: These represent the electrical power and thermal power of the gas turbine, respectively.

[0060] The fraction represents the conversion efficiency of the gas turbine;

[0061] This represents the gas energy input to the gas turbine at time t.

[0062] The operating power constraints and ramp rate constraints of gas turbines and gas boilers can be described as follows:

[0063] ;

[0064] ;

[0065] in, , Let represent the lower and upper limits of the operating power of the gas turbine at time t, respectively. This represents the actual operating power of the gas turbine at time t. , These represent the lower and upper limits of the gas turbine ramp rate, respectively. This represents the actual operating power of the gas turbine at time t-1.

[0066] By substituting historical operating data of the gas turbine, such as fuel consumption, runtime, inlet air temperature, and pressure, from multi-source historical data into the established conversion efficiency model, the output power of the gas turbine at different times is calculated. The output power calculated by the model can more accurately reflect the power generation capacity of the gas turbine under actual operating conditions and takes into account the impact of changes in operating conditions on the output power.

[0067] Preferably, the system sets a maximum and minimum allowable output power to ensure that the gas turbine will not experience equipment damage or operational instability due to excessively high or low power during operation. Simultaneously, the gas turbine's ramp rate limitation, i.e., the rate of change of output power per unit time, is considered. By analyzing historical operating data, a reasonable ramp rate range is determined and constrained when calculating output power, avoiding problems such as excessive equipment stress or grid frequency fluctuations caused by excessively rapid gas turbine power adjustments. Finally, gas turbine output power data that meets the upper and lower limits of operating power and ramp rate constraints is generated.

[0068] S12: Based on the residual power relationship model, define the charging and discharging status flags of the energy storage device charging and discharging records of multi-source historical data, update the charging and discharging power and constrain the charging and discharging power range and energy storage capacity, and output the charging and discharging constraint relationship of the energy storage device.

[0069] Specifically, for energy storage devices (ES), during operation, charge / discharge (thermal) status indicators are introduced for the energy storage device. Indicates heat storage indicator. The indicator signifies heat release, limiting it from simultaneous charging and releasing of heat; therefore, the following equation holds:

[0070] ;

[0071] ;

[0072] in, This indicates the remaining power of the battery. This represents the power at the end of the previous state. , These represent the charge and discharge coefficients of the battery, respectively. , These represent the charging and discharging power of the battery, respectively. The operating constraints and ramp-up constraints of energy storage devices can be described as follows:

[0073] ;

[0074] ;

[0075] ;

[0076] in, , These represent the lower and upper limits of the battery's charging power, respectively. and These represent the lower and upper limits of the battery's discharge power, respectively. Where: , These represent the lower and upper limits of the ramp rate during battery charging and discharging, respectively. For energy storage devices in operation, their capacity should also be constrained.

[0077] ;

[0078] in: , These represent the upper and lower limits of the energy storage device's capacity, respectively. This indicates the capacity of the energy storage device. After the above processing, the charging and discharging constraints of the energy storage device are output, clarifying the power and capacity limitations of the energy storage device under different operating conditions, and providing detailed characteristics of the energy storage device for subsequent optimized scheduling.

[0079] S13: Based on the power generation calculation model, the biomass fuel parameters from multiple historical data are calculated to determine the biomass power generation and carbon emission reduction, and the blending ratio is constrained to generate the power generation and carbon emission reduction of biomass power generation equipment.

[0080] Specifically, the utilization process of biomass considers the recycling and utilization of natural resources such as straw, and the generation of electricity through reasonable co-firing.

[0081] ;

[0082] ;

[0083] in, Indicates biomass power generation capacity. Indicates power generation efficiency. Indicates fuel conversion efficiency. Net calorific value The mass of biomass fuel is represented by T, and the time interval for power generation is represented by T. This indicates the total calorific value of biomass fuel. Here, represents the moisture content of the biomass fuel, and h represents the ash content. The following operational constraints must also be met for biomass power generation equipment:

[0084] ;

[0085] ;

[0086] in, , Let represent the lower and upper limits of the power generation capacity of the biomass power generation equipment at time t, respectively. This represents the actual power generation of the biomass power generation equipment at time t. , These represent the lower and upper limits of the ramp-up power of the biomass power generation equipment at time t, respectively. Considering the operational requirements of the biomass power generation equipment and the stability of fuel supply, constraints are imposed on the blending ratio. The blending ratio refers to the mixing ratio of biomass fuel with other auxiliary fuels. A reasonable blending ratio can ensure the stable operation and combustion efficiency of the power generation equipment while meeting environmental protection requirements. Based on the equipment's design parameters and actual operating experience, the allowable blending ratio range is determined, and corresponding constraints are imposed when calculating power generation and carbon emission reduction to ensure that the biomass power generation equipment operates under the condition of meeting the blending ratio limits, generating compliant power generation and carbon emission reduction data for the biomass power generation equipment.

[0087] S14: Based on cluster analysis, perform cluster analysis on the charging logs of new energy vehicles and construct the uncertainty set boundary to generate the charging and discharging power distribution set of new energy vehicles.

[0088] Specifically, unlike the uncertain output of renewable energy sources such as wind turbines and solar power, the interaction process of electric vehicles is relatively more complex and is greatly affected by subjective human factors. Therefore, its uncertainty set space is difficult to characterize. This embodiment uses the discrete scenario split-bar optimization method to perform cluster analysis on new energy vehicles, considering the probability distribution of different types of new energy vehicles, as described below:

[0089] ;

[0090] ;

[0091] in, and These represent the charging and discharging amounts of new energy vehicles, respectively. , All represent the number of new energy vehicles in the formula. Define the range of summation. Participated in the construction of the calculation relationship for the total number of new energy vehicles. This indicates the number of types of new energy vehicles. This represents the proportion of different types of electric vehicles; to describe the error between the distribution set of new energy vehicles and the actual distribution, the resulting uncertainty space is described as follows:

[0092] ;

[0093] in, Representing the uncertain set of new energy vehicles, Let represent the uncertainty distribution vector, and represent the proportion of each type of new energy vehicle. This represents the empirical distribution in historical data. and These represent the matrix boundaries of the uncertainty set. After the above processing, a set of charging and discharging power distributions for new energy vehicles is generated, reflecting the charging and discharging power distributions of different types of new energy vehicles at different time periods, providing basic data for subsequent vehicle-to-grid integration analysis and optimized scheduling.

[0094] S15: Integrate the data on gas turbine output power, charge-discharge constraint relationship, power generation, carbon emission reduction and charge-discharge power distribution set to generate a multi-energy coupled data set.

[0095] Specifically, the system integrates data on gas turbine output power, energy storage device charging and discharging constraints, biomass power generation and carbon emission reduction, and the charging and discharging power distribution of new energy vehicles. By integrating data from these different types of energy devices, a comprehensive multi-energy coupling dataset is constructed. This dataset contains key operating parameters and characteristics of various energy devices within the virtual power plant, accurately reflecting the multi-energy coupling relationships and overall operating status within the virtual power plant, providing a complete data foundation for subsequent optimized scheduling and collaborative trading.

[0096] In one embodiment, S2 of the multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption rewards provided by the present invention specifically includes the following steps:

[0097] S21: Based on the F-copula distribution function, perform joint probability modeling on the obtained historical data of wind turbine and photovoltaic power output to generate a joint probability distribution model of wind and solar power output.

[0098] Specifically, due to the uncertainties inherent in the large-scale integration of renewable energy and electric vehicles, operational risks are posed to the energy network. This embodiment constructs a vehicle-to-grid integration uncertainty risk management model, the principle diagram of which is shown below. Figure 3 As shown. Traditional uncertainty models often lack scenario generation for wind-solar correlation models. This system uses a multivariate kernel density function to construct the probability distribution function of wind turbines and photovoltaics under uncertainty conditions, and then performs an inverse transformation to obtain the sampled wind turbine and photovoltaic output for each time period, thereby ultimately generating typical daily curves that consider the randomness of wind and solar power. The data in this paper comes from one year of measured wind turbine and photovoltaic output data from a certain region in Southeast China. Based on this data, we further obtain the uncertain output of wind and solar power, which can further ensure the reliability of the wind-solar uncertainty model.

[0099] Based on the vehicle-to-grid (V2G) integration uncertainty risk management model, the system performs joint probability modeling on the acquired historical data of wind turbine and solar power output to generate a joint probability distribution model of wind and solar power output. Preferably, the system can utilize Sklar's theorem when constructing the joint probability distribution model, which sets random variables... The joint distribution function is Its marginal distribution function is Then there exists a Copula function. , so that:

[0100] ;

[0101] This embodiment selects the F-vine Copula function, which can take into account both positive and negative correlation characteristics. The F-vine Copula is a Copula model based on a vine structure. It constructs a cumulative distribution function for wind and solar power output by decomposing a multidimensional Copula function into a product of a series of binary Copula functions. The kernel density estimation method is used to determine the cumulative distribution function of the wind and solar power output using a random variable with a sample size of n. The sample points are Its probability density function is Its kernel density is estimated as follows:

[0102] ;

[0103] in, Indicates window width. The kernel function is then used to obtain the cumulative distribution function of wind and solar power output through integration. as follows:

[0104] ;

[0105] Then the cumulative distribution of the landscape was transformed into a uniform distribution. The following equation holds:

[0106] ;

[0107] in, It is a monotonically increasing function. This represents the probability that the cumulative distribution function F(P) of wind and light, under the probability measure Q, takes a value less than or equal to ξ. This means that, under the probability measure Q, the random variable P takes values ​​less than or equal to... The probability, Indicates passing through the inverse function After conversion, under the original cumulative distribution of scenery, the value is less than or equal to The probability of the unknown parameters is calculated using the maximum likelihood distribution method. The likelihood function for the random variable sample of wind and light is then determined. () can be described as:

[0108] ;

[0109] in, and These are the unknown parameters in the marginal distribution. , This represents the input sample data. and This represents the cumulative distribution function related to the random variable of wind and light. and Indicates and and The corresponding probability density function, θ, represents the unknown parameter in the Copula function. The system obtains the estimated value of parameter θ by maximizing the likelihood function L(θ). After the above steps, the system constructs a joint probability distribution model of wind turbine and photovoltaic power output covering 24 time periods of a day, laying the foundation for subsequent uncertainty analysis and scenario generation. The unknown parameters can then be obtained through the inverse function. Therefore, the expression for the joint probability distribution model for each time period in this paper can be derived as follows:

[0110] ;

[0111] in, This represents the joint probability distribution function of wind turbine and photovoltaic power output during time period t. Denotes the F-shaped Copula distribution function. and Let these represent the edge distribution functions of the wind turbine and photovoltaic power output during time period t, respectively. This represents the set of joint probability distribution functions for wind turbine and photovoltaic power output for each of the 24 time periods of a day. and Let represent the uniformly distributed variables obtained by transforming the wind turbine and photovoltaic power outputs during time period t through their respective marginal distribution functions.

[0112] S22: Based on the joint probability distribution model, random sampling processing is performed on the historical power output datasets of wind turbines and photovoltaics, and the sampled power output values ​​are calculated by the inverse function sampling method to generate a preliminary scene set.

[0113] Specifically, the system generates a series of uniformly distributed random numbers. and These random numbers correspond to the uniformly distributed variables of wind turbine and photovoltaic power output in time period t. Subsequently, using an inverse function sampling method, the system converts the uniformly distributed random numbers into the actual values ​​of wind turbine and photovoltaic power output. The specific conversion formula is as follows:

[0114] ;

[0115] ;

[0116] The system repeats the above process to generate multiple sample points, thus forming a preliminary scenario set. Each scenario fully includes wind turbine and solar power output data for 24 time periods of a day. Using this method, the system can generate a large number of preliminary scenarios that fully reflect the uncertainties in wind turbine and solar power output, providing a rich and detailed sample foundation for subsequent scenario evaluation and reduction efforts.

[0117] S23: Evaluate the initial scene set, calculate the scene proximity using the approximation ideal solution sorting method, and generate a typical scene set containing both the adverse scene set and the ideal scene set.

[0118] Specifically, the system employs the Top-Approximation-to-Ideal-Solution (TOPSIS) method, calculating the relative proximity of wind and solar power output values ​​based on the Euclidean distance between ideal and non-ideal scenario solutions. A larger value indicates closer proximity to the ideal power output, thus providing assessments for corresponding adverse and ideal scenarios. For the acquired n-dimensional wind and solar power output data, and m evaluation indicators for a single day, a wind and solar power output evaluation matrix is ​​constructed. as follows:

[0119] ;

[0120] Construct it into a standardized decision matrix form ,in as follows:

[0121] ;

[0122] in, It is the m-th element in the n-th column of the wind and solar power output evaluation matrix V. The minimum value in the nth column of the wind and solar power output evaluation matrix V. The maximum value in the nth column of the landscape output evaluation matrix V is then calculated, followed by its scene proximity value. The formula for calculating scene closeness is:

[0123] ;

[0124] ;

[0125] ;

[0126] in, Indicates the degree of resemblance to the scene. and These represent the distances between the output data and the solutions for the ideal and adverse scenarios, respectively. and Let represent the ideal scenario solution and the adverse scenario solution, respectively. This represents the value of the j-th indicator in the i-th scenario.

[0127] In one embodiment, such as Figure 4 As shown, S3 of the multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption rewards provided by the present invention specifically includes the following steps:

[0128] S31: Calculate the renewable energy-load correlation coefficient based on renewable energy output and load demand data in the multi-energy coupling dataset.

[0129] Specifically, the renewable energy-load correlation coefficient is obtained through the following steps:

[0130] S311: Based on renewable energy output and load demand data in a multi-energy coupled dataset, the covariance between renewable energy and load is calculated using a covariance calculation formula, generating uncertainty output of renewable energy and covariance indices of regional load.

[0131] Specifically, the calculation process for the covariance index between renewable energy and regional VPP is as follows:

[0132] ;

[0133] in, Represents the covariance index. This represents the renewable energy absorption capacity of the i-th region during time period t. This represents the load demand in the i-th region during time period t. , These represent the average output and load of renewable energy, respectively. Through this calculation, the system can accurately capture the dynamic relationship between renewable energy output and load demand, providing crucial data support for subsequent analysis and decision-making. In practical applications, the system processes large amounts of data from different regions and time periods to ensure that the covariance index comprehensively reflects the complex interaction between renewable energy and load, thus providing a reliable basis for the system's optimized scheduling and operation management.

[0134] S312: Based on the standard deviation calculation method, the complementarity calculation of renewable energy output and load demand data is performed to generate a standard deviation index containing the standard deviation of renewable energy output and the standard deviation of load demand.

[0135] Specifically, the formula for calculating the standard deviation index is:

[0136] ;

[0137] ;

[0138] in, This represents the standard deviation of the amount of renewable energy absorbed by wind turbines and photovoltaic power during the current period. The standard deviation represents the load demand. By utilizing this index, the system can gain a deeper understanding of the distribution characteristics and fluctuations of renewable energy output and load demand data. This is crucial for assessing system stability and developing appropriate control strategies. Through calculating the standard deviation of renewable energy output and load demand data, the system can better identify abnormal fluctuations in the data, prepare in advance, and ensure the safe and stable operation of the system.

[0139] S313: Based on the correlation coefficient calculation formula, the correlation degree of the covariance index and standard deviation index is calculated to generate the renewable energy-load correlation coefficient.

[0140] The formula for calculating the renewable energy-load correlation coefficient is as follows:

[0141] ;

[0142] in, This represents the renewable energy-load correlation coefficient. The covariance index representing the uncertainty of renewable energy output and regional load. This represents the standard deviation of the amount of renewable energy absorbed by wind turbines and photovoltaic power during the current period. This represents the standard deviation of load demand. The system needs to analyze a large amount of historical data when calculating the renewable energy-load correlation coefficient. This data includes wind turbine and solar power output data, as well as load demand data for various regions. Through in-depth analysis of this data, the system can identify potential relationships between renewable energy output and load demand. For example, at certain times, wind turbine output may be highly correlated with peak industrial loads, while at other times, solar power output may match peak residential electricity consumption. By calculating these correlations, the system provides crucial information for subsequent optimized scheduling.

[0143] S32: Generate the green energy consumption compensation amount based on the renewable energy-load correlation coefficient and the power consumption in typical scenarios.

[0144] Specifically, based on the renewable energy-load correlation coefficient, it is used as a measure of wind and solar power consumption in the regional VPP (Vehicle Power Plant) to provide certain compensation for the consumption of renewable energy. The formula for calculating the green energy consumption compensation amount is as follows:

[0145] ;

[0146] in, This indicates the amount of compensation for green energy consumption. Indicates the compensation standard parameters, Let represent the renewable energy consumption capacity of the i-th region during time period t, where T represents the time period. The system quantifies the economic value of renewable energy consumption by integrating the renewable energy-load correlation coefficient with the consumption capacity. The calculation of the compensation amount fully considers the uncertainty of renewable energy output and the volatility of load demand, aiming to incentivize virtual power plants to prioritize renewable energy consumption while meeting load demand, thereby achieving the goal of low-carbon operation. Specifically, within each scheduling cycle, the system dynamically calculates the compensation amount based on the actual renewable energy consumption and correlation coefficient, and incorporates it into the economic benefit model of the virtual power plant. This not only improves the utilization rate of renewable energy but also reduces carbon emissions, achieving a dual optimization of economic efficiency and low carbon emissions.

[0147] In calculating the compensation amount for green energy consumption, the system fully considers the differences in different regions and time periods. Compensation standard parameters The system can be adjusted according to policy requirements or market conditions to adapt to different incentive needs. In this way, the system can flexibly guide the operation strategy of virtual power plants, enabling them to actively absorb renewable energy and reduce dependence on traditional fossil fuels driven by economic interests. Ultimately, through this data-driven optimization method, the system achieves efficient collaborative scheduling of multiple virtual power plant systems, improving the flexibility, reliability, and environmental friendliness of the entire energy system.

[0148] In one embodiment, S4 of the multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption rewards provided by the present invention specifically includes the following steps:

[0149] S41: Based on the dynamic pricing model, the grid electricity price and supply-demand gap in the multi-energy coupled dataset are processed. The initial dynamic purchase and sale price is generated by constraining the upper and lower limits of the electricity price and the supply-demand balance conditions through the upper-level aggregator.

[0150] Specifically, the system processes grid electricity prices and supply-demand gaps in a multi-energy coupled dataset based on a dynamic pricing model. During this process, the system generates initial dynamic purchase and sale prices by constraining upper and lower price limits and supply-demand balance conditions through upper-level aggregators. The aggregator operator publishes prices for each time period. The main focus is on the interaction between the electricity market and VPPs, and the objective function can be described as follows:

[0151] ;

[0152] in, This is the initial dynamic purchase and sale price of electricity. and These represent the current market electricity price for selling electricity and the current electricity price for purchasing electricity. and These represent the electricity selling price and the electricity purchase price determined by the aggregator, respectively. and It refers to the electrical energy that aggregators sell and buy on the electricity market. and This refers to the electricity purchased and sold by the i-th entity in the interaction between the aggregator and the app. The supply-demand balance constraints and electricity price constraints are as follows:

[0153] ;

[0154] ;

[0155] in, This represents the total amount of electricity that the aggregator purchases from or sells to the electricity market at time t.

[0156] S42: Based on the optimization scheduling algorithm, the typical scenario set, green energy consumption compensation amount and initial dynamic purchase and sale electricity price are processed. The output of gas turbine, energy storage charging and discharging plan, biomass power generation and cross-regional transaction electricity are optimized through the lower-level virtual power plant to generate multi-source transaction current feedback signal.

[0157] Specifically, after generating the initial dynamic electricity purchase and sale price, the system further processes the typical scenario set, green energy consumption compensation amount, and initial dynamic electricity purchase and sale price based on an optimized scheduling algorithm. By optimizing gas turbine output, energy storage charging and discharging plans, biomass power generation, and cross-regional trading volume through lower-level virtual power plants, a multi-source trading current feedback signal is generated. Its objective function can be described as follows:

[0158] ;

[0159]

[0160] in, These represent the interaction costs with the aggregator, respectively. , , This represents the cost of gas turbines, energy storage, and flexible interruption loads. This indicates the cost of interaction between VPP and new energy vehicles. This represents the cost of carbon emissions. During the system optimization process, the operational constraints of gas turbines, energy storage devices, biomass power generation equipment, and new energy vehicles must be met, along with supply and demand balance conditions.

[0161] ;

[0162] ;

[0163] in, Let be the power of the i-th entity interacting with the electricity market at time t. Let be the output power of the i-th gas turbine unit at time t. Let be the power of the i-th energy storage device at time t. Let be the charging and discharging power of the i-th new energy vehicle at time t. Let be the output power of the i-th biomass power generation device at time t. Let be the power consumption of the i-th load at time t, representing the power consumption of the load at that time.

[0164] By optimizing the scheduling algorithm, the system generates multi-source trading current feedback signals, including gas turbine output, energy storage charging and discharging plans, biomass power generation, and cross-regional trading volume. These feedback signals reflect the optimal operating strategies of the virtual power plant under different operating conditions.

[0165] S43: Based on the iterative correction mechanism, the multi-source trading current feedback signal and supply-demand gap are processed. By dynamically adjusting the matching relationship between the purchase and sale price and the trading volume, the final optimized dynamic purchase and sale price and the converged trading current feedback signal are generated.

[0166] Specifically, during the iterative correction process, the system repeatedly adjusts the purchase and sale price of electricity and the trading volume until the convergence condition is met. Specifically, the system determines convergence by comparing the differences between the purchase and sale price and the trading volume in the previous and current rounds. If the difference is less than a set threshold, convergence is considered achieved; otherwise, iterative adjustments continue. Through this iterative correction mechanism, the system can dynamically adjust the purchase and sale price of electricity and the trading volume, gradually optimizing the operating strategies of aggregators and virtual power plants while meeting supply and demand balance and operational constraints, ultimately achieving the economical and efficient operation and low-carbon goals of the multi-virtual power plant system.

[0167] In one embodiment, S5 of the multi-virtual power plant collaborative trading method combining carbon metering and green energy consumption incentives provided by the present invention specifically includes the following steps:

[0168] S51: Based on a multi-level robust optimization algorithm, typical scenario sets, transaction current feedback signals, and charging and discharging power distribution sets of new energy vehicles are processed. The uncertainty problem of VPP renewable energy vehicle network is divided into main problem and sub-problem, and the optimal scheduling strategy for known scenarios and the search boundary for severe scenarios are solved respectively.

[0169] Specifically, regarding the uncertainty of wind and solar power, this embodiment employs multi-level robust optimization by integrating historical data of wind and solar power uncertainty variables to obtain different sample distributions. Fuzzy sets are constructed for different distribution parameters of the samples, and a scenario-based approach is used to reduce wind and solar power output under different probabilities. The uncertainty of electric vehicles is incorporated using a polyhedral uncertainty set. The compact form of the multi-level robust optimization algorithm presented in this paper is as follows:

[0170] ;

[0171] ;

[0172] in, , and These represent the electricity selling price and the electricity purchase price determined by the aggregator, respectively. This is the initial dynamic purchase and sale price of electricity. This represents the power allocation for the i-th device. This represents the parameter space determined based on the uncertainties inherent in wind and solar energy, along with relevant physical laws and empirical data. The feasible region is defined based on factors such as equipment operation constraints, grid constraints, and safety regulations of the virtual power plant. The aforementioned problem can be divided into multi-level robust optimization problems. Specifically, for the uncertainty problem of renewable energy vehicle-grid in each virtual power plant proposed in this paper, it can be divided into a main problem and sub-problems. The main problem obtains its optimal solution when the source load output is known, providing the lower boundary for the model and passing the result to the sub-problems. The subproblem provides an upper bound for the model by finding adverse scenarios in the uncertainty set and passing the corresponding scenario output to the main function.

[0173] The system decomposes the uncertainty problem of the VPP (Vehicle-Powered Plant) renewable energy network, with the main problem focusing on solving the optimal scheduling strategy for known scenarios. In this process, the system comprehensively considers multiple factors such as renewable energy output, energy storage device status, and the operating characteristics of traditional generator sets, striving to minimize operating costs and maximize economic efficiency while satisfying grid operation constraints. The system transforms the optimization of the scheduling strategy into a complex multi-variable, multi-constraint optimization problem by establishing a precise mathematical model and applying advanced optimization algorithms for solution. Simultaneously, the sub-problem focuses on determining the search boundary for adverse scenarios. The system simulates various extreme situations and analyzes their impact on the operation of the virtual power plant to determine the possible adverse scenario boundaries. This boundary not only covers extreme fluctuations in renewable energy output but also includes scenarios that may trigger system operation risks, such as sudden changes in load demand and large-scale simultaneous charging of new energy vehicles, providing comprehensive and rigorous boundary conditions for subsequent optimized scheduling. This process fully considers factors such as the volatility of renewable energy output, the uncertainty of load demand, and the randomness of new energy vehicle charging and discharging behavior, aiming to build a robust defense for the virtual power plant against various uncertain shocks.

[0174] S52: Based on the optimal scheduling strategy, the subproblems of the uncertainty problem are processed, and the dual transformation method is used to transform them into a single-layer optimization model. Based on the search boundary of the harsh scenario, the transaction constraints under the extreme scenario are generated by the convex relaxation technique.

[0175] Specifically, the system employs a dual transformation method to convert complex subproblems into easily solvable single-layer optimization models. This method simplifies the problem structure, enabling efficient solutions to problems that were previously difficult to solve directly. Simultaneously, based on the previously determined adverse scenario search boundaries, the system uses convex relaxation techniques to generate trading constraints under extreme scenarios. The application of convex relaxation ensures the mathematical solvability and stability of the generated constraints, guaranteeing that the system can still meet the basic requirements of power grid operation, such as power balance and voltage stability, even under extremely adverse scenarios. These trading constraints not only involve power trading between the virtual power plant and the external power grid but also the coordination and allocation of different energy resources within the virtual power plant, providing clear boundary conditions for subsequent optimized scheduling.

[0176] S53: Based on the optimal scheduling strategy and transaction constraints, an alternating iterative solution is performed. By updating the parameter boundaries of the main problem and sub-problems of the uncertainty problem, a cross-regional transaction strategy and equipment output plan under adverse scenarios are generated. The cross-regional transaction strategy is used to guide the collaborative scheduling of multiple virtual power plants.

[0177] Specifically, such as Figure 3 As shown, the system employs an alternating iterative solution method to further optimize the main problem and sub-problems of the uncertain problem. During each iteration, the system continuously updates the parameter boundaries of the main problem and sub-problems, enabling the main problem to better reflect the uncertainties in actual operation, while the sub-problems can more accurately capture the system characteristics under adverse scenarios. In this way, the system gradually approaches the optimal solution, ultimately generating a cross-regional trading strategy and equipment output plan suitable for adverse scenarios. The cross-regional trading strategy mainly guides the coordinated scheduling between multiple virtual power plants. By rationally arranging power transactions between virtual power plants in different regions, it achieves optimal resource allocation and improves the overall system's operating efficiency and economy. The equipment output plan, targeting various power generation and energy storage devices within the virtual power plant, clarifies the output arrangements and adjustment requirements at different time periods, ensuring that the virtual power plant can maximize the absorption of renewable energy and maintain system stability while meeting grid operation constraints.

[0178] In summary, this method, combining carbon metering and green energy consumption rewards, establishes a multi-virtual power plant collaborative trading model incorporating renewable energy sources such as wind and solar power, and also includes a biomass power generation virtual power plant model. This is achieved through a carbon trading mechanism to ensure low-carbon operation. To address the uncertainties of wind turbines and photovoltaics, an improved kernel distribution function is used for risk management, and the TOPSIS method is employed to assess energy output under normal and adverse weather conditions to address various practical operating scenarios. Furthermore, considering vehicle-to-grid integration, a model incorporating the dynamic interaction of new energy vehicles is established, paying particular attention to the uncertainties in the charging and discharging process between electric vehicles and VPPs, and constructing a corresponding interaction model. Finally, a multi-VPP trading mechanism considering dynamic electricity pricing is established. Besides ensuring economic viability, it compensates for renewable energy by establishing a correlation coefficient between renewable energy and energy consumption behavior, thereby guaranteeing the low-carbon nature of the proposed algorithm. Robust optimal scheduling of multi-VPPs is achieved under the background of considering the uncertainties of vehicle-to-grid integration and renewable energy. Simulations verify its effectiveness in improving the flexibility, reliability, and environmental friendliness of the power system. Future research will continue to focus on the deep correlation between renewable energy and load, as well as its characteristics across long and short time scales.

[0179] Preferably, such as Figure 5 As shown, this invention provides a multi-virtual power plant collaborative trading system 600 that combines carbon metering and green energy consumption rewards. This system is configured with the following modules:

[0180] VPP framework construction module 610 is used to construct a virtual power plant VPP low-carbon operation framework containing a multi-energy coupling model based on the acquired multi-source historical data, and to process the multi-source raw data based on the virtual power plant VPP low-carbon operation framework to generate a multi-energy coupling data set.

[0181] Among them, the multi-source historical data includes historical operating data of gas turbines, charging and discharging records of energy storage devices, biomass fuel parameters and charging logs of new energy vehicles, and the multi-energy coupled data set includes the output power of gas turbines, the charging and discharging constraints of energy storage devices, the power generation and carbon emission reduction of biomass power generation devices and the charging and discharging power distribution set of new energy vehicles;

[0182] The adverse-ideal scenario generation module 620 is used to perform joint probabilistic modeling and sampling sorting on the acquired historical power output datasets of wind turbines and photovoltaics, evaluate the historical scenario set of the historical power output datasets of wind turbines and photovoltaics, and generate a typical scenario set containing adverse scenario set and ideal scenario set.

[0183] The energy-load correlation module 630 is used to calculate the renewable energy-load correlation coefficient and compensation amount for multi-energy coupled data sets and typical scenario sets, and generate green energy consumption compensation amount.

[0184] The trading mechanism construction module 640 is used to process multi-energy coupled data sets, typical scenario sets and green energy consumption compensation amounts to establish an MVPP multi-entity trading mechanism, gradually iteratively optimize the purchase and sale electricity price from the upper layer and the multi-source trading current from the lower layer, and generate the final optimized dynamic purchase and sale electricity price and trading current feedback signal.

[0185] The robust optimization solution module 650 is used to perform robust scheduling processing on the charging and discharging power distribution set of new energy vehicles in typical scenario sets, trading current feedback signals and multi-energy coupling data sets based on a multi-level robust optimization solution algorithm. It generates cross-regional trading strategies and equipment output plans under adverse scenarios. The cross-regional trading strategies are used to guide the collaborative scheduling of multiple virtual power plants.

[0186] In summary, this invention provides a multi-virtual power plant collaborative trading system that combines carbon metering and green energy consumption rewards. By constructing a low-carbon operation framework for virtual power plants with a multi-energy coupling model and processing multi-source data, it comprehensively integrates information from gas turbines, energy storage, biomass, and new energy vehicles, generating a multi-energy coupling data set to lay a solid data foundation for subsequent optimization. Joint probabilistic modeling, sampling, sorting, and scenario assessment are performed on historical wind turbine and photovoltaic output data to effectively generate a typical scenario set containing both adverse and ideal scenarios, fully considering the uncertainty of renewable energy output and improving the robustness of the dispatch scheme. The renewable energy-load correlation coefficient and green energy consumption compensation amount are calculated to generate low-carbon incentive parameters, establishing a green energy consumption reward mechanism that matches load demand in time and space, enhancing the correlation between compensation incentives and low-carbon goals. Based on the multi-energy coupling data set, typical scenario set, and low-carbon incentive parameters, an MVPP multi-entity trading mechanism is constructed to achieve iterative optimization of dynamic electricity price interaction and traded electricity feedback, balancing economic benefits and system stability. Finally, a multi-level robust optimization algorithm is used for robust scheduling to generate cross-regional trading strategies and equipment output plans under adverse scenarios. This effectively guides the collaborative scheduling of multiple virtual power plants, significantly improves the flexibility, reliability, and environmental friendliness of the power system under the background of vehicle-grid integration and the uncertainty of renewable energy, effectively solves the problems existing in traditional virtual power plant scheduling, and realizes the synergistic quantitative optimization of cross-regional carbon trading costs and green energy absorption capacity.

[0187] Preferably, the VPP framework building module 610 is configured with the following units:

[0188] The gas turbine output power generation unit is used to calculate the historical operating data of the gas turbine based on the electric power and thermal power conversion efficiency model, constrain its upper and lower limits of operating power and ramp rate, and generate the gas turbine output power.

[0189] The energy storage device charge and discharge constraint relationship generation unit is used to define charge and discharge status flags for the energy storage device charge and discharge records of multi-source historical data based on the residual power relationship model, update the charge and discharge power and constrain the charge and discharge power range and energy storage capacity, and output the charge and discharge constraint relationship of the energy storage device.

[0190] The biomass power generation equipment data generation unit is used to calculate the biomass fuel parameters based on the power generation calculation model from multiple sources of historical data, optimize the biomass power generation and carbon emission reduction, and constrain the blending ratio to generate the power generation and carbon emission reduction of the biomass power generation equipment.

[0191] The new energy vehicle charging and discharging power distribution set generation unit is used to perform cluster analysis on the charging logs of new energy vehicles based on cluster analysis method and construct the uncertainty set boundary to generate the charging and discharging power distribution set of new energy vehicles.

[0192] The multi-energy coupled data set generation unit is used to integrate data on gas turbine output power, charge-discharge constraint relationship, power generation, carbon emission reduction and charge-discharge power distribution set to generate a multi-energy coupled data set.

[0193] Preferably, the adverse-ideal scene generation module 620 is configured with the following units:

[0194] The joint probability distribution model generation unit is used to perform joint probability modeling on the acquired historical data of wind turbine and photovoltaic power output based on the F-copula distribution function, and generate a joint probability distribution model of wind and solar power output.

[0195] The preliminary scenario set generation unit is used to randomly sample the historical power output datasets of wind turbines and photovoltaics based on the joint probability distribution model, calculate the sampled power output values ​​through the inverse function sampling method, and generate a preliminary scenario set.

[0196] The typical scene set generation unit is used to evaluate and process the preliminary scene set, calculate the scene closeness by using the approximation ideal solution sorting method, and generate a typical scene set containing the adverse scene set and the ideal scene set.

[0197] Preferably, the energy-load association module 630 is configured with the following units:

[0198] The correlation coefficient calculation unit is used to calculate the renewable energy-load correlation coefficient based on renewable energy output and load demand data in the multi-energy coupling dataset;

[0199] The compensation amount generation unit is used to generate green energy consumption compensation amount based on the renewable energy-load correlation coefficient and the consumption power of typical scenarios.

[0200] Preferably, the correlation coefficient calculation unit includes a covariance index generation subunit, a standard deviation index generation subunit, and a correlation coefficient generation subunit. The covariance index generation subunit is used to calculate the covariance between renewable energy and load based on renewable energy output and load demand data in a multi-energy coupled dataset, generating covariance indices for uncertain renewable energy output and regional load. The standard deviation index generation subunit is used to calculate the complementarity of renewable energy output and load demand data based on the standard deviation calculation method, generating a standard deviation index containing the standard deviation of renewable energy output and load demand. The correlation coefficient generation subunit is used to calculate the correlation degree between the covariance index and the standard deviation index based on the correlation coefficient calculation formula, generating a renewable energy-load correlation coefficient.

[0201] Preferably, the transaction mechanism construction module 640 is configured with the following units:

[0202] The initial dynamic purchase and sale price generation unit is used to process the grid electricity price and supply and demand gap in the multi-energy coupled data set based on the dynamic pricing model. It generates the initial dynamic purchase and sale price by constraining the upper and lower limits of the electricity price and the supply and demand balance conditions through the upper-level aggregator.

[0203] The multi-source trading current feedback signal generation unit is used to process typical scenario sets, green energy consumption compensation amount and initial dynamic purchase and sale electricity price based on the optimization scheduling algorithm. It generates multi-source trading current feedback signals by optimizing gas turbine output, energy storage charging and discharging plan, biomass power generation and cross-regional trading electricity through the lower-level virtual power plant.

[0204] The final optimization result generation unit is used to process the multi-source trading current feedback signal and supply-demand gap based on the iterative correction mechanism. By dynamically adjusting the matching relationship between the purchase and sale price and the trading volume, it generates the final optimized dynamic purchase and sale price and the converged trading current feedback signal.

[0205] Preferably, the robust optimization solution module 650 is configured with the following units:

[0206] The problem decomposition and solution unit is used to process typical scenario sets, transaction current feedback signals and charging and discharging power distribution sets of new energy vehicles based on multi-level robust optimization algorithms. By dividing the uncertainty problem of VPP renewable energy vehicle network into main problems and sub-problems, it solves the optimal scheduling strategy for known scenarios and the search boundary for adverse scenarios respectively.

[0207] The constraint generation unit is used to process the subproblems of the uncertainty problem based on the optimal scheduling strategy, transform them into a single-layer optimization model through the dual transformation method, and generate transaction constraints under extreme scenarios based on the search boundary of the harsh scenario through the convex relaxation technique.

[0208] The strategy and plan generation unit is used to perform alternating iterative solution processing based on the optimal scheduling strategy and transaction constraints. By updating the parameter boundaries of the main problem and sub-problems of the uncertainty problem, it generates cross-regional transaction strategies and equipment output plans under adverse scenarios. The cross-regional transaction strategy is used to guide the collaborative scheduling of multiple virtual power plants.

[0209] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described collaborative trading method for multiple virtual power plants.

[0210] In one embodiment, this application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described collaborative trading method for multiple virtual power plants.

[0211] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0212] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and 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 modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0213] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for coordinated transaction of multiple virtual power plants, characterized in that, The method comprises the following steps: S1: constructing a low-carbon operation framework of a virtual power plant containing a multi-energy coupling model based on obtained multi-source historical data, and processing multi-source original data based on the low-carbon operation framework of the virtual power plant to generate a multi-energy coupling data set; S2: performing joint probability modeling processing and sampling sorting on the obtained wind turbine and photovoltaic output historical data set, performing evaluation processing on a historical scenario set of the wind turbine and photovoltaic output historical data set, and generating a typical scenario set containing a severe scenario set and an ideal scenario set; S3: performing renewable energy-load correlation coefficient calculation and compensation amount calculation on the multi-energy coupling data set and the typical scenario set to generate a green energy consumption compensation amount; S4: establishing a multi-agent transaction mechanism based on the multi-energy coupling data set, the typical scenario set and the green energy consumption compensation amount, iteratively optimizing a purchase and sale electricity price from an upper layer and a multi-source transaction current from a lower layer by using the multi-agent transaction mechanism, and generating a final optimized dynamic purchase and sale electricity price and a transaction current feedback signal; S5: performing robust scheduling processing on a new energy vehicle charging and discharging power distribution set in the typical scenario set, the transaction current feedback signal and the multi-energy coupling data set by using a multi-level robust optimization solving algorithm to generate a cross-regional transaction strategy and a device output plan in a severe scenario, and using the cross-regional transaction strategy to guide multi-virtual power plant collaborative scheduling.

2. The method of claim 1, wherein, The low-carbon operation framework of the virtual power plant comprises an electric power and thermal power conversion efficiency model, a residual power relationship model and a power generation power calculation model; The processing of the multi-source original data based on the low-carbon operation framework of the virtual power plant to generate the multi-energy coupling data set comprises: S11: calculating historical operation data of a gas turbine in the multi-source historical data based on the electric power and thermal power conversion efficiency model, generating gas turbine output power by taking the upper and lower limits of the operation power of the gas turbine and the climbing rate as constraints; S12: defining a charging and discharging state flag based on the residual power relationship model for charging and discharging records of energy storage equipment in the multi-source historical data, updating the charging and discharging power and taking the charging and discharging power range and the energy storage capacity as constraints to output the charging and discharging constraint relationship of the energy storage equipment; S13: calculating biomass fuel parameters in the multi-source historical data based on the power generation power calculation model, updating biomass power generation power and carbon emission reduction amount, and generating the power generation power and carbon emission reduction amount of a biomass power generation device by taking the blending ratio as a constraint; S14: performing cluster analysis on new energy vehicle charging logs by using a cluster analysis method and constructing an uncertainty set boundary to generate a charging and discharging power distribution set of the new energy vehicle; S15: integrating the gas turbine output power, the charging and discharging constraint relationship, the power generation power and carbon emission reduction amount and the charging and discharging power distribution set to generate the multi-energy coupling data set.

3. The method of claim 1, wherein, The S2 comprises: S21: performing joint probability modeling processing on the obtained wind turbine and photovoltaic output historical data set based on an F-Feng Copula distribution function to generate a wind-solar output joint probability distribution model, and an expression of the joint probability distribution model is: ； wherein, denotes t the joint probability distribution function of the wind turbine and photovoltaic power outputs for a time period, denotes the F copula distribution function, and denote the marginal distribution functions of the wind turbine and photovoltaic power outputs, respectively, t denotes the joint probability distribution function of the wind turbine and photovoltaic power outputs for a time period, denotes the set of joint probability distribution functions of the wind turbine and photovoltaic power outputs for each of the 24 time periods of a day, and denote the uniform distribution variables obtained by transforming the wind turbine and photovoltaic power outputs for a time period through their respective marginal distribution functions, t denote the uniform distribution variables obtained by transforming the wind turbine and photovoltaic power outputs for a time period through their respective marginal distribution functions, S22: Random sampling processing is performed on the wind turbine and photovoltaic output historical data set based on the joint probability distribution model, sampled output values are calculated by inverse function sampling method, and a preliminary scenario set is generated; S23: The preliminary scenario set is evaluated, the scenario closeness is calculated by the approximation ideal solution sorting method, and a typical scenario set containing a severe scenario set and an ideal scenario set is generated, and the calculation formula of the scenario closeness is: ； ； ； wherein, denotes the closeness of the scenario, and denote the distance between the output data and the ideal scenario solution and the bad scenario solution, respectively, and denote the ideal scenario solution and the bad scenario solution, respectively, denotes the value of the i th indicator in the j th scenario.

4. The method of claim 1, wherein, The S3 comprises: S31: Based on the renewable energy output and load demand data in the multi-energy coupling data set, a renewable energy-load correlation coefficient is calculated, and the calculation formula of the renewable energy-load correlation coefficient is: ； wherein, represents the renewable energy-load correlation coefficient, represents the covariance index of the renewable output uncertainty and the regional load, represents the standard deviation of the wind and photovoltaic renewable energy consumption in the current period, represents the standard deviation of the load demand; S32: Based on the renewable energy-load correlation coefficient and the consumption power in the typical scenario set, a green energy consumption compensation amount is generated, and the calculation formula of the green energy consumption compensation amount is: ； wherein, represents the green energy consumption compensation amount, represents the compensation standard parameter, represents the renewable energy consumption power of the i-th region at the t period, T represents the time period.

5. The method of claim 4, wherein, The S31 comprises: S311: Based on the renewable energy output and load demand data in the multi-energy coupling data set, the covariance of renewable energy and load is calculated by the covariance calculation formula, and the covariance index of the uncertainty output of renewable energy and the regional load is generated, and the calculation formula of the covariance index is: ； wherein, denotes a covariance indicator, denotes the mean of the renewable energy output in the t time period in the i region, denotes the mean of the load demand in the t time period in the i region, , denote the mean of the renewable energy output and the load, respectively; S312: The complementarity of the renewable energy output and load demand data is calculated based on the standard deviation calculation method, and a standard deviation index containing the standard deviation of renewable energy output and the standard deviation of load demand is generated, and the calculation formula of the standard deviation index is: ； ； wherein , denotes the standard deviation of the load demand; S313: The correlation degree of the covariance index and the standard deviation index is calculated based on the correlation coefficient calculation formula, and a renewable energy-load correlation coefficient is generated.

6. The method of claim 1, wherein, The S4 comprises: S41: Based on the dynamic pricing model, the grid price and supply-demand gap in the multi-energy coupling data set are processed, the upper limit and lower limit of the price of the upper aggregator and the supply-demand balance condition are constrained, and an initial dynamic power purchase and sale price is generated; S42: Based on the optimal scheduling algorithm, the typical scenario set, the green energy consumption compensation amount and the initial dynamic power purchase and sale price are processed, the gas turbine output, the energy storage charging and discharging plan, the biomass power and the cross-regional transaction power of the lower virtual power plant are optimized, and a multi-source transaction current feedback signal is generated; S43: Based on the iterative correction mechanism, the multi-source transaction current feedback signal and the supply-demand gap are processed, the matching relationship between the power purchase and sale price and the transaction power is dynamically adjusted, and the final optimized dynamic power purchase and sale price and the converged transaction current feedback signal are generated.

7. The method of claim 1-6, wherein, The S5 comprises: S51: Based on the multi-level robust optimization algorithm, the typical scenario set, the transaction current feedback signal and the charging and discharging power distribution set of the new energy vehicle are processed, and the uncertainty problem of the VPP renewable energy vehicle grid is divided into a main problem and a sub-problem for solving the optimal scheduling strategy of the known scenario and searching the boundary of the severe scenario; S52: Based on the optimal scheduling strategy, the sub-problem of the uncertainty problem is processed, the dual transformation method is used to convert it into a single-layer optimization model, and based on the severe scenario search boundary, the transaction constraint condition under the limit scenario is generated by the convex relaxation technology. S53: alternately iteratively solving based on the optimal scheduling strategy and the transaction constraint condition, generating a cross-regional transaction strategy and a device output plan under adverse scenarios by updating the parameter boundary of the main problem and the sub-problem of the uncertainty problem, the cross-regional transaction strategy being used to guide the collaborative scheduling of the multiple virtual power plants.

8. A coordinated transaction system of multiple virtual power plants, characterized by, The system comprises: A framework construction module of a virtual power plant, configured to construct a low-carbon operation framework of the virtual power plant containing a multi-energy coupling model based on acquired multi-source historical data, and process multi-source original data based on the low-carbon operation framework of the virtual power plant to generate a multi-energy coupling data set; The multi-source historical data comprises gas turbine historical operation data, energy storage device charging and discharging records, biomass fuel parameters and new energy vehicle charging logs, and the multi-energy coupling data set comprises gas turbine output power, charging and discharging constraint relationship of the energy storage device, power generation power and carbon emission reduction of the biomass power generation device and charging and discharging power distribution set of the new energy vehicle; An adverse-ideal scenario generation module, configured to perform joint probability modeling processing and sampling sorting on acquired wind turbine and photovoltaic output historical data set, perform evaluation processing on a historical scenario set of the wind turbine and photovoltaic output historical data set, and generate a typical scenario set containing an adverse scenario set and an ideal scenario set; An energy-load association module, configured to perform renewable energy and load association coefficient calculation and compensation amount calculation on the multi-energy coupling data set and the typical scenario set to generate a green energy consumption compensation amount; A transaction mechanism construction module, configured to perform processing based on the multi-energy coupling data set, the typical scenario set and the green energy consumption compensation amount, establish a multi-agent transaction mechanism, gradually iteratively optimize a purchase and sale electricity price from an upper layer and a multi-source transaction current from a lower layer, and generate a final optimized dynamic purchase and sale electricity price and transaction current feedback signal; A robust optimization solving module, configured to perform robust scheduling processing on the typical scenario set, the transaction current feedback signal and the charging and discharging power distribution set of the new energy vehicle in the multi-energy coupling data set based on a multi-level robust optimization solving algorithm, and generate a cross-regional transaction strategy and a device output plan under adverse scenarios, the cross-regional transaction strategy being used to guide the collaborative scheduling of the multiple virtual power plants.

9. A device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to realize the steps in the collaborative transaction method of the multiple virtual power plants according to any one of claims 1 to 7.

10. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to realize the steps in the collaborative transaction method of the multiple virtual power plants according to any one of claims 1 to 7.

Images