Market-based virtual power plant planning method considering uncertainty

Abstract

The invention discloses a market-based virtual power plant planning method considering uncertainty. The method comprises the following steps: 1) proposing an MB-VPP energy transaction model; the system is composed of renewable energy sources, a controllable load participating in a demand response plan and an energy storage system running in a day-ahead energy market. And 2) the proposed method checks two different scenes, one is that the controllable load participates in the demand response plan, and the other is that the controllable load does not participate in the demand response plan. The influences of the two methods on social welfare, wind turbines, photovoltaic and energy storage system active power distribution and load transmission capacity are analyzed; and 3) modeling the uncertainty related to the wind turbine, photovoltaic and load requirements by using a scene-based method. And 4) solving an optimization problem by using a CPLEX solver in a GAMS environment. According to the method, the fluctuation of renewable energy sources can be effectively reduced, the reliability of a power system is improved, consumer benefits are maximized, and meanwhile the energy cost is minimized.

Description

Technical Field

[0001] The invention belongs to the field of electric power and is a market-based virtual power plant planning method taking uncertainty into consideration. Background Art

[0002] The benefits of renewable energy are twofold: first, they do not contribute to the carbonization of the environment, and second, they do not contribute to the warming effects of carbon emissions. Organizations, researchers, academics, and scientists are paying close attention to the migration of traditional energy systems to non-traditional energy systems. This is due to the ubiquity of renewable energy sources, which are environmentally friendly, non-polluting, and can be managed locally. In order to stimulate investment in renewable energy development, some countries have launched incentive-based renewable energy development programs. Enabling these small-scale electricity suppliers to exchange electricity or provide ancillary services to the power network minimizes the pressure on the grid during peak hours. Thus, improving the reliability of the power system.

[0003] However, integrating renewable energy systems into energy markets is quite challenging. The effectiveness of renewable energy systems, especially wind turbines and solar photovoltaic panels, is severely affected by climate change. Their generation is unreliable, intermittent, and limited by location. As a result, the efficiency of the operation of the power system may be affected, resulting in grid instability. In addition, renewable energy generation is usually limited, especially when considering photovoltaic power generation with a maximum output of 10 kW. This makes it challenging to predict whether renewable energy generation can keep up with the established schedule of the electricity market.

[0004] To overcome these challenges, a virtual power plant is a flexible decentralized network of small to medium-sized generators interconnected by a single centralized system. These units can be power sources, electricity consumers, demand response, and energy storage systems. The virtual power plant operator monitors and manages all the network's aggregated assets using a centralized control algorithm. The purpose of a virtual power plant is to jointly estimate, deliver, and trade the power and flexibility of assets in a day-ahead energy market. Thus, small units can access markets that they would not be able to enter on their own. Although demand response can be described as a shift in consumers' energy use to better balance the demand and supply of energy, it is a viable way to minimize peak demand.

[0005] Virtual power plants are a fairly new phenomenon that combine the capabilities of renewable and non-renewable energy and storage technologies to operate in future energy markets to improve system reliability. It also improves grid reliability by providing utilities with additional distribution-level operational capacity. Therefore, the use of energy storage systems to implement demand response programs is expected to have a profound positive impact on social welfare, wind turbines, solar photovoltaic panels and energy storage systems. Active power generation and transmission capabilities. Summary of the invention

[0006] The purpose of the invention is to overcome the shortcomings of the prior art and provide a market-based virtual power plant planning method that takes uncertainty into consideration, which is scientific, reasonable, highly applicable, accurate and effective. The method can effectively improve the active and active scheduling capabilities of wind power, photovoltaic and energy storage systems.

[0007] To achieve the above object, the present invention provides the following technical solution: a market-based virtual power plant planning method considering uncertainty, characterized in that the steps are:

[0008] Step 1: Use the hydropower planning method to collect historical data on market prices, renewable energy generation, energy storage systems, and load demand within the setting;

[0009] Step 2: Use the historical data from the mobile phone in step 1 to build an uncertainty model related to wind speed, solar radiation, and load demand;

[0010] 2.1) Wind speed modeling

[0011] Modeling the wind speed variation in a specific area: The probability density function relating wind speed to wind turbine module output power is expressed as:

[0012]

[0013] Wherein, v, c and k represent wind speed, probability density index and characteristic index respectively, and the output power of wind turbine components is detected through power curve;

[0014] In equation (2), P wt and P rated represent the wind turbine output power and rated power, respectively, v ci and v co The cut-in and cut-out speeds are represented by v respectively, and the rated speed of the wind turbine is represented by v r express;

[0015]

[0016] Therefore, the output power of the wind turbine module at bus i and scenario w is expressed as follows:

[0017]

[0018] in, is the percentage of active power produced by the wind turbine;

[0019] 2.2) Solar irradiance model

[0020] The probability density function of the β coefficient is used to simulate the solar irradiance as follows:

[0021]

[0022] The unit of solar irradiance is kW / m 2 , represented by S, α and β are determined as follows:

[0023]

[0024] Where μ and σ represent the mean and standard deviation, solar radiation and ambient temperature are key parameters for determining the output power of PV modules, expressed in formulas (7) and (8);

[0025]

[0026] Among them, P PV and P STC , respectively represent the output power of the solar photovoltaic panel module and the power under standard test conditions; the power temperature coefficient is represented by δ, T cell Indicates battery temperature; ambient temperature is T amb Indicates that the nominal operating temperature of the battery is denoted by NOCT, and G represents the solar irradiance;

[0027] 2.3) Load demand uncertainty modeling

[0028] The load demand of each bus is modeled using a normal probability density function: The normal distribution probability density function of the undetermined load l is given by formula (9) as follows:

[0029]

[0030] Among them, σ l is the standard deviation, μ l is the average value.

[0031] Step 3: Establish a market model and demand response plan, make corresponding objective functions and constraints, and obtain the interaction law between the virtual power plant and the electricity market.

[0032] 3.1) Establishing market model and method description

[0033] Assuming that all trading operations occurring within the virtual power plant structure are centrally managed, electricity can be sold and purchased on the day-ahead market the day before the scheduled delivery date, and sellers and buyers can submit bids to the market operator to participate in the day-ahead energy market; owners of virtual power plants participating in the day-ahead market submit energy bids and control the components of the total bid in a way that maximizes overall social welfare, thereby determining how much energy can be bought or sold on the electricity market;

[0034] 3.2) Make a demand response plan

[0035] The two basic categories of demand response programs are time-based programs and incentive-based programs;

[0036] 3.2.1) Incentive-based programs

[0037] The incentive-based scheme is broken down into six components, including direct load control, interruptible / curtailable services, emergency demand response, demand bidding, capacity market and ancillary services market;

[0038] Direct Load Control (DLC) applications allow system operators to remotely manage and change customer consumption. In the event of a power system emergency, customers participating in the “interruptible / curtailable” program are obliged to reduce their consumption to a minimum; otherwise, they risk being penalized.

[0039] In emergency demand response programs, the system operator compensates customers for reducing energy consumption when the system is under stress conditions; in demand-based bidding programs, large customers participate in the electricity market to gain financial benefits; in capacity market programs, consumers must reduce their pre-planned load or face penalties; in ancillary service market programs, consumers notify the system operator of their intention to reduce consumption and then compete as a reserve substitute in the ancillary service sector;

[0040] 3.2.2) Time-based planning

[0041] The time-based program is broken down into three parts, including time of use, real-time pricing, and critical peak pricing;

[0042] In a time-of-use plan, energy prices fluctuate with load consumption. There are generally three basic energy consumption segments; peak load periods have a high energy price, while off-peak and off-peak periods have average and low prices, respectively. The real-time pricing (RTP) program is the same as the time-of-use program (TOU), except that the pricing offered through RTP is wide-ranging. The use of critical peak pricing plans is limited to emergency situations.

[0043] 3.3) Establishment of objective function

[0044] The goal of virtual power plant operational planning is to maximize social welfare in equation (10), which maximizes the welfare of end users and minimizes the cost of energy production within the planning horizon, as follows:

[0045]

[0046] in, represents the active power demand, represents the bid price charged to consumers, is the amount of electricity exchanged with the grid, is the price of electricity exchanged with the grid, represent wind power generation and its production cost respectively; They represent photovoltaic power generation and its production cost respectively; They represent the active power reduction amount and the incentive price provided by the demand response program respectively. They represent the charging and discharging of the energy storage system (ESS), and their costs are respectively given by Indicates; η CH , η DIS They represent the charging and discharging efficiency of the energy storage system respectively;

[0047] 3.4) Constraints

[0048] Wind turbine installed capacity: Equation (11) shows the maximum and minimum installed capacity of the wind turbine generator;

[0049]

[0050] Photovoltaic installed capacity: Equation (12) represents the minimum and maximum power generation capacity of the photovoltaic unit;

[0051]

[0052] Maximum installed capacity: Equation (13) represents the maximum installed capacity of wind turbines and photovoltaics;

[0053]

[0054] The minimum and maximum allowed energy exchange limits between the upstream grid and the virtual power plant are expressed by equation (14);

[0055]

[0056] Demand response constraint: Equation (15) describes the contribution from the demand response provider as the demand drop in each time period and scenario;

[0057]

[0058] Voltage limit: Equation (16) represents the maximum and minimum voltage limits on each bus;

[0059]

[0060] Energy storage system operation: The energy storage system operation is represented by equations (17)-(23).

[0061]

[0062] in, Respectively represent the charging and discharging of the energy storage system ESS, is a binary variable, which means that charging and discharging of the energy storage system is not allowed in the same time period. E ess represents the maximum and minimum energy storage in the energy storage system. The initial and final energy storage levels in the energy storage system are represented by and Indicates; η CH , η DIS Respectively represent the charging and discharging efficiency of the energy storage system. (t) Indicates the charging status of the energy storage system;

[0063] Power balance constraint: Equation (23) represents the power balance constraint that applies to each bus and must be satisfied, stating the required generation and desired amount of power that should be satisfied for the virtual power plant;

[0064]

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. This paper proposes a new MB-VPP energy trading model. It consists of RESS participating in the disaster recovery plan, controllable loads and energy storage systems operating in the day-ahead energy market;

[0067] 2. The method proposed in this invention examines two different situations, one is the controllable load participating in the demand response plan, and the other is the controllable load not participating in the demand response plan. The impact of these two methods on social welfare, wind power, photovoltaic, energy storage system active scheduling and load transmission capacity is analyzed;

[0068] 3. Use scenario-based approaches to model the uncertainties associated with wind turbines, photovoltaics, and load demand. Use CPLEX solvers in the GAMS environment to solve optimization problems;

[0069] 4. The proposed approach can convince end users to participate in the electricity market to alleviate the peak load of the grid and, therefore, contribute to the reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 It is the optimization solution flow chart;

[0071] Figure 2 is the wind turbine wind speed power curve;

[0072] Figure 3 It is a diagram of the structure of the electricity market of a virtual power plant;

[0073] Figure 4It is a virtual power plant day-ahead market strategy diagram;

[0074] Figure 5 is a classification diagram of demand response programs;

[0075] Figure 6 It is a universal single-line diagram for 16 buses;

[0076] Figure 7 is the total active power diagram of wind turbine dispatch on the candidate bus in scenario AB;

[0077] Figure 8 It is the total active power diagram of the PV generators dispatched on the candidate bus in scenario AB;

[0078] Fig. 9 It is the charging power diagram of the energy storage unit within 24 hours when the demand response strategy is adopted in scenario AB;

[0079] Fig.10 It is the discharge power diagram of the energy storage unit within 24 hours when the demand response strategy is adopted in scenario AB;

[0080] Fig.11 This is a diagram of the social welfare situation under two scenarios. DETAILED DESCRIPTION

[0081] The present invention will be further described below using the accompanying drawings and embodiments.

[0082] Step 1: Collect historical data on market prices, renewable energy generation, energy storage systems, and load demand to provide a basis for modeling in Step 2.

[0083] The main objective of this study is to optimize the social welfare of virtual power plants by integrating demand response programs in the electricity market into the proposed model, e.g. Figure 1 As shown, in the electricity market, market-based optimal power flow is used to evaluate and solve the operational planning problem of a virtual power plant. The proposed method evaluates the effect of controllable loads participating in demand response programs, which affects the social welfare and the load demand of wind turbines, photovoltaics and energy storage systems in the planning horizon by maximizing the active generation and transmission capacity. Demand response programs and energy storage systems are good alternatives to solve the imbalance between supply and demand, mitigate the impact of uncertainties associated with renewable energy sources, and relieve load demand during peak hours. Figure 1 The step-by-step solution of the problem is shown. The stochastic programming method is used to initialize hour t and collect historical data on market prices, renewable energy generation, energy storage system and load demand. The objective function aims to maximize the social welfare in (10) under the equality and inequality constraints of the system.

[0084] Uncertainty modeling was then performed. An ARBO-based approach was used to model the uncertainties associated with solar radiation, wind speed, and load demand, where scenarios were defined as the likelihood of realization of the uncertain parameters. Probability density functions (PDFs) were used for each solar radiation, wind speed, and load demand to generate 24 scenarios.

[0085] Step 2: Build uncertainty models related to wind speed, solar radiation, and load demand.

[0086] Uncertainty Modeling

[0087] 2.1) Wind speed modeling

[0088] The wind speed variation in a specific area is modeled. The probability density function relating wind speed to wind turbine module output power is expressed as:

[0089]

[0090] Among them, v, c and k represent the wind speed, probability density index and characteristic index respectively. Therefore, the output power of the wind turbine component can be detected through the power curve.

[0091] In equation (2), P wt and P rated Represent the wind turbine output power and rated power respectively. ci and v co The rated speed of the wind turbine is expressed as v r express. Figure 2 The wind speed power curve for a wind turbine is shown.

[0092]

[0093] Therefore, the output power of the wind turbine module at bus i and scenario w can be expressed as follows:

[0094]

[0095] in, is the percentage of active power produced by the wind turbine.

[0096] 2.2) Solar irradiance model

[0097] The probability density function of the β coefficient is used to simulate the solar irradiance as follows:

[0098]

[0099] Solar irradiance (kW / m 2 ) is represented by S. α and β are determined as follows:

[0100]

[0101] Where μ and σ represent the mean and standard deviation. Solar radiation and ambient temperature are key parameters for determining the output power of PV modules, which are expressed in formulas (7) and (8).

[0102]

[0103] Among them, P PV and P STC They represent the output power of the solar photovoltaic panel module (unit: kilowatt) and the power under standard test conditions (unit: megawatt). The power temperature coefficient (% / ℃) is represented by δ, T cell Indicates the battery temperature. The ambient temperature (unit: °C) is T amb The nominal operating temperature of the battery (unit: °C) is represented by NOCT. G represents solar irradiance, in W / m 2 .

[0104] 2.3) Load demand uncertainty modeling

[0105] The load demand of each bus is modeled using a normal probability density function. The normal distribution probability density function of the undetermined load l is given by formula (9) as follows:

[0106]

[0107] Among them, σ l is the standard deviation, μ l is the average value.

[0108] Step 3: Establish a market model and demand response plan, make corresponding objective functions and constraints, and obtain the interaction law between the virtual power plant and the electricity market.

[0109] 3.1) Establishing market model and method description

[0110] Virtual power plants bring together different distributed energy technologies in the electricity market to improve the operational performance and reliability of the power system. In our proposed model, we assume that we will centrally manage all trading operations that take place within the virtual power plant structure. Electric energy can be sold and bought on the day-ahead market the day before the scheduled delivery date. Sellers and buyers can present their bids to the market operator to participate in the day-ahead energy market. The owners of virtual power plants participating in the day-ahead market present their energy bids and control the components of the total bid in a way that maximizes the overall social welfare. This determines how much energy can be bought or sold on the electricity market. The proposed model is as follows Figure 3The virtual power plant model includes energy storage, power loads, wind farms, photovoltaic power generation, wind energy and demand response programs. In this study, a direct control approach is used to ensure that distributed energy resources are coordinated and effectively represented in the market. Figure 4 The performance of the virtual power plant market strategy in the day-ahead electricity market is demonstrated.

[0111] 3.2) Make a demand response plan

[0112] The two basic categories of demand response programs are time-based programs and incentive-based programs. Figure 5 This classification is illustrated graphically.

[0113] 3.2.1) Incentive-based programs

[0114] The incentive-based programs are broken down into six segments, including direct load control, interruptible / curtailable services, emergency demand response, demand bidding, capacity market, and ancillary services market. The direct load control (DLC) application allows system operators to remotely manage and change client consumption. Customers participating in the interruptible / curtailable program are obliged to reduce their consumption to a minimum during power system emergencies; otherwise, they risk being penalized. In the emergency demand response program, system operators compensate users for reducing energy consumption when the system is under stress conditions. In the demand bidding program, large customers (such as generation companies) can participate in the power market to gain financial benefits. In the capacity market program, consumers must reduce their pre-planned loads or face penalties. Consumers in the ancillary services market program notify the system operator of their intention to reduce consumption. They can then compete as standby substitutes in the ancillary services sector.

[0115] 3.2.2) Time-based planning

[0116] Time-based programs are broken down into three segments including Time of Use, Real-Time Pricing, and Critical Peak Pricing. In a time-of-use plan, the price of energy fluctuates based on the load consumption. There are three basic energy consumption segments; peak load periods have a high energy price, while off-peak and off-peak periods have average and low prices, respectively. Real-time pricing (RTP) programs are the same as time-of-use (TOU) programs, with the main difference being the wide range of pricing offered through RTP. The use of critical peak pricing programs is limited to emergency situations. These types of programs are helpful when there is a lot of demand on the system.

[0117] 3.3) Establishment of objective function

[0118] The goal of virtual power plant operational planning is to maximize social welfare in equation (10). It maximizes the welfare of end users and minimizes the cost of energy production within the planning horizon as follows:

[0119]

[0120] in, represents the active power demand, represents the bid price charged to consumers, is the amount of electricity exchanged with the grid, is the price of electricity exchanged with the grid, represent wind power generation and its production cost respectively; They represent photovoltaic power generation and its production cost respectively; They represent the active power reduction amount and the incentive price provided by the demand response program respectively. They represent the charging and discharging of the energy storage system (ESS), and their costs are respectively given by Indicates; η CH , η DIS Respectively represent the charging and discharging efficiency of the energy storage system.

[0121] 3.4) Constraints

[0122] Wind turbine installed capacity: Equation (11) shows the maximum and minimum installed capacity of the wind turbine generator.

[0123]

[0124] PV installed capacity: Equation (12) represents the minimum and maximum power generation capacity of the PV system.

[0125]

[0126] Maximum installed capacity: Equation (13) represents the maximum installed capacity of wind turbines and photovoltaics.

[0127]

[0128] The minimum and maximum allowed energy exchange constraints between the upstream grid and the virtual power plant are expressed by equation (14).

[0129]

[0130] Demand response constraint: Equation (15) describes the contribution from the demand response provider as the demand drop in each time period and scenario.

[0131]

[0132] Voltage limits: Equation (16) expresses the maximum and minimum voltage limits on each bus.

[0133] V i MIN ≤V i ≤V i MAX (16)

[0134] Energy storage system operation: The energy storage system operation is represented by equations (17)-(23).

[0135]

[0136] in, represent the charging and discharging of the energy storage system (ESS), is a binary variable, which means that charging and discharging of the energy storage system is not allowed in the same time period. E ess represents the maximum and minimum energy storage in the energy storage system. The initial and final energy storage levels in the energy storage system are represented by and Indicates; η CH , η DIS Respectively represent the charging and discharging efficiency of the energy storage system. (t) Indicates the charging status of the energy storage system.

[0137] Power balance constraint: Equation (23) represents the power balance constraint that applies to each bus and must be satisfied. This means that the required generation and the desired amount of power of the virtual power plant should be met.

[0138]

[0139] Step 4: The effectiveness of the proposed strategy is verified through scenario simulation and case analysis.

[0140] (1) Case analysis

[0141] The effectiveness of the proposed strategy is illustrated by case analysis and the results are verified. For numerical analysis, a general 16-node 33KV general distribution system is used. The single-line schematic diagram of the 16-way general distribution system is shown in Figure 6As shown. During normal operation, the virtual power plant is connected to the main grid. The final capacity of photovoltaic, energy storage system and wind turbines is determined by applying the optimal power flow within the planning horizon. In this study, we assume that the distribution network consists of three photovoltaic units and four wind turbines. In the distribution network, there are three photovoltaic candidate buses with a rated power generation of 440kW, located at positions 12, 14 and 16, and wind turbine candidate buses are located at positions 3, 6, 9 and bus 11 with a nominal power generation of 660kW. A 200kW energy storage system is installed on the candidate bus at the location 15 they selected. The upper and lower voltage values ​​are considered as V max =1.06pu、V min =0.94pu. The active power cost of the disaster recovery plan is assumed to be $10 / kW, which is paid to users to reduce the load demand of each bus. In this study, we study the operation planning problem of virtual power plants in the electricity market. The virtual power plant operation planning problem in the energy environment is studied in a novel way, which maximizes social welfare while respecting the constraints of the problem. In order to reduce the amount of computation and improve the efficiency and accuracy of problem solving, the number of scenarios in this study is limited to 24. The problem is formulated as a mixed integer linear programming (MILP) problem using General Algebraic Modeling Simulation (GAMS), optimization software, and CPLEX solver to produce the best results.

[0142] (2) Simulation results

[0143] The highest power dispatch bus of each energy system is shown in Table 1.

[0144] Table 1 Two scenarios with and without demand response

[0145]

[0146] Two different scenarios are considered in this paper to evaluate the impact of demand response schemes on the active power, social welfare and energy storage systems of virtual power plants aggregating wind turbines and photovoltaic dispatch. The total active power of wind power, photovoltaic and energy storage systems under scenario A and scenario B is studied. Figure 7 and Figure 8 The total amount of active power dispatched by wind turbines and PV at each candidate bus for scenarios A and B is shown in Figure 2. In scenario B, where the demand response scheme is adopted, the total active output of wind power and PV is significantly higher than that in scenario A. The rationality of the highest power dispatch bus for each energy system is shown in Table 2.

[0147] Table 2 Bus with the highest power dispatch

[0148]

[0149] The energy storage unit has two operation modes: when the energy price is low, it operates as a load and stores energy, and when the energy price is high, it operates as a producer of energy and sells the energy at a profit on the market. Fig. 9 and Fig.10 Scenario B, which adopts a demand response strategy, is shown to have more charging and discharging of the energy storage unit than in scenario A. This is because the demand response scheme allows some loads to be shifted to other low demand periods. In addition, according to Fig. 9 As shown, the energy storage device is scheduled to be charged at 1-6h and 12-17h (off-peak hours). During these hours, the energy storage device is charged with a total power of 49.3kW. Fig.10 As shown, the energy storage device is scheduled to discharge during the hours 7-11h and 18-24h (peak hours). The energy storage device released a total of 49.2 kW of power. The energy storage system performed well in flattening the load curve and improving grid stability. This reduced the virtual power plant's dependence on the utility grid and also reduced its overall energy costs. The social benefits in the two cases are as follows Fig.11 It can be observed that the social welfare in scenario B is greater than that in scenario A. This is mainly attributed to the integration of demand response scheme and the increased active power dispatched in scenario B, which allows for higher social welfare.

[0150] A new MB-VPP energy trading model is proposed. It consists of renewable energy, controllable loads participating in demand response programs, and energy storage systems operating in the day-ahead energy market. The proposed approach examines two different scenarios, one in which the controllable loads participate in the demand response program and the other in which the controllable loads do not participate in the demand response program. The impact of these two approaches on social welfare, wind turbines, photovoltaic generators, and active power distribution and delivery capacity to the loads are analyzed. A scenario-based approach is used to model the uncertainties associated with wind turbines, photovoltaic generators, and load demand. The CPLEX solver is used in the GAMS environment to solve the optimization problem. The simulation results show that the MB-VPP is able to effectively power the loads.

[0151] 1. In scenario B, the active power dispatch capacity of wind turbines and PV generators at each candidate bus is improved by 3% during the planning period due to the integration of demand response programs in the proposed model.

[0152] 2. Similarly, the controllable load participating in the demand response program used in scenario B also increases the social welfare of the virtual power plant by 3%.

[0153] 3. The proposed method can convince end users to participate in the electricity market to alleviate the peak load of the power grid, thus helping to improve the reliability of the power system.

Claims

1. A market-based virtual power plant planning method considering uncertainty, characterized in that: The steps are: Step 1: Use the hydropower planning method to collect historical data on market prices, renewable energy generation, energy storage systems, and load demand within the setting; Step 2: Use the historical data from the mobile phone in step 1 to build an uncertainty model related to wind speed, solar radiation, and load demand; Step 3: Establish a market model and demand response plan, make corresponding objective functions and constraints, and obtain the interaction law between the virtual power plant and the electricity market.

2. A market-based virtual power plant planning method considering uncertainty according to claim 1, characterized in that: The specific method in step 2 is: 2.1) Wind speed modeling Modeling the wind speed variation in a specific area: The probability density function relating wind speed to wind turbine module output power is expressed as: Wherein, v, c and k represent wind speed, probability density index and characteristic index respectively, and the output power of wind turbine components is detected through power curve; In equation (2), P wt and P rated represent the wind turbine output power and rated power, respectively, v ci and v co The cut-in and cut-out speeds are represented by v respectively, and the rated speed of the wind turbine is represented by v r express; Therefore, the output power of the wind turbine module at bus i and scenario w is expressed as follows: in, is the percentage of active power produced by the wind turbine; 2.2) Solar irradiance model The probability density function of the β coefficient is used to simulate the solar irradiance as follows: The unit of solar irradiance is kW / m 2 , represented by S, α and β are determined as follows: Where μ and σ represent the mean and standard deviation, solar radiation and ambient temperature are key parameters for determining the output power of PV modules, which are expressed by formulas (7) and (8); Among them, P PV and P STC , respectively represent the output power of the solar photovoltaic panel module and the power under standard test conditions; the power temperature coefficient is represented by δ, T cell Indicates battery temperature; ambient temperature is T amb Indicates that the nominal operating temperature of the battery is denoted by NOCT, and G represents the solar irradiance; 2.3) Load demand uncertainty modeling The load demand of each bus is modeled using a normal probability density function: The normal distribution probability density function of the undetermined load l is given by formula (9) as follows: Among them, oh l is the standard deviation, μ l is the average value.

3. According to the market-based virtual power plant planning method considering uncertainty according to claim 1, it is characterized in that the specific method in step 3 is: 3.1) Establishing market model and method description Assuming that all trading operations occurring within the virtual power plant structure are centrally managed, electricity can be sold and purchased on the day-ahead market the day before the scheduled delivery date, and sellers and buyers can submit bids to the market operator to participate in the day-ahead energy market; owners of virtual power plants participating in the day-ahead market submit energy bids and control the components of the total bid in a way that maximizes overall social welfare, thereby determining how much energy can be bought or sold on the electricity market; The proposed model is shown in Figure 3. The wholesale energy market interacts with the virtual power plant market operator through electricity prices and energy exchange, while the market-based virtual power plant and its operator interact with each other. The virtual power plant model includes energy storage, power load, wind farm, photovoltaic power generation, wind energy and demand response plan. In this study, a direct control method is used to ensure the coordination of distributed energy and effective representation in the market. Figure 4 shows the performance of the virtual power plant market strategy in the day-ahead power market. 3.2) Make a demand response plan The two basic categories of demand response programs are time-based programs and incentive-based programs; 3.2.1) Incentive-based programs The incentive-based scheme is broken down into six components, including direct load control, interruptible / curtailable services, emergency demand response, demand bidding, capacity market and ancillary services market; Direct Load Control DLC application allows system operators to remotely manage and change customer consumption. In case of power system emergency, customers participating in the "interruptible / curtailable" program are obliged to reduce their consumption to the minimum; otherwise, they have the chance of being penalized; In emergency demand response programs, the system operator compensates customers for reducing energy consumption when the system is under stress conditions; in demand-based bidding programs, large customers participate in the electricity market to gain financial benefits; in capacity market programs, consumers must reduce their pre-planned load or face penalties; in ancillary service market programs, consumers notify the system operator of their intention to reduce consumption and then compete as a reserve substitute in the ancillary service sector; 3.2.2 Time-based planning The time-based program is broken down into three parts, including time of use, real-time pricing, and critical peak pricing; In a time-of-use plan, energy prices fluctuate with load consumption. There are generally three basic energy consumption segments; peak load periods have a high energy price, while off-peak and off-peak periods have average and low prices, respectively. The real-time pricing (RTP) program is the same as the time-of-use program (TOU), except that the pricing offered through RTP is wide-ranging. The use of critical peak pricing plans is limited to emergency situations. 3.3) Establishment of objective function The goal of virtual power plant operational planning is to maximize social welfare in equation (10), which maximizes the welfare of end users and minimizes the cost of energy production within the planning horizon, as follows: in, represents the active power demand, represents the bid price charged to consumers, is the amount of electricity exchanged with the grid, is the price of electricity exchanged with the grid, represent wind power generation and its production cost respectively; They represent photovoltaic power generation and its production cost respectively; They represent the active power reduction amount and the incentive price provided by the demand response program respectively. They represent the charging and discharging of the energy storage system (ESS), and their costs are respectively given by Indicates; η CH , η DIS They represent the charging and discharging efficiency of the energy storage system respectively; 3.4) Constraints Wind turbine installed capacity: Equation (11) shows the maximum and minimum installed capacity of the wind turbine generator; Photovoltaic installed capacity: Equation (12) represents the minimum and maximum power generation capacity of the photovoltaic unit; Maximum installed capacity: Equation (13) represents the maximum installed capacity of wind turbines and photovoltaics; The minimum and maximum allowed energy exchange limits between the upstream grid and the virtual power plant are expressed by equation (14); Demand response constraint: Equation (15) describes the contribution from the demand response provider as the demand drop in each time period and scenario; Voltage limit: Equation (16) represents the maximum and minimum voltage limits on each bus; In i MIN ≤V i ≤V i MAX (16) Energy storage system operation: The energy storage system operation is represented by equations (17)-(23). in, Respectively represent the charging and discharging of the energy storage system ESS, is a binary variable, which means that charging and discharging of the energy storage system is not allowed in the same time period. E ess represents the maximum and minimum energy storage in the energy storage system. The initial and final energy storage levels in the energy storage system are represented by and Indicates; η CH , η DIS Respectively represent the charging and discharging efficiency of the energy storage system. (t) Indicates the charging status of the energy storage system; Power balance constraint: Equation (23) represents the power balance constraint that applies to each bus and must be satisfied, stating the required generation and desired amount of power that should be satisfied for the virtual power plant;

Images