Virtual power plant adjustable capacity boundary construction method considering conditional value-at-risk

Abstract

The invention relates to a virtual power plant adjustable capacity boundary construction method considering conditional value-at-risk, and the method comprises the steps: firstly constructing a virtual power plant electric energy dispatching model based on time-of-use electricity price guidance by taking the minimum electricity purchasing cost of a virtual power plant spot market as an objective function; constructing a virtual power plant regulation capacity boundary depicting model by taking maximum output and minimum output of the virtual power plant in the regulation time period as objective functions; the maximum output and the minimum output of the virtual power plant in the adjustment period respectively correspond to the upper limit and the lower limit of the adjustment capacity boundary of the virtual power plant; solving the virtual power plant electric energy scheduling model to obtain a scheduling strategy of the virtual power plant in the electric energy market; and finally, according to a scheduling strategy of the virtual power plant in the electric energy market, performing global optimization on the virtual power plant adjustment space capacity description model to obtain an adjustable capacity boundary of the virtual power plant in an adjustment time period. The invention is beneficial to improving the flexibility of the power system, promoting the green and low-carbon development of the power industry and optimizing the energy structure.

Description

Technical Field

[0001] The invention relates to a method for constructing an adjustable capacity boundary of a virtual power plant considering conditional risk value, and belongs to the technical field of energy distribution. Background Art

[0002] In recent years, with the goal of building a new power system, a large number of heterogeneous distributed energy sources have been connected to the power grid. By the end of 2022, the cumulative installed capacity of distributed photovoltaic, distributed wind power, distributed gas, and small hydropower as the main distributed energy sources in my country has reached 250.0049 million kilowatts, a year-on-year increase of 25.06%. Therefore, the urgency and importance of the integrated development of source, grid, load and storage are important. Virtual power plants aggregate various resources scattered in the power grid, such as distributed power sources, energy storage, and loads, and conduct coordinated optimization, operation control, and market transactions. They can better tap the potential of distributed resources and are an important method to promote the organic interaction of source, grid, load and storage and enhance system flexibility.

[0003] Virtual power plants generally participate in multiple types of electricity markets by declaring external characteristic parameters such as operating curves, ramp rates, and feasible domains. At present, virtual power plant external characteristic extraction technologies mostly focus on using vertex search or Minkowski sum methods to extract the PQ feasible domain of virtual power plants, and using full probability criteria and key parameter extraction methods to characterize the probability distribution of virtual power plant output. However, in the process of declaring the adjustable capacity of virtual power plants, it is necessary to comprehensively consider the risks caused by the uncertainty of new energy. Therefore, it is of great engineering significance to study the method of constructing the adjustable capacity boundary of virtual power plants considering conditional risk value. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a method for constructing an adjustable capacity boundary of a virtual power plant by introducing conditional risk value to deal with the uncertainty risk of new energy output.

[0005] In order to achieve the above object, the technical solution proposed by the present invention is: a method for constructing an adjustable capacity boundary of a virtual power plant considering conditional risk value, comprising the following steps:

[0006] Step 1: In the electric energy dispatching stage, the objective function is to minimize the purchase cost of electricity in the virtual power plant spot market, and to construct an electric energy dispatching model of the virtual power plant guided by time-of-use electricity prices;

[0007] Step 2: Taking the maximum and minimum output of the virtual power plant during the regulation period as the objective function, and taking the conditional risk value constraint including the regulation power compensation price and the regulation cost opportunity constraint including the loss risk aversion coefficient as the constraint conditions, a virtual power plant regulation capacity boundary characterization model is constructed; the maximum and minimum output of the virtual power plant during the regulation period correspond to the upper and lower limits of the virtual power plant regulation capacity boundary respectively;

[0008] Step 3: Solve the electric energy dispatch model of the virtual power plant to obtain the dispatch strategy of the virtual power plant in the electric energy market;

[0009] Step 4: According to the dispatch strategy of the virtual power plant in the electric energy market, the virtual power plant regulation space capacity characterization model is globally optimized to obtain the adjustable capacity boundary of the virtual power plant during the regulation period.

[0010] A further design of the above technical solution is as follows: the objective function of the virtual power plant electric energy scheduling model is:

[0011]

[0012] in, represents the electricity cost of distributed photovoltaic p at time t; represents the electricity cost of gas turbine m at time t; represents the electricity cost of energy storage e at time t; represents the electricity cost of the flexible load h at time t; represents the electricity cost of electric vehicle v at time t; represents the revenue from electricity purchase and sale of the virtual power plant at time t;

[0013] The electricity cost of distributed photovoltaic is:

[0014]

[0015] Among them, c pv The grid-connected electricity price of distributed photovoltaic power generation is RMB 391 / MWh; represents the output of distributed photovoltaic p at time t; λ pv Indicates the unit abandonment penalty cost; It represents the predicted value of distributed photovoltaic p output;

[0016] The cost of electricity from a gas turbine is:

[0017]

[0018] in, represents the g-th stage cost operation state variable of gas turbine m at time t; l m,g , b m,g represents the slope and intercept of the g-th cost operation of gas turbine m at time t; represents the output of gas turbine m at time t;

[0019] The electricity cost of energy storage is:

[0020]

[0021] in, Indicates the unit energy cost of energy storage; represents the charging and discharging power of energy storage e at time t;

[0022] The electricity cost of flexible load is:

[0023]

[0024] Among them, TEP t represents the time-of-use electricity price at time t; represents the actual power consumption of the flexible load h at time t; represents the subsidy price of flexible load response to virtual power plant dispatch at time t; represents the power of the flexible load h in response to the virtual power plant dispatch at time t;

[0025] The electricity cost of electric vehicles is:

[0026]

[0027] in, represents the unit energy cost of electric vehicle v at time t, and takes represents the actual power consumption of electric vehicle v at time t; represents the subsidized electricity price of electric vehicle v in response to the virtual power plant dispatch at time t; represents the power of the flexible load v in response to the virtual power plant dispatch at time t;

[0028] The revenue from purchasing and selling electricity of a virtual power plant is:

[0029]

[0030] in, represents the purchase and sale price of electricity of the virtual power plant, Represents the amount of electricity purchased and sold by the virtual power plant at time t.

[0031] The constraints of the virtual power plant electric energy dispatch model include power balance constraints, gas turbine operation constraints, energy storage constraints, photovoltaic output constraints, electric vehicle constraints and flexible load constraints.

[0032] The power balance constraints are as follows:

[0033]

[0034] The gas turbine operating constraints are as follows:

[0035]

[0036] In the formula, It represents the upper and lower limits of the output of the gas turbine unit m of the virtual power plant; It represents the upper limit of the ramp rate of the gas turbine unit m in the virtual power plant;

[0037] The energy storage constraints are as follows:

[0038]

[0039] SOC e,T =SOC e,0

[0040] In the formula, is a Boolean variable, indicating the charge and discharge status of the energy storage e; Indicates the maximum charge and discharge power of energy storage e; SOC e,t Indicates the amount of energy stored e at time t; SOC e,min , SOC e,max They are the maximum and minimum power allowed by energy storage e; SOC e,0 is the initial capacity of energy storage e, set to half of the maximum capacity of energy storage; η e is the charging and discharging efficiency of energy storage e;

[0041] The photovoltaic output constraints are as follows:

[0042]

[0043] The flexible load constraints are as follows:

[0044]

[0045] In the formula, represents the predicted value of the electric energy demand of the flexible load h at time t; It represents the cumulative amount of electric energy demand of flexible load h that is not met at time t;

[0046] The electric vehicle constraints are as follows:

[0047]

[0048] In the formula, represents the predicted value of the charging demand of electric vehicle v at time t; It represents the cumulative amount of electric energy demand of flexible load v that is not met at time t.

[0049] The objective function of the virtual power plant regulation capacity boundary characterization model is as follows:

[0050]

[0051] in, It means that after the virtual power plant provides regulation services, c Load demand at any given moment; Indicates the power purchase and sale adjustment power of the virtual power plant.

[0052] The conditional VaR constraints are:

[0053]

[0054]

[0055] v(s)≥0

[0056] Among them, C control (s) represents the comprehensive cost of the virtual power plant providing regulation services under the scenario of s renewable energy output; ε pv , represents the adjustment power subsidy of the virtual power plant to photovoltaic, electric vehicles and flexible loads; α represents the adjustment power compensation price in the regulatory market; represents the probability of occurrence of the new energy output scenario; L VaR,τ , L CVaR,τ represents the risk value and conditional risk value under the confidence level τ; ζ, v(s) are auxiliary variables for calculating risk value and conditional risk value;

[0057] The adjustment cost opportunity constraint is:

[0058]

[0059] Among them, γ represents the virtual power plant's aversion to loss risk. The larger γ is, the more the power sales company averses loss risk.

[0060] The constraint conditions of the virtual power plant regulation capacity boundary characterization model also include power balance constraints after regulation, gas turbine regulation constraints, energy storage regulation constraints, photovoltaic output regulation constraints, electric vehicle regulation constraints, and flexible load regulation constraints.

[0061] The adjusted power balance constraint is as follows:

[0062]

[0063] in, It represents the regulation power of photovoltaic, gas turbine, flexible load, energy storage charging, energy storage discharging, and electric vehicle charging scheduling strategy under the scenario s at time t;

[0064] The gas turbine regulation constraints are:

[0065]

[0066]

[0067] The energy storage regulation constraint is:

[0068]

[0069] SOC e,T,s =SOC e,0,s

[0070] The photovoltaic regulation constraint is:

[0071]

[0072] in, It represents the predicted output value of photovoltaic p at time t in scenario s.

[0073] The electric vehicle regulation constraints are:

[0074]

[0075] The flexible load adjustment constraint is:

[0076]

[0077] In step 3, the GUROBI solver is used to solve the virtual power plant electric energy dispatch model to obtain the dispatch strategy of the virtual power plant in the electric energy market, including: the amount of electricity purchased and sold by the virtual power plant, the power generation and consumption plan of the internal members of the virtual power plant, and the market revenue of the virtual power plant.

[0078] In step 4, the speed pause particle swarm algorithm is used to globally optimize the virtual power plant regulation capacity boundary characterization model to obtain the virtual power plant adjustable capacity boundary during the regulation period, including the upper and lower limits of the virtual power plant output during the regulation period.

[0079] The present invention has the following beneficial effects:

[0080] The present invention constructs a virtual power plant regulation space boundary characterization model that considers conditional risk value, characterizes the single-time adjustable capacity boundary of the virtual power plant under different risk aversion coefficients, and then constructs the adjustable capacity boundary of the virtual power plant. The virtual power plant operator can declare according to the constructed adjustable capacity boundary to provide regulatory services such as peak shaving and valley filling for the power system, and makes full use of the internal coordination mechanism of the virtual power plant. It is of great significance in improving the flexibility of the power system, optimizing the energy structure, saving energy and reducing consumption, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 A flowchart of the method for constructing the adjustable capacity boundary of a virtual power plant according to the present invention;

[0082] Figure 2 This is a schematic diagram of the uncertainty scenario of distributed photovoltaics;

[0083] Figure 3 is a schematic diagram of the flexible load demand curve;

[0084] Figure 4 A schematic diagram of the demand curve for charging electric vehicles;

[0085] Figure 5 This is a schematic diagram of time-of-use electricity prices in the electricity energy market;

[0086] Figure 6 A schematic diagram of the power purchase and sales strategy for virtual power plants in the electric energy market;

[0087] Figure 7 Schematic diagram of the results of constructing the adjustable capacity boundary of the virtual power plant considering the conditional value at risk. DETAILED DESCRIPTION

[0088] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0089] Example

[0090] The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value in this embodiment is as follows: Figure 1 As shown, the following steps are included:

[0091] Step 1: In the electric energy dispatching stage, the objective function is to minimize the purchase cost of electricity in the virtual power plant spot market, and to construct an electric energy dispatching model of the virtual power plant guided by time-of-use electricity prices;

[0092] Step 1.1: Construct a virtual power plant energy dispatch model with the goal of minimizing the virtual power plant spot market electricity purchase cost. The specific construction method is as follows:

[0093] Step 1.1.1: Determine the objective function of the virtual power plant energy scheduling model as:

[0094]

[0095] in, represents the electricity cost of distributed photovoltaic p at time t; represents the electricity cost of gas turbine m at time t; represents the electricity cost of energy storage e at time t; represents the electricity cost of the flexible load h at time t; represents the electricity cost of electric vehicle v at time t; Represents the revenue from purchasing and selling electricity of the virtual power plant at time t.

[0096] The cost function of distributed photovoltaic energy consumption aggregated by the virtual power plant is:

[0097]

[0098] Among them, c pv The grid-connected electricity price of distributed photovoltaic power generation is RMB 391 / MWh; represents the output of distributed photovoltaic p at time t; λ pv Indicates the unit abandonment penalty cost; Represents the predicted value of distributed photovoltaic p output.

[0099] The energy cost function of the gas turbine aggregated by the virtual power plant is:

[0100]

[0101] in, represents the g-th stage cost operation state variable of gas turbine m at time t; l m,g , b m,g represents the slope and intercept of the g-th cost operation of gas turbine m at time t; Represents the output of gas turbine m at time t.

[0102] The energy cost function of energy storage aggregated by the virtual power plant is:

[0103]

[0104] in, Indicates the unit energy cost of energy storage; Represents the charging and discharging power of energy storage e at time t.

[0105] The energy cost function of the flexible load aggregated by the virtual power plant is:

[0106]

[0107] Among them, TEP t represents the time-of-use electricity price at time t; represents the actual power consumption of the flexible load h at time t; represents the subsidy price of flexible load response to virtual power plant dispatch at time t; It represents the power of flexible load h in response to virtual power plant dispatch at time t.

[0108] The energy cost function of electric vehicles aggregated by the virtual power plant is:

[0109]

[0110] in, represents the unit energy cost of electric vehicle v at time t, and takes represents the actual power consumption of electric vehicle v at time t; represents the subsidized electricity price of electric vehicle v in response to the virtual power plant dispatch at time t; It represents the power of flexible load v in response to virtual power plant dispatch at time t.

[0111] The revenue function of the virtual power plant energy market is:

[0112]

[0113] in, represents the purchase and sale price of electricity of the virtual power plant, Represents the amount of electricity purchased and sold by the virtual power plant at time t.

[0114] Step 1.1.2: Determine the constraints of the virtual power plant energy dispatch model, which include: power balance constraints, gas turbine operation constraints, energy storage constraints, photovoltaic output constraints, electric vehicle constraints, and flexible load constraints, as follows:

[0115] (1) The power balance constraints are as follows:

[0116]

[0117] (2) Gas turbine operating constraints are as follows:

[0118]

[0119] In the formula, It represents the upper and lower limits of the output of the gas turbine unit m of the virtual power plant; It represents the upper limit of the ramp rate of the gas turbine unit m in the virtual power plant;

[0120] (3) Energy storage constraints are as follows:

[0121]

[0122] SOC e,T =SOC e,0

[0123] In the formula, is a Boolean variable, indicating the charge and discharge status of the energy storage e; Indicates the maximum charge and discharge power of energy storage e; SOC e,t Indicates the amount of energy stored e at time t; SOC e,min , SOC e,max They are the maximum and minimum power allowed by energy storage e; SOC e,0 is the initial capacity of energy storage e, set to half of the maximum capacity of energy storage; η e is the charging and discharging efficiency of energy storage e;

[0124] (4) The photovoltaic output constraints are as follows:

[0125]

[0126] (5) The flexible load constraints are as follows:

[0127]

[0128] In the formula, represents the predicted value of the electric energy demand of the flexible load h at time t; It represents the cumulative amount of electric energy demand of flexible load h that is not met at time t.

[0129] (6) Electric vehicle restrictions are as follows:

[0130]

[0131] In the formula, represents the predicted value of the charging demand of electric vehicle v at time t; It represents the cumulative amount of electric energy demand of flexible load v that is not met at time t.

[0132] Step 2: Taking the maximum and minimum output of the virtual power plant during the regulation period as the objective function, and taking the conditional risk value constraint including the regulation power compensation price and the regulation cost opportunity constraint including the loss risk aversion coefficient as the constraint conditions, a virtual power plant regulation capacity boundary characterization model is constructed; the maximum and minimum output of the virtual power plant during the regulation period correspond to the upper and lower limits of the virtual power plant regulation capacity boundary respectively;

[0133] Step 2.1: Construct a virtual power plant regulation space boundary characterization model, with the maximum or minimum output during the regulation period as the goal, and achieve it through the virtual power plant power energy scheduling strategy and the sum of the regulation power. The specific construction method is as follows:

[0134] Step 2.1.1: Determine the objective function of the virtual power plant regulation space boundary characterization model:

[0135]

[0136] in, It means that after the virtual power plant provides regulation services, c Load demand at any given moment; Indicates the power purchase and sale adjustment power of the virtual power plant.

[0137] Step 2.1.2: Determine the constraints of the virtual power plant regulation space boundary characterization model, which include: adjusted power balance constraint, gas turbine regulation constraint, energy storage regulation constraint, photovoltaic output regulation constraint, electric vehicle regulation constraint, flexible load regulation constraint, conditional risk value constraint, and regulation cost opportunity constraint, as follows:

[0138] (1) The power balance constraints after adjustment are as follows:

[0139]

[0140] in, It represents the regulation power of photovoltaic, gas turbine, flexible load, energy storage charging, energy storage discharging, and electric vehicle charging scheduling strategy under the scenario s at time t.

[0141] (2) The gas turbine regulation constraints are as follows:

[0142]

[0143] (3) Energy storage regulation constraints:

[0144]

[0145] SOC e,T,s =SOC e,0,s

[0146] (4) The photovoltaic regulation constraints are as follows:

[0147]

[0148] in, It represents the predicted output value of photovoltaic p at time t in scenario s.

[0149] (5) Electric vehicle regulation constraints:

[0150]

[0151] (6) Flexible load adjustment constraints:

[0152]

[0153] (7) Conditional Value at Risk Constraints:

[0154]

[0155] v(s)≥0

[0156] Among them, C control (s) represents the comprehensive cost of the virtual power plant providing regulation services under the scenario of s renewable energy output; ε pv , represents the adjustment power subsidy of the virtual power plant to photovoltaic, electric vehicles and flexible loads; α represents the adjustment power compensation price in the regulatory market; represents the probability of occurrence of the new energy output scenario; L VaR,τ , L CVaR,τ represents the risk value and conditional risk value under the confidence level τ; ζ and v(s) are auxiliary variables for calculating the risk value and conditional risk value.

[0157] (8) Adjustment cost opportunity constraints:

[0158]

[0159] Among them, γ represents the virtual power plant's aversion to loss risk. The larger γ is, the more the power sales company averses loss risk.

[0160] Step 3: Construct a test case of the virtual power plant energy dispatch model suitable for step 1, use the GUROBI solver to solve the virtual power plant energy dispatch model, and obtain the dispatch strategy of the virtual power plant in the energy market, including: the amount of electricity purchased and sold by the virtual power plant, the power generation and consumption plans of the internal members of the virtual power plant, and the market revenue of the virtual power plant.

[0161] Step 4: Based on the virtual power plant electricity market dispatch strategy obtained in step 3, combined with the adjustment electricity subsidy price, the speed pause particle swarm algorithm is used to globally optimize the virtual power plant adjustment capacity boundary characterization model constructed in step 2 to obtain the virtual power plant adjustable capacity boundary during the adjustment period, including the upper and lower limits of the virtual power plant's output during the adjustment period and the adjustment plan for subsequent periods.

[0162] Application Examples

[0163] The virtual power plant constructed in this example is composed of distributed photovoltaics, distributed gas turbines, distributed energy storage, flexible loads, and electric vehicles. The parameters of the virtual power plant members are shown in Table 1. The uncertainty scenario of distributed photovoltaics is shown in Figure 2 , the flexible load and electric vehicle charging demand curve is shown in Figure 3 , Figure 4 Virtual power plants obtain revenue by participating in the purchase and sale of electricity in the energy market and providing services such as peak shaving and valley filling in the regulation market. Virtual power plants purchase and sell electricity in the energy market based on the time-of-use electricity price. Figure 5 The amount of electricity that the virtual power plant provides regulation services in the regulation market will be subsidized at 5,500 yuan / MWh.

[0164] Table 1 Virtual power plant member parameters

[0165]

[0166] This embodiment is based on the MATLAB platform in the win11 operating system, R9CPU, 4.50GHz processor environment to carry out simulation and optimization analysis, the details are as follows:

[0167] According to the method of the above embodiment, a virtual power plant electric energy dispatching model and a virtual power plant regulating capacity boundary characterization model are constructed, and the virtual power plant electric energy dispatching model is solved. The obtained virtual power plant power purchase and sale strategy in the electric energy market is as follows: Figure 6 As shown in the figure, the construction result of the adjustable capacity boundary of the virtual power plant considering the conditional risk value when facing the regulation market is as follows Figure 7 As shown, Figure 7 The adjustable space in is composed of the upper and lower limits of the output of the virtual power plant at the fifth moment. Figure 7 It can be seen that the virtual power plant maximizes the adjustable space of the virtual power plant at the fifth moment by adjusting the subsequent output plan; after considering the conditional risk value, in order to reduce the risk cost brought by photovoltaic uncertainty to a certain extent, the adjustment space that the virtual power plant can declare is also reduced. Therefore, the virtual power plant operator can use this patent to select different adjustable capacity declaration strategies and adjust the subsequent output plan based on its own aversion to uncertainty risks, electricity energy trading prices and adjustment power compensation prices.

[0168] According to the theoretically derived formula and simulation results, the following conclusions can be drawn:

[0169] The present invention considers the process of virtual power plants formulating power purchase and sales strategies in the electric energy market and adjusting their power purchase and sales strategies in the process of adjustable capacity declaration, thereby considering the guidance of time-of-use electricity prices on the output plan of virtual power plants, and obtaining the initial scheduling plan of virtual power plants, including the output plan of virtual power plants and the output plan of their aggregated resources;

[0170] This patent introduces the conditional risk value theory to deal with the risk of uncertainty in new energy output, considers the conditional risk value constraints including the adjustment of power compensation prices, and proposes a model for constructing the adjustable capacity boundary of a virtual power plant that considers conditional risk value and cost guidance. By solving the upper and lower limits of the acceptable output of the virtual power plant at a certain moment, the adjustable space of the virtual power plant is constructed. According to this model, virtual power plant operators can choose different adjustable capacity declaration strategies and adjust subsequent output plans based on their own aversion to uncertainty risks and the intensity of market compensation, so as to maximize the benefits of virtual power plants in multi-type electricity markets.

[0171] The technical solutions of the present invention are not limited to the above-mentioned embodiments, and any technical solutions obtained by equivalent replacement methods fall within the scope of protection required by the present invention.

Claims

1. A method for constructing an adjustable capacity boundary of a virtual power plant considering conditional risk value, characterized in that: The steps include: Step 1: In the electric energy dispatching stage, the objective function is to minimize the purchase cost of electricity in the virtual power plant spot market, and to construct an electric energy dispatching model of the virtual power plant guided by time-of-use electricity prices; Step 2: Taking the maximum and minimum output of the virtual power plant during the regulation period as the objective function, and taking the conditional risk value constraint including the regulation power compensation price and the regulation cost opportunity constraint including the loss risk aversion coefficient as the constraint conditions, a virtual power plant regulation capacity boundary characterization model is constructed; the maximum and minimum output of the virtual power plant during the regulation period correspond to the upper and lower limits of the virtual power plant regulation capacity boundary respectively; Step 3: Solve the electric energy dispatch model of the virtual power plant to obtain the dispatch strategy of the virtual power plant in the electric energy market; Step 4: According to the dispatch strategy of the virtual power plant in the electric energy market, the virtual power plant regulation space capacity characterization model is globally optimized to obtain the adjustable capacity boundary of the virtual power plant during the regulation period.

2. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 1, characterized in that: The objective function of the virtual power plant electric energy scheduling model is: in, represents the electricity cost of distributed photovoltaic p at time t; represents the electricity cost of gas turbine m at time t; represents the electricity cost of energy storage e at time t; represents the electricity cost of the flexible load h at time t; represents the electricity cost of electric vehicle v at time t; represents the revenue from electricity purchase and sale of the virtual power plant at time t; The electricity cost of distributed photovoltaic is: Among them, c pv The grid-connected electricity price of distributed photovoltaic power generation is RMB 391 / MWh; represents the output of distributed photovoltaic p at time t; λ pv Indicates the unit abandonment penalty cost; It represents the predicted value of distributed photovoltaic p output; The cost of electricity from a gas turbine is: in, represents the g-th stage cost operation state variable of gas turbine m at time t; l m,g , b m,g represents the slope and intercept of the g-th cost operation of gas turbine m at time t; represents the output of gas turbine m at time t; The electricity cost of energy storage is: in, Indicates the unit energy cost of energy storage; represents the charging and discharging power of energy storage e at time t; The electricity cost of flexible load is: Among them, TEP t represents the time-of-use electricity price at time t; represents the actual power consumption of the flexible load h at time t; represents the subsidy price of flexible load response to virtual power plant dispatch at time t; represents the power of the flexible load h in response to the virtual power plant dispatch at time t; The electricity cost of electric vehicles is: in, represents the unit energy cost of electric vehicle v at time t, and takes represents the actual power consumption of electric vehicle v at time t; represents the subsidized electricity price of electric vehicle v in response to the virtual power plant dispatch at time t; represents the power of the flexible load v in response to the virtual power plant dispatch at time t; The revenue from purchasing and selling electricity of a virtual power plant is: in, represents the purchase and sale price of electricity of the virtual power plant, Represents the amount of electricity purchased and sold by the virtual power plant at time t.

3. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 2 is characterized in that: The constraints of the virtual power plant electric energy dispatch model include power balance constraints, gas turbine operation constraints, energy storage constraints, photovoltaic output constraints, electric vehicle constraints and flexible load constraints.

4. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 3 is characterized by: The power balance constraints are as follows: The gas turbine operating constraints are as follows: In the formula, It represents the upper and lower limits of the output of the gas turbine unit m of the virtual power plant; It represents the upper limit of the ramp rate of the gas turbine unit m in the virtual power plant; The energy storage constraints are as follows: SOCIETY e,T =SOC e,0 In the formula, is a Boolean variable, indicating the charge and discharge status of the energy storage e; Indicates the maximum charge and discharge power of energy storage e; SOC e,t Indicates the amount of energy stored e at time t; SOC e,min , SOC e,max They are the maximum and minimum power allowed by energy storage e; SOC e,0 is the initial capacity of energy storage e, set to half of the maximum capacity of energy storage; η e is the charging and discharging efficiency of energy storage e; The photovoltaic output constraints are as follows: The flexible load constraints are as follows: In the formula, represents the predicted value of the electric energy demand of the flexible load h at time t; It represents the cumulative amount of electric energy demand of flexible load h that is not met at time t; The electric vehicle constraints are as follows: In the formula, represents the predicted value of the charging demand of electric vehicle v at time t; It represents the cumulative amount of electric energy demand of flexible load v that is not met at time t.

5. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 4, characterized in that: The objective function of the virtual power plant regulation capacity boundary characterization model is as follows: in, It means that after the virtual power plant provides regulation services, c Load demand at any given moment; Indicates the power purchase and sale adjustment power of the virtual power plant.

6. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 5 is characterized in that: The conditional VaR constraints are: v(s)≥0 Among them, C control (s) represents the comprehensive cost of the virtual power plant providing regulation services under the scenario of s renewable energy output; ε pv , represents the adjustment power subsidy of the virtual power plant to photovoltaic, electric vehicles and flexible loads; α represents the adjustment power compensation price in the regulatory market; represents the probability of occurrence of the new energy output scenario; L VaR,τ , L CVaR,τ represents the risk value and conditional risk value under the confidence level τ; ζ, v(s) are auxiliary variables for calculating risk value and conditional risk value; The adjustment cost opportunity constraint is: Among them, γ represents the virtual power plant's aversion to loss risk. The larger γ is, the more the power sales company averses loss risk.

7. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 6 is characterized in that: The constraint conditions of the virtual power plant regulation capacity boundary characterization model also include power balance constraints after regulation, gas turbine regulation constraints, energy storage regulation constraints, photovoltaic output regulation constraints, electric vehicle regulation constraints, and flexible load regulation constraints.

8. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 7, characterized in that: The adjusted power balance constraint is as follows: in, It represents the regulation power of photovoltaic, gas turbine, flexible load, energy storage charging, energy storage discharging, and electric vehicle charging scheduling strategy under the scenario s at time t; The gas turbine regulation constraints are: The energy storage regulation constraint is: SOCIETY e,T,s =SOC e,0,s The photovoltaic regulation constraint is: in, It represents the predicted output value of photovoltaic p at time t in scenario s. The electric vehicle regulation constraints are: The flexible load adjustment constraint is:

9. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 1, characterized in that: In step 3, the GUROBI solver is used to solve the virtual power plant electric energy dispatch model to obtain the dispatch strategy of the virtual power plant in the electric energy market, including: the amount of electricity purchased and sold by the virtual power plant, the power generation and consumption plan of the internal members of the virtual power plant, and the market revenue of the virtual power plant.

10. The method for constructing the adjustable capacity boundary of a virtual power plant considering conditional risk value according to claim 1, characterized in that: In step 4, the speed pause particle swarm algorithm is used to globally optimize the virtual power plant regulation capacity boundary characterization model to obtain the virtual power plant adjustable capacity boundary during the regulation period, including the upper and lower limits of the virtual power plant output during the regulation period.

Images