A dual game method and system for multiple virtual power plants and power distribution systems

Abstract

The present invention discloses a dual game method and system of multiple virtual power plants and distribution systems, the method comprising: constructing a distribution system operation model and a virtual power plant operation model respectively with the goal of minimizing the operation cost; converting the distribution system operation model and the virtual power plant operation model into corresponding distributed blue-robust optimization models respectively according to the uncertainty of distributed photovoltaics; constructing a first game model between multiple virtual power plants and the distribution system with the electricity price strategy and the grid-connected power as the interactive variables; constructing a second game model between the virtual power plants with the active power transmitted between the virtual power plants as the interactive variables; constructing a dual game model of multiple virtual power plants and the distribution system based on the first game model and the second game model; solving the dual game model to obtain the optimal power generation plan and the optimal energy interaction strategy. The present invention can balance the supply and demand relationship of the power system and improve the resource allocation efficiency of the entire power system.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a dual game method and system for multiple virtual power plants and a power distribution system. Background Art

[0002] With the vigorous promotion of clean energy, a large number of distributed renewable energy sources are connected to the distribution system. These distributed renewable energy sources are characterized by dispersion, intermittency and uncertainty, which poses challenges to the stable operation of the distribution system. As a flexible energy management method, multi-virtual power plants can aggregate many distributed renewable energy sources to form virtual power plants, interact with the distribution system, determine the optimal power generation plan and trading strategy, and maximize their own interests.

[0003] The existing game methods between virtual power plants and distribution systems are mainly based on two-level planning and Stackelberg game methods. The game method based on two-level planning is that the lower virtual power plant optimizes the scheduling strategy of internal distributed energy and energy storage equipment according to the constraints given by the upper distribution system, and then feeds the results back to the upper distribution system. The upper distribution system then adjusts its decision variables according to the feedback from the lower virtual power plant, and obtains the optimal solution of the two-level planning after multiple rounds of iterative solutions. The method based on Stackelberg game is to regard the distribution system as a leader and the virtual power plant as a follower. The leader first formulates a strategy, such as formulating the rules of the power market, determining the price range for purchasing electricity from the virtual power plant, etc. The follower adjusts its own power generation plan, energy storage charging and discharging plan, etc. according to the leader's strategy to maximize its own interests.

[0004] However, existing game methods are often based on some simplified assumptions. For example, it is assumed that the distributed energy within the virtual power plant is completely predictable and controllable. But in reality, the power generation of renewable energy is highly intermittent and uncertain, and the load demand is also highly volatile. These assumptions may cause large deviations in the model in practical applications, resulting in the inability of the game results to accurately reflect the actual interaction situation, which in turn affects the optimal operation decisions of the virtual power plant and the distribution system. At present, there are mainly robust optimization methods and stochastic optimization methods for dealing with uncertainty. The robust optimization method is to find the solution in the worst case in the constructed uncertainty set, and there is often a problem of being too conservative. Stochastic optimization models the probability distribution of uncertain factors and finds the solution when the expected optimal is mainly dependent on whether an accurate probability distribution can be constructed. It may not be possible to find the global optimal solution due to the influence of randomness. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the existing game methods, the present invention provides a dual game method and system of multiple virtual power plants and distribution systems.

[0006] In a first aspect, an embodiment of the present invention provides a dual game method of multiple virtual power plants and a power distribution system, comprising:

[0007] With the goal of minimizing operating costs, the distribution system operation model and the virtual power plant operation model are constructed respectively;

[0008] Constructing a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and converting the distribution system operation model and the virtual power plant operation model into a distribution system operation distributed robust optimization model and a virtual power plant operation distributed robust optimization model respectively based on the fuzzy set;

[0009] Taking the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables, a first-level game model between multiple virtual power plants and the distribution system is constructed;

[0010] Taking the active power transmitted between the virtual power plants determined based on the electricity price strategy and the blue optimization model of the virtual power plant operation distribution as the interactive variable, a second game model between the virtual power plants is constructed;

[0011] Based on the first game model and the second game model, a dual game model of multiple virtual power plants and distribution systems is constructed;

[0012] The double game model is solved to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0013] Preferably, the step of constructing a power distribution system operation model and a virtual power plant operation model respectively with the goal of minimizing the operation cost includes:

[0014] Taking minimizing the operating cost of the distribution system as the goal, establishing an objective function, constructing an initial distribution system operation model, and using a first constraint condition to constrain the initial distribution system operation model to obtain a distribution system operation model, wherein the first constraint condition includes a first distributed photovoltaic operation constraint, a first gas turbine operation constraint, a first energy storage operation constraint, a first reactive compensation device operation constraint, a virtual power plant price setting constraint, and a power flow constraint;

[0015] With the goal of minimizing the operating cost of the virtual power plant, an objective function is established, an initial virtual power plant operating model is constructed, and the initial virtual power plant operating model is constrained by a second constraint condition to obtain a virtual power plant operating model, wherein the second constraint condition includes a second distributed photovoltaic operating constraint, a second gas turbine operating constraint, a second energy storage operating constraint, a second reactive compensation device operating constraint, energy exchange constraints between virtual power plants, and a power balance constraint.

[0016] Preferably, the objective function of the power distribution system operation model is characterized by the following formula:

[0017]

[0018] in, represents the operating cost of the power distribution system, T represents the total operating time, represents the operating cost of distributed photovoltaic power generation in the power distribution system at time t, represents the gas turbine operating cost in the power distribution system at time t, represents the energy storage operation cost in the distribution system at time t, represents the energy exchange cost between the distribution system and the virtual power plant at time t;

[0019] The following formula is used to characterize the first distributed photovoltaic constraint:

[0020]

[0021]

[0022] in, represents the active power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the reactive power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i in the distribution system at time t, represents the power generation capacity of distributed photovoltaic at node i in the distribution system at time t;

[0023] The first gas turbine operation constraint is characterized by the following formula:

[0024]

[0025]

[0026] in, represents the active power transmission of the gas turbine at node i in the power distribution system at time t, represents the reactive power transmission of the gas turbine at node i in the power distribution system at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i in the power distribution system at time t, represents the power factor of the gas turbine at node i at time t in the power distribution system;

[0027] The following formula is used to characterize the first energy storage operation constraint:

[0028]

[0029]

[0030]

[0031] in, represents the energy storage charging state at node i in the power distribution system at time t, represents the energy storage discharge state at node i in the distribution system at time t, represents the energy storage charging power at node i in the distribution system at time t, represents the energy storage discharge power at node i in the distribution system at time t, represents the lower limit of energy storage charging power at node i in the power distribution system at time t, represents the upper limit of energy storage charging power at node i in the power distribution system at time t, represents the lower limit of energy storage discharge power at node i in the distribution system at time t, represents the upper limit of energy storage discharge power at node i in the distribution system at time t;

[0032] The following formula is used to characterize the operation constraint of the first reactive power compensation device:

[0033]

[0034] in, It represents the reactive power provided by the reactive power compensation device at node i in the power distribution system at time t. represents the lower limit of reactive power provided by the reactive power compensation device at node i in the distribution system at time t, represents the upper limit of reactive power provided by the reactive power compensation device at node i in the power distribution system at time t;

[0035] The following formula is used to characterize the virtual power plant price setting constraints:

[0036]

[0037]

[0038] in, represents the active power price of the nth virtual power plant in the distribution system at time t, represents the reactive power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the upper limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of reactive power price of the nth virtual power plant in the distribution system at time t, represents the upper limit of reactive power price of the nth virtual power plant in the distribution system at time t;

[0039] The power flow constraint is characterized by the following formula:

[0040]

[0041]

[0042]

[0043] in, represents the active node injection amount at node j in the distribution system at time t, represents the reactive node injection amount at node j in the distribution system at time t, represents the active node injection amount at node i in the distribution system at time t, represents the reactive node injection amount at node i in the distribution system at time t, It indicates the active power at the head end of line jk of the power distribution system at time t. It represents the reactive power of the first end of line jk at time t in the power distribution system. It represents the active power at the head end of line ij of the power distribution system at time t. It represents the reactive power of the line ij at the time t of the power distribution system. represents the set of nodes in the power distribution system, represents the active load at node i in the distribution system at time t, represents the reactive load at node i in the distribution system at time t, represents the active amount of interaction between the distribution system and the virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant, represents the resistance of line ij, represents the line ij reactance, represents the voltage at node i in the power distribution system at time t, represents the lower limit of voltage at node i in the power distribution system at time t, represents the upper voltage limit at node i in the power distribution system at time t, Indicates the upper limit of the current that line ij can withstand. Represents the current on line ij at time t in the distribution system.

[0044] Preferably, the objective function of the virtual power plant operation model is characterized by the following formula:

[0045]

[0046] in, represents the operating cost of the nth virtual power plant, represents the operating cost of distributed photovoltaic power generation in the nth virtual power plant at time t, represents the operating cost of the gas turbine in the nth virtual power plant at time t, represents the energy storage operation cost in the nth virtual power plant at time t, represents the energy exchange cost between the nth virtual power plant and the distribution system at time t, represents the energy interaction cost between the nth virtual power plant and other virtual power plants at time t;

[0047] The following formula is used to characterize the second distributed photovoltaic constraint:

[0048]

[0049]

[0050] in, represents the active power transmission of distributed photovoltaic at node i of the virtual power plant at time t, represents the reactive power transmission of distributed photovoltaic at node i of the virtual power plant at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i of the virtual power plant at time t, represents the power generation capacity of distributed photovoltaic at node i of the virtual power plant at time t;

[0051] The second gas turbine operation constraint is characterized by the following formula:

[0052]

[0053]

[0054] in, represents the active power transmission of the gas turbine at node i in the virtual power plant at time t, represents the reactive power transmission of the gas turbine at node i of the virtual power plant at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i of the virtual power plant at time t, represents the power factor of the gas turbine at node i at time t in the virtual power plant;

[0055] The following formula is used to characterize the second energy storage operation constraint:

[0056]

[0057]

[0058]

[0059] in, represents the energy storage charging state at node i of the virtual power plant at time t, represents the energy storage discharge state at node i of the virtual power plant at time t, represents the energy storage charging power at node i of the virtual power plant at time t, represents the energy storage discharge power at node i of the virtual power plant at time t, represents the lower limit of energy storage charging power at node i of the virtual power plant at time t, represents the upper limit of energy storage charging power at node i of the virtual power plant at time t, represents the lower limit of energy storage discharge power at node i of the virtual power plant at time t, represents the upper limit of energy storage discharge power at node i of the virtual power plant at time t;

[0060] The following formula is used to characterize the operation constraint of the second reactive power compensation device:

[0061]

[0062] in, represents the reactive power provided by the reactive power compensation device at node i of the virtual power plant at time t, represents the lower limit of reactive power provided by the reactive compensation device at node i of the virtual power plant at time t, represents the reactive power upper limit provided by the reactive power compensation device at node i of the virtual power plant at time t;

[0063] The following formula is used to characterize the energy exchange constraints between the virtual power plants:

[0064]

[0065]

[0066] in, represents the energy transmitted from the mth virtual power plant to the nth virtual power plant at time t, represents the energy transmitted from the nth virtual power plant to the mth virtual power plant at time t, represents the active load of the nth virtual power plant at time t;

[0067] The power balance constraint is characterized by the following formula:

[0068]

[0069]

[0070] in, represents the reactive load of the nth virtual power plant at time t, represents the active amount of interaction between the distribution system and the virtual power plant, represents other virtual power plants except the nth virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant.

[0071] Preferably, the step of constructing a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and converting the distribution system operation model and the virtual power plant operation model into a distribution system operation distributed robust optimization model and a virtual power plant operation distributed robust optimization model respectively based on the fuzzy set, comprising:

[0072] Performing scene clustering on distributed photovoltaic historical data to obtain multiple characteristic scenes and initial probability distribution of each characteristic scene, wherein the distributed photovoltaic historical data includes distributed photovoltaic output historical data and distributed photovoltaic load historical data;

[0073] Based on the constraint of KL divergence on the error between the initial probability distribution and the true probability distribution of each of the feature scenes, a probability distribution uncertainty set is constructed, and the probability distribution uncertainty set is characterized as a fuzzy set;

[0074] Based on the fuzzy set and the two-stage decision variable setting rule, the distribution system operation model is converted into a distribution system operation distributed robust optimization model, wherein the two-stage decision variable setting rule is to set the discrete variables as the first-stage decision variables and the non-discrete variables as the second-stage decision variables;

[0075] Based on the fuzzy sets and the two-stage decision variable setting rules, the virtual power plant operation model is converted into a virtual power plant operation distribution robust optimization model.

[0076] Preferably, the following formula is used to characterize the distribution system operation distribution robust optimization model or the virtual power plant operation distribution robust optimization model:

[0077]

[0078]

[0079]

[0080] Among them, s represents the sequence number of the feature scene, represents the total number of feature scenes, represents the true probability distribution of feature scene s, represents a fuzzy set, represents the first-stage decision variable, represents the second-stage decision variable, represents the set of decision variable constraints in the first stage, represents the set of decision variable constraints in the second stage, , , , , , , , , , , , , , Represents the constant matrix corresponding to the decision variable constraints.

[0081] Preferably, the first game model between multiple virtual power plants and the distribution system is constructed by taking the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables, including:

[0082] Solving the blue stick optimization model for the operation of the virtual power plant, and transferring the solved grid-connected power to the distribution system;

[0083] Solving the distributed robust optimization model of the power distribution system based on the grid-connected power, and transmitting the solved electricity price strategy to the virtual power plant;

[0084] The electricity price strategy and the grid-connected power are used as interactive variables to iterate and interact until a Nash equilibrium is reached, thereby obtaining a first-level game model between multiple virtual power plants and a distribution system.

[0085] Preferably, the active power transmitted between the virtual power plants determined based on the electricity price strategy and the virtual power plant operation distribution robust optimization model is used as an interactive variable to construct a second game model between the virtual power plants, including:

[0086] Solving the blue-robust optimization model of the virtual power plant operation distribution based on the electricity price strategy to obtain the active power of each virtual power plant;

[0087] The active power transmitted between the virtual power plants is used as an interactive variable for iterative interaction until a Nash equilibrium is reached, thereby obtaining a second-level game model between the virtual power plants.

[0088] Preferably, solving the dual game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system includes:

[0089] The double game model is solved by a solver to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0090] In a second aspect, an embodiment of the present invention provides a dual game system of multiple virtual power plants and a power distribution system, including:

[0091] An operation model building module is used to build a distribution system operation model and a virtual power plant operation model respectively with the goal of minimizing the operation cost;

[0092] A distributed robust optimization model construction module is used to construct a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and based on the fuzzy set, the distribution system operation model and the virtual power plant operation model are converted into a distributed robust optimization model for distribution system operation and a distributed robust optimization model for virtual power plant operation respectively;

[0093] A first game model construction module is used to construct a first game model between multiple virtual power plants and the distribution system, using the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables;

[0094] A second game model construction module, used to construct a second game model between the virtual power plants, taking the active power transmitted between the virtual power plants determined based on the electricity price strategy and the virtual power plant operation distribution blue optimization model as an interactive variable;

[0095] A dual game model construction module, used to construct a dual game model of multiple virtual power plants and distribution systems based on the first game model and the second game model;

[0096] The output result determination module is used to solve the double game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0097] Compared with the prior art, the dual game method and system of multiple virtual power plants and distribution systems in the embodiment of the present invention have the following beneficial effects: based on the KL divergence, the uncertainty of distributed photovoltaic output and load is taken into account, and the conservative control of distributed blue-robust optimization is achieved by adjusting the KL divergence, thereby improving the applicability of the dual game method of multiple virtual power plants and distribution systems; a dual game framework of multiple virtual power plants and distribution systems is constructed by an interactive game strategy between the distribution system and multiple virtual power plants established with electricity price strategy and grid-connected power as interactive variables and a cooperative game strategy between multiple virtual power plants established with active power as interactive variables, which helps to better integrate distributed energy into the power system and improve the resource allocation efficiency of the entire power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0098] Figure 1 It is a flow chart of a dual game method of multiple virtual power plants and power distribution systems according to an embodiment of the present invention;

[0099] Figure 2 It is a schematic diagram of a process of constructing an operation model according to an embodiment of the present invention;

[0100] Figure 3 It is a schematic diagram of a process of constructing a distributed blue bar optimization model according to an embodiment of the present invention;

[0101] Figure 4 It is a schematic diagram of the process of constructing a first-level game model according to an embodiment of the present invention;

[0102] Figure 5 It is a schematic diagram of the framework of the first game model of an embodiment of the present invention;

[0103] Figure 6 It is a schematic diagram of the process of constructing the second game model according to an embodiment of the present invention;

[0104] Figure 7 It is a schematic diagram of the framework of the second game model of an embodiment of the present invention;

[0105] Figure 8 It is a schematic diagram of the framework of the double game model of the embodiment of the present invention;

[0106] Fig. 9 1. is a schematic diagram of a topology of a simulation system of an example of an embodiment of the present invention;

[0107] Fig.10 It is a structural schematic diagram of a dual game system of multiple virtual power plants and power distribution systems according to an embodiment of the present invention;

[0108] in, Fig. 9 In the figure, VPP1 is the first virtual power plant; VPP2 is the second virtual power plant; VPP3 is the third virtual power plant; 1-7, 15-18, 22-25, 33-34, 40-41, 49 are load nodes in the distribution system; 8-14, 19-21 are load nodes of the first virtual power plant; 26-32, 35-39 are load nodes of the second virtual power plant; 42-48, 50-51 are load nodes of the third virtual power plant. DETAILED DESCRIPTION

[0109] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0110] In the description of the present invention, it should be understood that the terms "first" and "second" etc. are used in the present invention to distinguish different objects rather than to describe a specific order.

[0111] like Figure 1 As shown, an embodiment of the present invention provides a dual game method of multiple virtual power plants and distribution systems, comprising the steps of:

[0112] S1. With the goal of minimizing operating costs, the distribution system operation model and the virtual power plant operation model are constructed respectively;

[0113] Specifically, Figure 2 As shown, step S1 includes:

[0114] S101, with the goal of minimizing the operating cost of the distribution system, establishing an objective function, constructing an initial distribution system operating model, and using a first constraint condition to constrain the initial distribution system operating model to obtain a distribution system operating model;

[0115] The present invention considers the distributed photovoltaic operation cost, the gas turbine operation cost, the energy storage operation cost and the energy exchange cost to establish the objective function of the distribution system operation model.

[0116] Specifically, the objective function of the distribution system operation model is characterized by the following formula:

[0117]

[0118] in, represents the operating cost of the power distribution system, T represents the total operating time, represents the operating cost of distributed photovoltaic power generation in the power distribution system at time t, represents the gas turbine operating cost in the power distribution system at time t, represents the energy storage operation cost in the distribution system at time t, Represents the energy exchange cost between the distribution system and the virtual power plant at time t.

[0119] Further, there are:

[0120]

[0121]

[0122]

[0123]

[0124] in, represents the distributed photovoltaic cost coefficient, Represents the set of distributed photovoltaic nodes in the power distribution system, represents the active power generation of distributed photovoltaic at node i in the distribution system at time t, and represents the gas turbine cost factor, represents the set of gas turbine nodes in the power distribution system, represents the active power generation of the gas turbine at node i in the power distribution system at time t, represents the energy storage cost coefficient, represents the set of energy storage nodes in the power distribution system, represents the charge amount of energy storage at node i at time t in the power distribution system, represents the charging rate of the energy storage, represents the discharge amount of energy storage at node i in the power distribution system at time t, represents the discharge rate of stored energy, represents the set of virtual power plants, represents the active transaction cost coefficient of the distribution system and virtual power plant, represents the active amount of interaction between the distribution system and the virtual power plant, represents the reactive power transaction cost coefficient of the distribution system and virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant.

[0125] Specifically, the first constraint condition includes a first distributed photovoltaic operation constraint, a first gas turbine operation constraint, a first energy storage operation constraint, a first reactive compensation device operation constraint, a virtual power plant price setting constraint and a flow constraint.

[0126] The following formula is used to characterize the first distributed photovoltaic constraint:

[0127]

[0128]

[0129] in, represents the active power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the reactive power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i in the distribution system at time t, Represents the power generation capacity of distributed photovoltaic at node i in the distribution system at time t.

[0130] The following formula is used to characterize the first gas turbine operation constraint:

[0131]

[0132]

[0133] in, represents the active power transmission of the gas turbine at node i in the power distribution system at time t, represents the reactive power transmission of the gas turbine at node i in the power distribution system at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i in the power distribution system at time t, Represents the power factor of the gas turbine at node i in the distribution system at time t.

[0134] The following formula is used to characterize the first energy storage operation constraint:

[0135]

[0136]

[0137]

[0138] in, represents the energy storage charging state at node i in the power distribution system at time t, represents the energy storage discharge state at node i in the distribution system at time t, represents the energy storage charging power at node i in the distribution system at time t, represents the energy storage discharge power at node i in the distribution system at time t, represents the lower limit of the energy storage charging power at node i in the distribution system at time t, represents the upper limit of energy storage charging power at node i in the power distribution system at time t, represents the lower limit of energy storage discharge power at node i in the distribution system at time t, It represents the upper limit of energy storage discharge power at node i in the distribution system at time t.

[0139] The following formula is used to characterize the operation constraints of the first reactive power compensation device:

[0140]

[0141] in, It represents the reactive power provided by the reactive power compensation device at node i in the power distribution system at time t. represents the lower limit of reactive power provided by the reactive power compensation device at node i in the distribution system at time t, It represents the upper limit of reactive power provided by the reactive power compensation device at node i in the distribution system at time t.

[0142] The virtual power plant price setting constraints are described by the following formula:

[0143]

[0144]

[0145] in, represents the active power price of the nth virtual power plant in the distribution system at time t, represents the reactive power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the upper limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of reactive power price of the nth virtual power plant in the distribution system at time t, Represents the upper limit of reactive power price of the nth virtual power plant in the distribution system at time t.

[0146] The following formula is used to represent the power flow constraint:

[0147]

[0148]

[0149]

[0150] in, represents the active node injection amount at node j in the distribution system at time t, represents the reactive node injection amount at node j in the distribution system at time t, represents the active node injection amount at node i in the distribution system at time t, represents the reactive node injection amount at node i in the distribution system at time t, It indicates the active power at the head end of line jk of the power distribution system at time t. It represents the reactive power of the first end of line jk at time t in the power distribution system. It represents the active power at the head end of line ij of the power distribution system at time t. It represents the reactive power of the line ij at the time t of the power distribution system. represents the set of nodes in the power distribution system, represents the active load at node i in the distribution system at time t, represents the reactive load at node i in the distribution system at time t, represents the active amount of interaction between the distribution system and the virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant, represents the resistance of line ij, represents the line ij reactance, represents the voltage at node i in the power distribution system at time t, represents the lower limit of voltage at node i in the power distribution system at time t, represents the upper voltage limit at node i in the power distribution system at time t, Indicates the upper limit of the current that line ij can withstand. Represents the current on line ij at time t in the distribution system.

[0151] S102. With the goal of minimizing the operating cost of the virtual power plant, an objective function is established, an initial virtual power plant operating model is constructed, and the initial virtual power plant operating model is constrained by a second constraint condition to obtain a virtual power plant operating model.

[0152] The present invention considers the operating cost of distributed photovoltaics, the operating cost of gas turbines, the operating cost of energy storage, the energy exchange cost between virtual power plants and distribution systems, and the energy interaction cost between virtual power plants to establish the objective function of the virtual power plant operation model.

[0153] Specifically, the objective function of the virtual power plant operation model is characterized by the following formula:

[0154]

[0155] in, represents the operating cost of the nth virtual power plant, represents the operating cost of distributed photovoltaic power generation in the nth virtual power plant at time t, represents the operating cost of the gas turbine in the nth virtual power plant at time t, represents the energy storage operation cost in the nth virtual power plant at time t, represents the energy exchange cost between the nth virtual power plant and the distribution system at time t, Represents the energy interaction cost between the nth virtual power plant and other virtual power plants at time t.

[0156] Further, there are:

[0157]

[0158]

[0159]

[0160]

[0161]

[0162] in, represents the set of distributed photovoltaic nodes in the nth virtual power plant, represents the active power generation of distributed photovoltaics at node i at time t in the virtual power plant, represents the set of gas turbine nodes in the nth virtual power plant, represents the active power generation of the gas turbine at node i in the virtual power plant at time t, represents the set of energy storage nodes in the nth virtual power plant, represents the charge amount of energy storage at node i in the virtual power plant at time t, represents the discharge amount of energy storage at node i in the virtual power plant at time t, represents the cost coefficient of direct energy trading by the virtual power plant, represents the energy transmitted from the mth virtual power plant to the nth virtual power plant at time t, Represents other virtual power plants except the nth virtual power plant.

[0163] Specifically, the second constraint conditions include a second distributed photovoltaic operation constraint, a second gas turbine operation constraint, a second energy storage operation constraint, a second reactive compensation device operation constraint, energy exchange constraints between virtual power plants, and a power balance constraint.

[0164] The following formula is used to characterize the second distributed photovoltaic constraint:

[0165]

[0166]

[0167] in, represents the active power transmission of distributed photovoltaic at node i of the virtual power plant at time t, represents the reactive power transmission of distributed photovoltaic at node i of the virtual power plant at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i of the virtual power plant at time t, Represents the power generation capacity of distributed photovoltaic at node i of the virtual power plant at time t.

[0168] The following formula is used to characterize the operating constraints of the second gas turbine:

[0169]

[0170]

[0171] in, represents the active power transmission of the gas turbine at node i in the virtual power plant at time t, represents the reactive power transmission of the gas turbine at node i of the virtual power plant at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i of the virtual power plant at time t, Represents the power factor of the gas turbine at node i of the virtual power plant at time t.

[0172] The following formula is used to characterize the second energy storage operation constraint:

[0173]

[0174]

[0175]

[0176] in, represents the energy storage charging state at node i of the virtual power plant at time t, represents the energy storage discharge state at node i of the virtual power plant at time t, represents the energy storage charging power at node i of the virtual power plant at time t, represents the energy storage discharge power at node i of the virtual power plant at time t, represents the lower limit of energy storage charging power at node i of the virtual power plant at time t, represents the upper limit of energy storage charging power at node i of the virtual power plant at time t, represents the lower limit of energy storage discharge power at node i of the virtual power plant at time t, Represents the upper limit of energy storage discharge power at node i of the virtual power plant at time t.

[0177] The following formula is used to characterize the operation constraints of the second reactive power compensation device:

[0178]

[0179] in, represents the reactive power provided by the reactive power compensation device at node i of the virtual power plant at time t, represents the lower limit of reactive power provided by the reactive compensation device at node i of the virtual power plant at time t, It represents the upper limit of reactive power provided by the reactive compensation device at node i of the virtual power plant at time t.

[0180] The following formula is used to characterize the energy exchange constraints between virtual power plants:

[0181]

[0182]

[0183] in, represents the energy transmitted from the mth virtual power plant to the nth virtual power plant at time t, represents the energy transmitted from the nth virtual power plant to the mth virtual power plant at time t, Represents the active load of the nth virtual power plant at time t.

[0184] The power balance constraint is characterized by the following formula:

[0185]

[0186]

[0187] in, represents the reactive load of the nth virtual power plant at time t, represents the active amount of interaction between the distribution system and the virtual power plant, represents other virtual power plants except the nth virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant.

[0188] S2. According to the uncertainty of distributed photovoltaics, a fuzzy set based on KL divergence is constructed, and based on the fuzzy set, the distribution system operation model and the virtual power plant operation model are transformed into a distribution system operation distributed robust optimization model and a virtual power plant operation distributed robust optimization model respectively;

[0189] Specifically, Figure 3 As shown, step S2 includes:

[0190] S201, clustering the distributed photovoltaic historical data into scenes to obtain multiple characteristic scenes and an initial probability distribution of each characteristic scene;

[0191] The present invention takes into account the uncertainty of distributed photovoltaic output and load. Specifically, the distributed photovoltaic historical data includes distributed photovoltaic output historical data and distributed photovoltaic load historical data. The Gaussian clustering algorithm is used to cluster the distributed photovoltaic output historical data and the distributed photovoltaic load historical data to obtain multiple representative feature scenes and their initial probability distributions.

[0192] S202, based on the KL divergence constraint on the error between the initial probability distribution and the true probability distribution of each feature scene, construct a probability distribution uncertainty set, and characterize the probability distribution uncertainty set as a fuzzy set;

[0193] Since there may be a certain error between the initial probability distribution of the feature scene after scene clustering and the true probability distribution, the KL divergence is used to constrain the error between the initial probability distribution and the true probability distribution, and the probability distribution uncertainty set of the feature scene, namely the fuzzy set, is constructed:

[0194]

[0195] in, represents a fuzzy set, s represents the sequence number of the feature scene, represents the total number of feature scenes, represents the true probability distribution of feature scene s, represents the initial probability distribution of feature scene s, Represents the KL divergence value between the initial probability distribution and the true probability distribution.

[0196] S203, based on fuzzy sets and two-stage decision variable setting rules, transforming the distribution system operation model into a distribution system operation distribution robust optimization model;

[0197] On the basis of fuzzy sets, the first-stage decision variables and the second-stage decision variables are set respectively according to the two-stage decision variable setting rules, and the distribution system operation model is converted into a distributed robust optimization model in the form of a matrix. Among them, the two-stage decision variable setting rule is to set the discrete variables as the first-stage decision variables and the non-discrete variables as the second-stage decision variables. Specifically, the present invention sets the discrete type of energy storage charge and discharge state as the first-stage decision variable, and other variables as the second-stage decision variables. The second-stage decision variables include the active power generation of distributed photovoltaics, the reactive power generation of distributed photovoltaics, the active power generation of gas turbines, the reactive power generation of gas turbines, the charging amount of energy storage, the discharging amount of energy storage, the reactive compensation amount of reactive compensation devices, the active power price and reactive power price interacting with virtual power plants, the active and reactive injection amount of distribution system nodes, the active and reactive power at the head end of the line, the node voltage, and the line current.

[0198] The following formula is used to characterize the distribution system operation distribution optimization model:

[0199]

[0200]

[0201]

[0202] Among them, s represents the sequence number of the feature scene, represents the total number of feature scenes, represents the true probability distribution of feature scene s, represents a fuzzy set, represents the first-stage decision variable, represents the second-stage decision variable, represents the set of decision variable constraints in the first stage, represents the set of decision variable constraints in the second stage, , , , , , , , , , , , , , Represents the constant matrix corresponding to the decision variable constraints.

[0203] S204. Based on fuzzy sets and two-stage decision variable setting rules, the virtual power plant operation model is converted into a virtual power plant operation distribution robust optimization model.

[0204] On the basis of fuzzy sets, the first-stage decision variables and the second-stage decision variables are set respectively according to the two-stage decision variable setting rules, and the virtual power plant operation model is converted into a distributed robust optimization model in matrix form. Specifically, the present invention sets the discrete type of energy storage charging and discharging state as the first-stage decision variable, and other variables are set as the second-stage decision variables. The second-stage decision variables include the active power generation of distributed photovoltaics, the reactive power generation of distributed photovoltaics, the active power generation of gas turbines, the reactive power generation of gas turbines, the charging amount of energy storage, the discharging amount of energy storage, the reactive compensation amount of reactive compensation devices, and the active and reactive power interacting with the distribution system.

[0205] The following formula is used to characterize the robust optimization model of virtual power plant operation:

[0206]

[0207]

[0208]

[0209] Among them, s represents the sequence number of the feature scene, represents the total number of feature scenes, represents the true probability distribution of feature scene s, represents a fuzzy set, represents the first-stage decision variable, represents the second-stage decision variable, represents the set of decision variable constraints in the first stage, represents the set of decision variable constraints in the second stage, , , , , , , , , , , , , , Represents the constant matrix corresponding to the decision variable constraints.

[0210] It should be noted that the distributed robust optimization model for the distribution system operation and the distributed robust optimization model for the virtual power plant operation are characterized by the same mathematical formula. When the distributed photovoltaic uncertainty of the distribution system operation is considered, the mathematical formula characterizes the distributed robust optimization model for the distribution system operation; when the distributed photovoltaic uncertainty of the virtual power plant operation is considered, the mathematical formula characterizes the distributed robust optimization model for the virtual power plant operation. Compared with the uncertainty handling methods of robust optimization and random optimization, distributed robust optimization combines the characteristics of both and can flexibly adjust the range of the distribution set according to actual conditions, thereby achieving a balance between conservatism and optimality.

[0211] S3. Taking the electricity price strategy determined by the distributed robust optimization model of the distribution system operation and the grid-connected power determined by the distributed robust optimization model of the virtual power plant operation as interactive variables, a first-level game model between multiple virtual power plants and the distribution system is constructed;

[0212] Specifically, Figure 4 As shown, step S3 includes:

[0213] S301, solving a blue stick optimization model for the operation of a virtual power plant, and transmitting the solved grid-connected power to a distribution system;

[0214] The robust optimization model of virtual power plant operation distribution is a two-stage optimization problem, which can be well solved by using C&CG algorithm.

[0215] S302, solving a distributed robust optimization model for the distribution system operation based on the grid-connected power, and transmitting the solved electricity price strategy to the virtual power plant;

[0216] The robust optimization model of distribution system operation is a two-stage optimization problem, which can be well solved by using C&CG algorithm.

[0217] S303, iteratively interact with the electricity price strategy and the grid-connected power as interactive variables until a Nash equilibrium is reached, thereby obtaining a first-level game model between multiple virtual power plants and the distribution system.

[0218] Figure 5 The framework diagram of the first level game model is shown. The present invention adopts a cooperative game mode. The distribution system and the virtual power plant both aim to minimize their respective operating costs. The distribution system solves the optimal electricity price strategy with the lowest operating cost under the grid-connected power delivered by the virtual power plant. The virtual power plant solves the optimal grid-connected power with the lowest operating cost under the electricity price strategy delivered by the distribution system. When the game reaches the Nash equilibrium state, the optimal electricity price strategy and grid-connected power are obtained. Among them, the convergence criterion of Nash equilibrium is that the change in the interactive variable between the previous and subsequent iterations is less than the given error. Specifically, the following formula is used to characterize the convergence criterion of Nash equilibrium:

[0219]

[0220] in, represents the interactive game error, represents the number of interactive game iterations, Represents the given error of the interactive game.

[0221] With the continuous development of distributed energy, the grid power of these distributed energy sources has the characteristics of intermittency and uncertainty. The electricity price strategy can be used as a means of regulation to adapt to this characteristic. For example, through the dynamic electricity price strategy, the distributed energy in the virtual power plant is encouraged to go online at the right time. Taking the grid power and electricity price strategy as interactive variables helps to better integrate distributed energy into the power system, thereby improving the resource allocation efficiency of the entire power system.

[0222] S4. Taking the active power transmitted between virtual power plants determined by the robust optimization model based on the electricity price strategy and the virtual power plant operation distribution as the interactive variable, a second game model between virtual power plants is constructed;

[0223] Specifically, Figure 6 As shown, step S4 includes:

[0224] S401, solving a blue stick optimization model for the operation of virtual power plants based on the electricity price strategy to obtain the active power of each virtual power plant;

[0225] Based on the electricity price strategy determined by the first game model, with the output and load of distributed photovoltaics as uncertain factors, the C&CG algorithm is used to solve the distributed robust optimization model of each virtual power plant to obtain the active power of each virtual power plant.

[0226] S402, using the active power transmitted between the virtual power plants as an interactive variable to perform iterative interaction until a Nash equilibrium is reached, thereby obtaining a second-level game model between the virtual power plants.

[0227] Figure 7 The framework diagram of the second game model is shown. The present invention adopts a cooperative game approach, and each virtual power plant pursues the lowest operating cost. The interests of all parties are coordinated through the active power exchange mechanism. When the cooperative game reaches the Nash equilibrium, the strategy of each player (virtual power plant) is obtained, that is, the grid-connected power transmitted to the distribution system. Among them, the convergence criterion of Nash equilibrium is that the change in active power between the two iterations of the virtual power plant is less than the given error. Specifically, the following formula is used to characterize the convergence criterion of Nash equilibrium:

[0228]

[0229] in, represents the cooperative game error, represents the number of cooperative game iterations, represents the given error of cooperative game, Represents a collection of virtual power plants.

[0230] Active power directly reflects the actual energy transmission in the power system. In a virtual power plant, the active power output or consumption of each participant can accurately reflect its actual contribution to the energy supply and demand of the entire system. By exchanging and adjusting active power, the power balance between the virtual power plant and the power grid and between the various entities within the virtual power plant can be better achieved to ensure the stable operation of the system.

[0231] S5. Based on the first game model and the second game model, a dual game model of multiple virtual power plants and distribution systems is constructed;

[0232] Figure 8 The framework diagram of the dual game model is shown. A dual game model of multiple virtual power plants and distribution systems is constructed. The first game uses the electricity price strategy formulated by the distribution system and the grid-connected power determined by the virtual power plant as interactive variables to establish an interactive game framework between multiple virtual power plants and the distribution system. The second game uses the active power transmitted between virtual power plants as an interactive variable to establish a cooperative game framework between multiple virtual power plants.

[0233] S6. Solve the double game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0234] The dual game model is solved by using a solver to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system. This embodiment is based on the MATLAB simulation platform and uses the Yalmip toolbox to call the commercial solver Gurobi to solve the dual game model of multiple virtual power plants and distribution systems, and outputs the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0235] The embodiment of the present invention provides a dual game method for multiple virtual power plants and distribution systems. The uncertainty of distributed photovoltaic output and load is considered based on KL divergence, and conservative control of distributed blue-robust optimization is achieved by adjusting the KL divergence, thereby improving the applicability of the dual game method for multiple virtual power plants and distribution systems. A dual game framework for multiple virtual power plants and distribution systems is constructed by an interactive game strategy between the distribution system and multiple virtual power plants established with electricity price strategy and grid-connected power as interactive variables and a cooperative game strategy between multiple virtual power plants established with active power as interactive variables, which helps to better integrate distributed energy into the power system and improve the resource allocation efficiency of the entire power system.

[0236] In order to verify the effectiveness of the dual game method of multiple virtual power plants and distribution systems in the embodiment of the present invention, Fig. 9 A simulation analysis was performed on the test example shown, and the solution results are shown in Table 1.

[0237] Table 1 Comparison of single game and double game data

[0238]

[0239] By comparing the commonly used single game methods of distribution systems and virtual power plants, the dual game method of multiple virtual power plants and distribution systems proposed in the present invention reduces the distribution system operating cost by 3.45%, the virtual power plant operating cost by 7.41%, and the total power system operating cost by 5.58%, which proves the effectiveness and economy of the method proposed in the present invention.

[0240] Based on the above dual game method of multiple virtual power plants and distribution systems, Fig.10 As shown, an embodiment of the present invention provides a dual game system of multiple virtual power plants and power distribution systems, including:

[0241] Operation model building module 01, used to build a distribution system operation model and a virtual power plant operation model respectively with the goal of minimizing operation cost;

[0242] The distributed robust optimization model construction module 02 is used to construct a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and transform the distribution system operation model and the virtual power plant operation model into a distributed robust optimization model of the distribution system operation and a distributed robust optimization model of the virtual power plant operation respectively based on the fuzzy set;

[0243] The first game model construction module 03 is used to construct the first game model between multiple virtual power plants and the distribution system by taking the electricity price strategy determined by the distributed robust optimization model based on the distribution system operation and the grid-connected power determined by the distributed robust optimization model based on the virtual power plant operation as interactive variables;

[0244] The second game model construction module 04 is used to construct a second game model between the virtual power plants by taking the active power transmitted between the virtual power plants determined by the power price strategy and the blue optimization model of the virtual power plant operation distribution as the interaction variable;

[0245] A dual game model construction module 05 is used to construct a dual game model of multiple virtual power plants and distribution systems based on the first game model and the second game model;

[0246] The output result determination module 06 is used to solve the double game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

[0247] It should be noted that each unit in the above-mentioned dual game system of multiple virtual power plants and distribution systems can be fully or partially implemented by software, hardware and their combination. The above-mentioned units can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. For the specific definition of a dual game system of multiple virtual power plants and distribution systems, please refer to the definition of a dual game method of multiple virtual power plants and distribution systems in the above text. The two have the same functions and effects, which will not be repeated here.

[0248] In summary, the embodiment of the present invention provides a dual game method and system for multiple virtual power plants and distribution systems. The uncertainty of distributed photovoltaic output and load is considered based on KL divergence, and conservative control of distributed robust optimization is achieved by adjusting the KL divergence, thereby improving the applicability of the dual game method for multiple virtual power plants and distribution systems. A dual game framework for multiple virtual power plants and distribution systems is constructed by an interactive game strategy between the distribution system and multiple virtual power plants established with electricity price strategy and grid-connected power as interactive variables and a cooperative game strategy between multiple virtual power plants established with active power as interactive variables, which helps to better integrate distributed energy into the power system and improve the resource allocation efficiency of the entire power system.

[0249] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0250] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention. These improvements and substitutions should also be regarded as the scope of protection of the present invention.

Claims

1. A dual game method of multiple virtual power plants and distribution systems, characterized in that: include: With the goal of minimizing operating costs, the distribution system operation model and the virtual power plant operation model are constructed respectively; Constructing a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and converting the distribution system operation model and the virtual power plant operation model into a distribution system operation distributed robust optimization model and a virtual power plant operation distributed robust optimization model respectively based on the fuzzy set; Taking the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables, a first-level game model between multiple virtual power plants and the distribution system is constructed; Taking the active power transmitted between the virtual power plants determined based on the electricity price strategy and the blue optimization model of the virtual power plant operation distribution as the interactive variable, a second game model between the virtual power plants is constructed; Based on the first game model and the second game model, a dual game model of multiple virtual power plants and distribution systems is constructed; The double game model is solved to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

2. The dual game method of multiple virtual power plants and power distribution systems according to claim 1 is characterized in that: The objective of minimizing the operating cost is to construct a distribution system operating model and a virtual power plant operating model, respectively, including: Taking minimizing the operating cost of the distribution system as the goal, establishing an objective function, constructing an initial distribution system operation model, and using a first constraint condition to constrain the initial distribution system operation model to obtain a distribution system operation model, wherein the first constraint condition includes a first distributed photovoltaic operation constraint, a first gas turbine operation constraint, a first energy storage operation constraint, a first reactive compensation device operation constraint, a virtual power plant price setting constraint, and a power flow constraint; With the goal of minimizing the operating cost of the virtual power plant, an objective function is established, an initial virtual power plant operating model is constructed, and the initial virtual power plant operating model is constrained by a second constraint condition to obtain a virtual power plant operating model, wherein the second constraint condition includes a second distributed photovoltaic operating constraint, a second gas turbine operating constraint, a second energy storage operating constraint, a second reactive compensation device operating constraint, energy exchange constraints between virtual power plants, and a power balance constraint.

3. The dual game method of multiple virtual power plants and power distribution systems according to claim 2 is characterized in that: The objective function of the distribution system operation model is characterized by the following formula: in, represents the operating cost of the power distribution system, T represents the total operating time, represents the operating cost of distributed photovoltaic power generation in the power distribution system at time t, represents the gas turbine operating cost in the power distribution system at time t, represents the energy storage operation cost in the distribution system at time t, represents the energy exchange cost between the distribution system and the virtual power plant at time t; The following formula is used to characterize the first distributed photovoltaic constraint: in, represents the active power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the reactive power transmission of distributed photovoltaic at node i in the distribution system at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i in the distribution system at time t, represents the power generation capacity of distributed photovoltaic at node i in the distribution system at time t; The first gas turbine operation constraint is characterized by the following formula: in, represents the active power transmission of the gas turbine at node i in the power distribution system at time t, represents the reactive power transmission of the gas turbine at node i in the power distribution system at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i in the power distribution system at time t, represents the power factor of the gas turbine at node i at time t in the power distribution system; The following formula is used to characterize the first energy storage operation constraint: in, represents the energy storage charging state at node i in the distribution system at time t, represents the energy storage discharge state at node i in the distribution system at time t, represents the energy storage charging power at node i in the distribution system at time t, represents the energy storage discharge power at node i in the distribution system at time t, represents the lower limit of energy storage charging power at node i in the power distribution system at time t, represents the upper limit of energy storage charging power at node i in the power distribution system at time t, represents the lower limit of energy storage discharge power at node i in the distribution system at time t, represents the upper limit of energy storage discharge power at node i in the distribution system at time t; The following formula is used to characterize the operation constraint of the first reactive power compensation device: in, It represents the reactive power provided by the reactive power compensation device at node i in the power distribution system at time t. represents the lower limit of reactive power provided by the reactive power compensation device at node i in the distribution system at time t, represents the upper limit of reactive power provided by the reactive power compensation device at node i in the power distribution system at time t; The following formula is used to characterize the virtual power plant price setting constraints: in, represents the active power price of the nth virtual power plant in the distribution system at time t, represents the reactive power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the upper limit of the active power price of the nth virtual power plant in the distribution system at time t, represents the lower limit of reactive power price of the nth virtual power plant in the distribution system at time t, represents the upper limit of reactive power price of the nth virtual power plant in the distribution system at time t; The power flow constraint is characterized by the following formula: in, represents the active node injection amount at node j in the distribution system at time t, represents the reactive node injection amount at node j in the distribution system at time t, represents the active node injection amount at node i in the distribution system at time t, represents the reactive node injection amount at node i in the distribution system at time t, It indicates the active power at the head end of line jk of the power distribution system at time t. It represents the reactive power of the first end of line jk at time t in the power distribution system. It represents the active power at the head end of line ij of the power distribution system at time t. It represents the reactive power of the line ij at the time t of the power distribution system. represents the set of nodes in the power distribution system, represents the active load at node i in the distribution system at time t, represents the reactive load at node i in the distribution system at time t, represents the active amount of interaction between the distribution system and the virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant, represents the resistance of line ij, represents the line ij reactance, represents the voltage at node i in the power distribution system at time t, represents the lower limit of voltage at node i in the power distribution system at time t, represents the upper voltage limit at node i in the power distribution system at time t, Indicates the upper limit of the current that line ij can withstand. Represents the current on line ij at time t in the distribution system.

4. The dual game method of multiple virtual power plants and power distribution systems according to claim 2 is characterized in that: The objective function of the virtual power plant operation model is characterized by the following formula: in, represents the operating cost of the nth virtual power plant, represents the operating cost of distributed photovoltaic power generation in the nth virtual power plant at time t, represents the operating cost of the gas turbine in the nth virtual power plant at time t, represents the energy storage operation cost in the nth virtual power plant at time t, represents the energy exchange cost between the nth virtual power plant and the distribution system at time t, represents the energy interaction cost between the nth virtual power plant and other virtual power plants at time t; The following formula is used to characterize the second distributed photovoltaic constraint: in, represents the active power transmission of distributed photovoltaic at node i of the virtual power plant at time t, represents the reactive power transmission of distributed photovoltaic at node i of virtual power plant at time t, represents the upper limit of the active power transmission capacity of distributed photovoltaic at node i of the virtual power plant at time t, represents the power generation capacity of distributed photovoltaic at node i of the virtual power plant at time t; The second gas turbine operation constraint is characterized by the following formula: in, represents the active power transmission of the gas turbine at node i in the virtual power plant at time t, represents the reactive power transmission of the gas turbine at node i of the virtual power plant at time t, represents the upper limit of the active power transmission capacity of the gas turbine at node i of the virtual power plant at time t, represents the power factor of the gas turbine at node i at time t in the virtual power plant; The following formula is used to characterize the second energy storage operation constraint: in, represents the energy storage charging state at node i of the virtual power plant at time t, represents the energy storage discharge state at node i of the virtual power plant at time t, represents the energy storage charging power at node i of the virtual power plant at time t, represents the energy storage discharge power at node i of the virtual power plant at time t, represents the lower limit of energy storage charging power at node i of the virtual power plant at time t, represents the upper limit of energy storage charging power at node i of the virtual power plant at time t, represents the lower limit of energy storage discharge power at node i of the virtual power plant at time t, represents the upper limit of energy storage discharge power at node i of the virtual power plant at time t; The following formula is used to characterize the operation constraint of the second reactive power compensation device: in, represents the reactive power provided by the reactive power compensation device at node i of the virtual power plant at time t, represents the lower limit of reactive power provided by the reactive compensation device at node i of the virtual power plant at time t, represents the reactive power upper limit provided by the reactive power compensation device at node i of the virtual power plant at time t; The following formula is used to characterize the energy exchange constraints between the virtual power plants: in, represents the energy transmitted from the mth virtual power plant to the nth virtual power plant at time t, represents the energy transmitted from the nth virtual power plant to the mth virtual power plant at time t, represents the active load of the nth virtual power plant at time t; The power balance constraint is characterized by the following formula: in, represents the reactive load of the nth virtual power plant at time t, represents the active amount of interaction between the distribution system and the virtual power plant, represents other virtual power plants except the nth virtual power plant, Represents the reactive power of the interaction between the distribution system and the virtual power plant.

5. The dual game method of multiple virtual power plants and power distribution systems according to claim 1 is characterized in that: The method of constructing a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and converting the distribution system operation model and the virtual power plant operation model into a distribution system operation distributed robust optimization model and a virtual power plant operation distributed robust optimization model respectively based on the fuzzy set, comprises: Performing scene clustering on distributed photovoltaic historical data to obtain multiple characteristic scenes and initial probability distribution of each characteristic scene, wherein the distributed photovoltaic historical data includes distributed photovoltaic output historical data and distributed photovoltaic load historical data; Based on the constraint of KL divergence on the error between the initial probability distribution and the true probability distribution of each of the feature scenes, a probability distribution uncertainty set is constructed, and the probability distribution uncertainty set is characterized as a fuzzy set; Based on the fuzzy set and the two-stage decision variable setting rule, the distribution system operation model is converted into a distribution system operation distributed robust optimization model, wherein the two-stage decision variable setting rule is to set the discrete variables as the first-stage decision variables and the non-discrete variables as the second-stage decision variables; Based on the fuzzy sets and the two-stage decision variable setting rules, the virtual power plant operation model is converted into a virtual power plant operation distribution robust optimization model.

6. The dual game method of multiple virtual power plants and power distribution systems according to claim 5 is characterized in that: The following formula is used to characterize the distribution system operation distribution robust optimization model or the virtual power plant operation distribution robust optimization model: Among them, s represents the sequence number of the feature scene, represents the total number of feature scenes, represents the true probability distribution of feature scene s, represents a fuzzy set, represents the first-stage decision variable, represents the second-stage decision variable, represents the set of decision variable constraints in the first stage, represents the set of decision variable constraints in the second stage, , , , , , , , , , , , , , Represents the constant matrix corresponding to the decision variable constraints.

7. The dual game method of multiple virtual power plants and power distribution systems according to claim 1 is characterized in that: The first game model between multiple virtual power plants and the distribution system is constructed by taking the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables, including: Solving the blue stick optimization model for the operation of the virtual power plant, and transferring the solved grid-connected power to the distribution system; Solving the distributed robust optimization model of the power distribution system based on the grid-connected power, and transmitting the solved electricity price strategy to the virtual power plant; The electricity price strategy and the grid-connected power are used as interactive variables to iterate and interact until a Nash equilibrium is reached, thereby obtaining a first-level game model between multiple virtual power plants and a distribution system.

8. The dual game method of multiple virtual power plants and power distribution systems according to claim 1 is characterized in that: The active power transmitted between the virtual power plants determined based on the electricity price strategy and the virtual power plant operation distribution robust optimization model is used as an interactive variable to construct a second game model between the virtual power plants, including: Solving the blue-robust optimization model of the virtual power plant operation distribution based on the electricity price strategy to obtain the active power of each virtual power plant; The active power transmitted between the virtual power plants is used as an interactive variable for iterative interaction until a Nash equilibrium is reached, thereby obtaining a second-level game model between the virtual power plants.

9. The dual game method of multiple virtual power plants and power distribution systems according to claim 1, characterized in that: Solving the dual game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system includes: The double game model is solved by a solver to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

10. A dual game system of multiple virtual power plants and power distribution systems, characterized in that: include: An operation model building module is used to build a distribution system operation model and a virtual power plant operation model respectively with the goal of minimizing the operation cost; A distributed robust optimization model construction module is used to construct a fuzzy set based on KL divergence according to the uncertainty of distributed photovoltaics, and based on the fuzzy set, the distribution system operation model and the virtual power plant operation model are converted into a distributed robust optimization model for distribution system operation and a distributed robust optimization model for virtual power plant operation respectively; A first game model construction module is used to construct a first game model between multiple virtual power plants and the distribution system, using the electricity price strategy determined based on the distributed robust optimization model of the distribution system operation and the grid-connected power determined based on the distributed robust optimization model of the virtual power plant operation as interactive variables; A second game model construction module, used to construct a second game model between the virtual power plants, taking the active power transmitted between the virtual power plants determined based on the electricity price strategy and the virtual power plant operation distribution blue optimization model as an interactive variable; A dual game model construction module, used to construct a dual game model of multiple virtual power plants and distribution systems based on the first game model and the second game model; The output result determination module is used to solve the double game model to obtain the optimal power generation plan of the upper power grid and the optimal energy interaction strategy between the virtual power plant and the distribution system.

Images