Virtual power plant optimal bidding strategy considering operation reliability of power distribution network

Abstract

The invention discloses a virtual power plant optimal bidding strategy considering operation reliability of a power distribution network, and belongs to the technical field of new energy. According to the method, firstly, a virtual power plant optimal bidding strategy process considering the operation reliability of the power distribution network is constructed; secondly, constructing a virtual power plant optimal bidding strategy model considering the operation reliability of the power distribution network based on a double-layer optimization theory; and finally, solving the virtual power plant optimal operation model considering the operation reliability of the power distribution network, and carrying out linear solving on the proposed double-layer optimization problem. According to the method, the profitability of the virtual power plant is improved, meanwhile, the operation reliability of the power grid is considered, a large amount of data and computing resources are saved, and the method has the advantages of being good in expansibility, high in convergence speed and the like.

Description

Technical Field

[0001] The present invention belongs to the field of new energy technology, and in particular relates to an optimal bidding strategy for a virtual power plant taking into account the operating reliability of a distribution network. Background Art

[0002] Through advanced information and communication technologies and software systems, virtual power plants can effectively aggregate distributed energy such as renewable energy, energy storage systems, self-contained power plants, and controllable loads to participate in the power market, promote the elastic consumption of renewable energy, enhance the flexibility of the power system, and provide guarantees for the safe and reliable operation of the distribution network. It has received widespread attention from researchers at home and abroad. When responding to grid demand, virtual power plants usually consider the actual location of distributed energy in the distribution network, and consider the network operation constraints of distributed energy in decision-making, and provide auxiliary services such as congestion management, voltage control, and peak shaving to the distribution network on the premise of providing reliable services. At the same time, the occurrence of extreme emergencies in the distribution network may have a bad impact on the operation of virtual power plants, and bring great challenges to the formulation of the optimal bidding strategy for virtual power plants. Therefore, incorporating the operational reliability of the distribution network into the optimal bidding framework of the virtual power plant, constructing the optimal operation model of the virtual power plant considering the operational reliability of the distribution network, and formulating the optimal bidding strategy of the virtual power plant considering the operational reliability of the distribution network are of great significance to the safe and reliable operation of the distribution network.

[0003] The current research methods on the optimal bidding strategy of virtual power plants mainly rely on precise optimization models and optimization algorithms, mainly including mixed integer linear programming, reinforcement learning and dynamic programming, to maximize the economic benefits of virtual power plants in the electricity market. Mixed integer linear programming accurately optimizes resource scheduling and market bidding strategies by establishing linear objective functions and constraints, but may encounter computational complexity problems when facing large-scale problems. Reinforcement learning can dynamically adjust decision-making strategies to cope with market uncertainties through the interaction between intelligent agents and the environment, but its learning process requires a large amount of data and computing resources. Dynamic programming optimizes resource scheduling on time series by decomposing complex problems into multiple staged sub-problems, but may face the curse of dimensionality when dealing with high-dimensional problems.

[0004] However, the above technologies mainly focus on how to build a more sophisticated virtual power plant operation model for operational decision-making. The research on the optimal bidding strategy for virtual power plants considering the reliability of distribution network operation still has the following problems:

[0005] Existing research focuses on maximizing the operational benefits of virtual power plants participating in distribution network demand response, ignoring the important impact of network operation reliability on the formulation of virtual power plant bidding strategies;

[0006] Taking into account both the reliability of distribution network operation and the optimal operating efficiency of virtual power plants, it is difficult to construct an optimal operation model of virtual power plants that takes into account the reliability of distribution network operation;

[0007] The formulation of a virtual power plant bidding strategy is a collaborative optimization problem involving multiple objectives and multiple subject decision coupling. It is difficult to decouple the decision-making behaviors of multiple subjects, and it is difficult to find the optimal operating benefits of the virtual power plant.

[0008] Therefore, a new technical solution is urgently needed in the prior art to solve this problem. Summary of the invention

[0009] The technical problem to be solved by the present invention is to provide an optimal bidding strategy for virtual power plants that takes into account the operating reliability of the distribution network, so as to solve the technical problem that the algorithms and models in the prior art fail to take into account both the operating reliability of the distribution network and the optimal operating efficiency of the virtual power plant.

[0010] The technical solution adopted by the present invention is to provide an optimal bidding strategy for a virtual power plant taking into account the reliability of distribution network operation, comprising the following steps:

[0011] Step 1: Construct the optimal bidding strategy process of virtual power plants considering the reliability of distribution network operation;

[0012] Step 2: Based on the double-layer optimization theory, the optimal bidding model of virtual power plants considering the reliability of distribution network operation is constructed;

[0013] Step 3: Solve the optimal operation model of the virtual power plant considering the reliability of distribution network operation.

[0014] The specific process in step 1 is as follows:

[0015] 1) Select the distributed energy target entities that the virtual power plant needs to aggregate, including but not limited to photovoltaic units, micro turbines, electric vehicles and interruptible loads. The virtual power plant aims to maximize operating profits while improving the reliability of the distribution network and formulate the optimal bidding strategy for participating in the day-ahead power market;

[0016] 2) Obtain the actual data set of virtual power plants participating in the day-ahead power market, focusing on the distribution network topology and its operation constraint data, load volume and interruptible load data, as well as the predicted power of photovoltaic units, electric vehicle charging and discharging curves, and distributed energy output data of micro turbines;

[0017] 3) Investigate the reliability analysis of extreme emergencies in the distribution network. For isolated islands that do not have distributed power sources or cannot meet their full load demand, additional power switching actions are required to restore the users who have been cut as soon as possible;

[0018] 4) Based on the double-layer optimization theory, the upper-layer optimization problem is constructed with the goal of optimizing the operating efficiency of the virtual power plant considering the reliability of distribution network operation;

[0019] 5) From the perspective of independent system operators, the competitive day-ahead electricity market is simulated and the underlying optimization problem is constructed with the goal of optimizing social welfare.

[0020] The upper layer of the optimal bidding model for virtual power plants considering the reliability of distribution network operation in step 2 specifically includes:

[0021] 1) Construction of the upper problem: The upper problem aims to optimize the operating efficiency of the virtual power plant considering the reliability of the distribution network. Its objective function includes the profit obtained from participating in the day-ahead market for electricity transactions, the profit obtained from electricity transactions with the distribution network, and the load reduction and load interruption costs under extreme emergencies, which can be expressed as

[0022]

[0023] Where: T is the scheduling period set, t is the scheduling period index; S D , S C , S R are the bidding uncertainty of rivals, the aggregation uncertainty of distributed energy resources, and the uncertainty scenario set of emergencies, s D 、s C 、s R is the uncertainty scenario index; F is the number of market participants’ bidding curve segments, and f is the segment index; b T is the transmission network node index, For node b T The locational marginal price of are the load reduction and load interruption costs of the virtual power plant at node b in period t, is the time-of-use electricity price in the distribution station area z during period t, is the location marginal price of the virtual power plant connecting to the distribution network node b in period t; are the number of electricity sales bids accepted by the ith market participant and the number of electricity purchase bids accepted by the dth market participant, respectively; are the active load power, active load reduction power and active load interruption power of node b in time period t respectively; Cost of generating electricity for distributed generation;

[0024] 2) Upper-level problem constraints: In the day-ahead market, each electricity producer will provide the market with a segmented quantity-price pair (ψ ift ,π ift ) submits a power supply bidding contract, which includes the power supply and price of each round of bidding in each period of the next day. At the same time, each power consumer will also submit a power consumption bidding contract (Ωdft ,π dft ), including the power consumption and price of each round of quotations in each period of the next day. After the market participants submit the bidding bids, the independent system operator clears the market in the form of payment bidding and determines the acceptable amount of the participants' bids. The bidding power constraints of the bidding contracts are as follows:

[0025]

[0026]

[0027] Where: ift With Ω dft The bid amounts of electricity producer i and electricity consumer d are respectively, is the active output power of photovoltaic unit p, is the active output power of distributed power source u, and are the charging and discharging active power of electric vehicle e; U N , P N 、E N They are a collection of distributed power sources, a collection of photovoltaic units, and a collection of electric vehicles interconnected with the power grid;

[0028] Day-ahead market participants need to meet the power balance constraints:

[0029]

[0030]

[0031] Where: FP lt With FQ lt are the active power flow and reactive power flow of line l, C lb is the distribution network fault matrix; is the reactive output power of photovoltaic unit p, is the reactive output power of distributed power source u, and are the charging and discharging reactive power of electric vehicle e respectively; They are the active load power, active load reduction power and active load interruption power of node b respectively.

[0032] The lower layer of the optimal bidding model for virtual power plants considering the reliability of distribution network operation in step 2 specifically includes:

[0033] 1) Construction of the lower-level problem: From the perspective of the independent system operator, the lower-level problem aims to clear the day-ahead market and achieve optimal social welfare. The solution to the lower-level problem is the amount of electricity purchased and sold between the bidding participants in each period and the location marginal price of each node in the distribution network, expressed as

[0034]

[0035] 2) Constraints of the lower problem: The lower problem must satisfy the constraints on the number of bids accepted, the node power balance, the transmission line flow, and the node voltage phase angle, as shown below:

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042] Where: b T , l T Respectively represent the nodes and lines of the transmission network, For node b T The voltage phase angle of B X-T represents the relationship between the injected power and the voltage phase angle at each node; Formula (9) represents the power balance relationship of each node in the transmission network. In this framework, the location marginal price at each node is endogenously generated in each time period. For transmission line T The active power, For transmission line T The maximum active power that can be carried; Formula (11) represents the voltage phase angle constraint of each node, and Formula (12) represents the voltage phase angle of the relaxed node to be 0; μ, γ, ζ, χ are the Lagrangian coefficients corresponding to the above constraints;

[0043] The bidding curves for power supply and consumption of virtual power plants are increasing curves and decreasing curves respectively, which need to meet the following constraints:

[0044]

[0045]

[0046] The model conversion and solution in step 3 are as follows:

[0047] 1) Using the KKT optimality condition and strong duality theory, the nonlinear part of the objective function of the upper problem is converted into a linearized form as shown below:

[0048]

[0049] 2) On this basis, auxiliary binary variables and the big M method can be used to linearize the KKT complementary relaxation conditions. Thus, the two-level optimization problem is transformed into a single-level mixed integer linear programming model, which can be solved by using the Cplex / Gurobi solver.

[0050] Through the above design scheme, the present invention can bring the following beneficial effects:

[0051] 1. The optimal bidding strategy process for virtual power plants considering the reliability of distribution network operation is constructed to design the optimal bidding strategy for virtual power plants to participate in the day-ahead market, so as to improve the profitability of virtual power plants while taking into account the reliability of power grid operation;

[0052] 2. Based on the double-layer optimization theory, the optimal bidding strategy model of virtual power plants considering the reliability of distribution network operation is constructed to characterize the game relationship between multiple entities participating in the day-ahead market of virtual power plants, taking into account both the maximization of social welfare and the optimization of virtual power plant operation efficiency;

[0053] 3. The proposed two-layer optimization problem is linearly solved, which saves a lot of data and computing resources and has the advantages of good scalability and fast convergence speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of an optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation according to the present invention;

[0055] Figure 2 This is an overall framework diagram of virtual power plants participating in the day-ahead market according to the optimal bidding strategy for virtual power plants that takes into account the reliability of distribution network operation. DETAILED DESCRIPTION

[0056] The present invention is further described below with reference to the accompanying drawings and specific embodiments:

[0057] like Figure 1 As shown, the optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation of the present invention specifically includes the following:

[0058] Step 1: The process of constructing the optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation is as follows:

[0059] 1) Select the distributed energy target entities that the virtual power plant needs to aggregate, including but not limited to photovoltaic units, micro turbines, electric vehicles and interruptible loads. The virtual power plant aims to maximize operating profits while improving the reliability of the distribution network and formulate the optimal bidding strategy for participating in the day-ahead power market. Figure 2 The overall framework of virtual power plants participating in the day-ahead market is given.

[0060] 2) Obtain the actual data set of virtual power plants participating in the day-ahead power market, focusing on the distribution network topology and its operating constraint data, load volume and interruptible load data, as well as photovoltaic unit predicted power, electric vehicle charging and discharging curves, and distributed energy output data such as micro turbines.

[0061] 3) Investigate extreme emergencies in the distribution network and perform reliability analysis. Extreme emergencies in the distribution network may cause some grids to be disconnected from the main grid, forming islands. Some islands contain distributed generators and renewable energy sources that can meet their load needs. For islands that do not have distributed power sources or cannot meet their full load needs, additional power switching actions are required to restore the cut users as soon as possible.

[0062] 4) Based on the double-layer optimization theory, the upper-layer optimization problem is constructed with the goal of optimizing the operating efficiency of the virtual power plant considering the reliability of the distribution network operation. Specifically, in the bidding stage, the virtual power plant and other competitors submit bids to the market. The bidding content is the electricity clearing price and its results in participating in the day-ahead power market, which is directly related to the operating efficiency of the virtual power plant.

[0063] 5) From the perspective of independent system operators, the competitive day-ahead power market is simulated, and the lower-level optimization problem is constructed with the goal of optimizing social welfare. After receiving the bids, the independent operator performs power market clearing with the goal of optimizing social welfare, and feeds back the final clearing price and its results to the virtual power plant and other competitors.

[0064] Step 2: Based on the two-level optimization theory, the optimal bidding model of virtual power plants considering the reliability of distribution network operation is constructed.

[0065] 1) Construction of the upper problem: The upper problem aims to optimize the operating efficiency of the virtual power plant considering the reliability of the distribution network. Its objective function mainly includes the profit obtained from participating in the day-ahead market for electricity transactions, the profit obtained from electricity transactions with the distribution network, and the load reduction and load interruption costs under extreme emergencies, which can be expressed as

[0066]

[0067] Where: T is the scheduling period set, t is the scheduling period index; S D , S C , S R are the bidding uncertainty of rivals, the aggregation uncertainty of distributed energy resources, and the uncertainty scenario set of emergencies, s D 、s C 、s R is the uncertainty scenario index; F is the number of market participants’ bidding curve segments, and f is the segment index; b T is the transmission network node index, For node bT The locational marginal price of are the load reduction and load interruption costs of the virtual power plant at node b in period t, is the time-of-use electricity price in the distribution station area z during period t, is the location marginal price of the virtual power plant connecting to the distribution network node b in period t; are the number of electricity sales bids accepted by the ith market participant and the number of electricity purchase bids accepted by the dth market participant, respectively; are the active load power, active load reduction power and active load interruption power of node b in time period t respectively; The cost of distributed generation.

[0068] 2) Upper-level problem constraints: In the day-ahead market, each electricity producer will provide the market with a segmented quantity-price pair (ψ ift ,π ift ) submits a power supply bidding contract, which includes the power supply and price of each round of bidding in each period of the next day. At the same time, each power consumer will also submit a power consumption bidding contract (Ω dft ,π dft ), including the electricity consumption and price of each round of quotations in each period of the next day. After market participants submit their competitive bids, the independent system operator clears the market in the form of paid bids and determines the acceptable amount of the participants' bids. Therefore, when designing a bidding strategy for a virtual power plant, its available resources and user load should be fully considered. In addition, in the bidding for the virtual power plant market, it is necessary to consider the situation where emergencies cause the distributed energy to be disconnected from the main grid to form an island. For this reason, the bidding power constraints for the bidding contracts are constructed as shown below.

[0069]

[0070]

[0071] Where: ift With Ω dft The bid amounts of electricity producer i and electricity consumer d are respectively, is the active output power of photovoltaic unit p, is the active output power of distributed power source u, and are the charging and discharging active power of electric vehicle e; U N , P N 、E N They are a collection of distributed power sources interconnected with the power grid, a collection of photovoltaic units, and a collection of electric vehicles.

[0072] In addition, day-ahead market participants need to meet power balance constraints:

[0073]

[0074]

[0075] Where: FP lt With FQ lt are the active power flow and reactive power flow of line l, C lb is the distribution network fault matrix; is the reactive output power of photovoltaic unit p, is the reactive output power of distributed power source u, and are the charging and discharging reactive power of electric vehicle e respectively; They are the active load power, active load reduction power and active load interruption power of node b respectively.

[0076] 3) Construction of the lower-level problem: The lower-level problem aims to clear the day-ahead market and achieve optimal social welfare from the perspective of the independent system operator. The solution to the lower-level problem is the amount of electricity purchased and sold between the bidding participants in each period and the location marginal price of each node in the distribution network, expressed as

[0077]

[0078] 4) Constraints of the lower problem: The lower problem must satisfy the constraints on the number of bids accepted, the node power balance, the transmission line flow, and the node voltage phase angle, as shown below:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] Where: b T , l T Respectively represent the nodes and lines of the transmission network, For node b T The voltage phase angle of B X-T It represents the relationship between the injected power and the voltage phase angle of each node; Formula (9) represents the power balance relationship of each node in the transmission network. In this framework, the location marginal price at each node is endogenously generated in each time period. For transmission lineT The active power, For transmission line T The maximum active power that can be carried; Formula (11) represents the voltage phase angle constraint of each node, and Formula (12) represents the voltage phase angle of the relaxed node to be 0; μ, γ, ζ, χ are the Lagrangian coefficients corresponding to the above constraints.

[0086] Virtual power plants need to obtain the economic bidding curves of all participants in the process of participating in the day-ahead market clearing. However, since virtual power plants cannot obtain this information, scenario simulation methods are used to predict competitors' bidding curves and consider their bidding uncertainty. Therefore, when solving the underlying problem, the uncertainty of competitors' bids needs to be considered to extract the optimal bidding strategy for virtual power plants in multiple periods. The bidding curves for power supply and consumption of virtual power plants should be increasing curves and decreasing curves, respectively, and must meet the following constraints:

[0087]

[0088]

[0089] Step 3: Solve the optimal operation model of the virtual power plant considering the reliability of distribution network operation

[0090] 1) Since the bid price of the virtual power plant participating in the day-ahead market is ultimately specified by the independent system operator, the lower-level problem is a linear problem and a convex optimization problem. Therefore, the Carlo-Kuhn-Tucker (KKT) optimality condition is adopted to transform the lower-level problem into a set of constraints of the upper-level problem through the Lagrangian equation related to the lower-level problem, and to generate the equilibrium constrained programming (MPEC) problem equivalently, which has the upper-level problem objective function, i.e., equation (1) and its nonlinear constraints, as well as the KKT complementary condition nonlinear problem.

[0091] 2) The nonlinearity of the objective function of the upper problem is given by the term and Therefore, by using the KKT optimality condition and strong duality theory, the nonlinear part of the objective function of the upper problem can be converted into a linear form as shown below.

[0092]

[0093] 3) On this basis, auxiliary binary variables and the big M method can be used to linearize the KKT complementary relaxation conditions. Thus, the two-level optimization problem is transformed into a single-level mixed integer linear programming model, and the model can be solved using the Cplex / Gurobi solver.

[0094] The implementation methods of the present invention are not limited to the above-mentioned embodiments, and any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. An optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation, characterized by: The following steps are involved: Step 1: Construct the optimal bidding strategy process of virtual power plants considering the reliability of distribution network operation; Step 2: Based on the double-layer optimization theory, the optimal bidding model of virtual power plants considering the reliability of distribution network operation is constructed; Step 3: Solve the optimal operation model of the virtual power plant considering the reliability of distribution network operation.

2. According to claim 1, the optimal bidding strategy for a virtual power plant taking into account the reliability of distribution network operation is characterized in that: The specific process in step 1 is as follows: 1) Select the distributed energy target entities that the virtual power plant needs to aggregate, including but not limited to photovoltaic units, micro turbines, electric vehicles and interruptible loads. The virtual power plant aims to maximize operating profits while improving the reliability of the distribution network and formulate the optimal bidding strategy for participating in the day-ahead power market; 2) Obtain the actual data set of virtual power plants participating in the day-ahead power market, focusing on the distribution network topology and its operation constraint data, load volume and interruptible load data, as well as the predicted power of photovoltaic units, electric vehicle charging and discharging curves, and distributed energy output data of micro turbines; 3) Investigate the reliability analysis of extreme emergencies in the distribution network. For isolated islands that do not have distributed power sources or cannot meet their full load demand, additional power switching actions are required to restore the users who have been cut as soon as possible; 4) Based on the double-layer optimization theory, the upper-layer optimization problem is constructed with the goal of optimizing the operating efficiency of the virtual power plant considering the reliability of distribution network operation; 5) From the perspective of independent system operators, the competitive day-ahead electricity market is simulated and the underlying optimization problem is constructed with the goal of optimizing social welfare.

3. The optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation according to claim 1 is characterized in that: The upper layer of the optimal bidding model for virtual power plants considering the reliability of distribution network operation in step 2 specifically includes: 1) Construction of the upper problem: The upper problem aims to optimize the operating efficiency of the virtual power plant considering the reliability of the distribution network. Its objective function includes the profit obtained from participating in the day-ahead market for electricity transactions, the profit obtained from electricity transactions with the distribution network, and the load reduction and load interruption costs under extreme emergencies, which can be expressed as Where: T is the scheduling period set, t is the scheduling period index; S D , S C , S R are the bidding uncertainty of rivals, the aggregation uncertainty of distributed energy resources, and the uncertainty scenario set of emergencies, s D 、s C 、s R is the uncertainty scenario index; F is the number of market participants’ bidding curve segments, and f is the segment index; b T is the transmission network node index, For node b T The locational marginal price of are the load reduction and load interruption costs of the virtual power plant at node b in period t, is the time-of-use electricity price in the distribution station area z during period t, is the location marginal price of the virtual power plant connecting to the distribution network node b in period t; are the number of electricity sales bids accepted by the ith market participant and the number of electricity purchase bids accepted by the dth market participant, respectively; are the active load power, active load reduction power and active load interruption power of node b in time period t respectively; Cost of generating electricity for distributed generation; 2) Upper-level problem constraints: In the day-ahead market, each electricity producer will provide the market with a segmented quantity-price pair (ψ ift ,π ift ) submits a power supply bidding contract, which includes the power supply and price of each round of bidding in each period of the next day. At the same time, each power consumer will also submit a power consumption bidding contract (Ω dft ,π dft ), including the power consumption and price of each round of quotations in each period of the next day. After the market participants submit the bidding bids, the independent system operator clears the market in the form of payment bidding and determines the acceptable amount of the participants' bids. The bidding power constraints of the bidding contracts are as follows: Where: ift With Ω dft The bid amounts of electricity producer i and electricity consumer d are respectively, is the active output power of photovoltaic unit p, is the active output power of distributed power source u, and are the charging and discharging active power of electric vehicle e; U N , P N 、E N They are a collection of distributed power sources, a collection of photovoltaic units, and a collection of electric vehicles interconnected with the power grid; Day-ahead market participants need to meet the power balance constraints: Where: FP lt With FQ lt are the active power flow and reactive power flow of line l, C lb is the distribution network fault matrix; is the reactive output power of photovoltaic unit p, is the reactive output power of distributed power source u, and are the charging and discharging reactive power of electric vehicle e respectively; They are the active load power, active load reduction power and active load interruption power of node b respectively.

4. The optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation according to claim 1 is characterized in that: The lower layer of the optimal bidding model for virtual power plants considering the reliability of distribution network operation in step 2 specifically includes: 1) Construction of the lower-level problem: From the perspective of the independent system operator, the lower-level problem aims to clear the day-ahead market and achieve optimal social welfare. The solution to the lower-level problem is the amount of electricity purchased and sold between the bidding participants in each period and the location marginal price of each node in the distribution network, expressed as 2) Constraints of the lower problem: The lower problem must satisfy the constraints on the number of bids accepted, the node power balance, the transmission line flow, and the node voltage phase angle, as shown below: Where: b T , l T Respectively represent the nodes and lines of the transmission network, For node b T The voltage phase angle of B X-T represents the relationship between the injected power and the voltage phase angle at each node; Formula (9) represents the power balance relationship of each node in the transmission network. In this framework, the location marginal price at each node is endogenously generated in each time period. For transmission line T The active power, For transmission line T The maximum active power that can be carried; Formula (11) represents the voltage phase angle constraint of each node, and Formula (12) represents the voltage phase angle of the relaxed node to be 0; μ, γ, ζ, χ are the Lagrangian coefficients corresponding to the above constraints; The bidding curves for power supply and consumption of virtual power plants are increasing curves and decreasing curves respectively, which need to meet the following constraints:

5. The optimal bidding strategy for a virtual power plant considering the reliability of distribution network operation according to claim 1 is characterized by: The model conversion and solution in step 3 are as follows: 1) Using the KKT optimality condition and strong duality theory, the nonlinear part of the objective function of the upper problem is converted into a linearized form as shown below: 2) On this basis, auxiliary binary variables and the big M method can be used to linearize the KKT complementary relaxation conditions. Thus, the two-level optimization problem is transformed into a single-level mixed integer linear programming model, which can be solved by using the Cplex / Gurobi solver.

Images