Virtual power plant aggregation method and system considering space-time coupling characteristics

Abstract

The invention relates to the technical field of virtual power plant aggregation, and particularly discloses a virtual power plant aggregation method and system considering space-time coupling characteristics, and the method comprises the steps: selecting a single network node of a virtual power plant region; selecting a single type of distributed resources accessed by the node to obtain a control variable parameter matrix and a vector formed by constrained constants; selecting a search direction set according to power-energy characteristics of the distributed resources on different time sequences; establishing an aggregation parameter optimal linear decision model by taking maximization of the sum of the extension lengths in all the dimension directions as an optimization target; solving the model to obtain an optimal odd polyhedron approximate feasible region; and further obtaining a chinonol polyhedron approximate aggregation feasible region of various types of resources of the node. According to the method, the Minkowski sum process of the polyhedron is converted into linear superposition by selecting different search directions, and feasibility is provided for feasible region aggregation of massive multi-element distributed resources in a high-dimensional space.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plant aggregation, and in particular to a virtual power plant aggregation method and system taking into account time-space coupling characteristics. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] A virtual power plant (VPP) is a model that aggregates dispersed distributed energy resources (DERs) through advanced information and communication technology (ICT) and energy management systems to form a unified dispatchable power supply or demand response capability.

[0004] Regarding the virtual power plant aggregation method, existing studies have proposed using a semi-plane representation of convex polytopes to describe resource flexibility. Compared with the traditional vertex-based representation, this improves computational efficiency, but it still cannot meet the aggregation needs of large-scale resources. The precise aggregation of high-dimensional feasible domain space still faces the problem of inaccurate calculation.

[0005] The prior art proposes an aggregation method for approximating the feasible domain. However, in the approximate solution process, the following technical problems often exist: The differences in electricity demand and physical characteristics of various resources in the virtual power plant and their coupling in time and space make the description process of the general resource model complicated; at the same time, the high-dimensional space of the feasible domain of various distributed resources easily leads to problems such as low computational efficiency of feasible domain aggregation methods such as Minkowski summation. Summary of the invention

[0006] In order to solve the above problems, the present invention proposes a virtual power plant aggregation method and system taking into account the time and space coupling characteristics, taking into account the time characteristics of different distributed resources and the coupling characteristics of their power and energy, and utilizing the affine-based security domain concept to effectively handle the spatial coupling relationship between resources, thereby improving the computational efficiency and realizing the effective characterization of the feasible domain of virtual power plant aggregation in the time / space dimension.

[0007] In some embodiments, the following technical solutions are adopted: A virtual power plant aggregation method considering spatiotemporal coupling characteristics, comprising: Select a single type of distributed resource accessed by a node in the virtual power plant area network, and select a set of search directions based on the power-energy characteristics of the distributed resource at different time sequences based on the Chino polyhedron approximation principle; Taking maximizing the sum of the extension lengths in each dimensional direction as the optimization goal, an optimal linear decision model for aggregation parameters is established; the model is solved to obtain the optimal Chino polyhedron approximate feasible domain of the distributed resource; Calculating the optimal Chino polyhedron approximate feasible domain of each type of distributed resource of the network node, and converting the Minkowski sum process of the polyhedron into a linear superposition by selecting different search directions to obtain the optimal Chino polyhedron approximate feasible domain of the network node; Considering the spatial coupling characteristics of virtual power plants, the optimal Chino polyhedron approximate feasible domain is modified to obtain the final aggregate feasible domain with the goal of maximizing the sum of the maximum extension interval lengths of the feasible domain of the common coupling point.

[0008] As a further solution, the distributed resources include at least: air-conditioning load, distributed energy storage, micro-turbines and distributed photovoltaics; for distributed photovoltaics, the constraints containing random variables are converted into deterministic constraints by using chance constraints, and the moment information of random variables is used to convert the chance constraints into processable linear constraints.

[0009] As a further solution, based on the Chino polyhedron approximation principle, a search direction set is selected according to the power-energy characteristics of the distributed resources at different time sequences, specifically: The search direction set of air conditioning load and distributed energy storage is composed of and The distributed photovoltaic search direction set is composed of The search direction set of the micro-turbine is composed of and constitute; in, It includes 2N search directions, each of which represents the power operation boundary of the common coupling point at a certain moment, and is used to approximate the extension direction of the aggregated resources at each decision moment; It includes N-1 search directions, each of which can represent the difference between the power of the next time period and the current time period, and is used to approximate the extension direction of the power-time coupling constraint of the aggregated resources: It includes N-1 search directions, representing the sum of energy changes in a certain time period, and is expressed as the sum of power values ​​over a certain period of time, which is used to approximate the extension direction of the power-energy coupling constraint of the aggregated resources; N is the time scale.

[0010] As a further solution, the optimal linear decision model for aggregation parameters is specifically: ; Where L is the sum of the extension lengths in each dimension, , The maximum extension length vector in each dimension direction, is the search direction matrix; are the coordinates of the center point of the polyhedron; is a non-negative Farkas multiplier, is the control variable parameter matrix, is a vector of constrained constants, is the original feasible domain of distributed resources, i for P Identification of the constraints in It is j non-negative Farkas multipliers in dimensions, It is the Nth+ j non-negative Farkas multipliers in dimensions, is the control variable parameter matrix j dimensions, is the coordinate of the center point of the polyhedron j dimensions.

[0011] As a further solution, considering the spatial coupling characteristics of the virtual power plant, the optimal Chino polyhedron approximate feasible domain is modified with the goal of maximizing the sum of the maximum extension interval lengths of the feasible domain of the common coupling point, specifically: The objective function is constructed with the constraints of the equivalent adjustable output of internal resources, participation factor constraints, the maximum extension of the feasible domain of the common coupling point, the internal network constraints of the virtual power plant and the power constraints of the coupling point as the constraints, and the maximum sum of the length of the feasible domain of the common coupling point as the goal. The nonlinear constraints in the constraints are linearized; the optimization model consisting of the objective function and constraints is solved using a commercial solver to obtain the revised aggregate feasible domain.

[0012] As a further solution, the objective function is specifically: ; in, and Point of Common Coupling o The lower and upper limits of the maximum extension of the feasible domain; t refers to a certain time period, and N refers to the number of decision moments for day-ahead scheduling.

[0013] In other embodiments, the following technical solutions are adopted: A virtual power plant aggregation system considering time-space coupling characteristics, comprising: Search direction selection module, which selects a single type of distributed resource connected to the virtual power plant regional network node. Based on the principle of approximation of the Chino polyhedron, according to the power-energy characteristics of the distributed resource at different time sequences, a set of search directions is selected; Feasible region solution module, which takes maximizing the sum of the extension lengths in each dimension direction as the optimization goal, and establishes an optimal linear decision model for aggregation parameters; solves the model to obtain the optimal Chino polyhedron approximation feasible region of the distributed resource; Feasible region aggregation module, which calculates the optimal Chino polyhedron approximation feasible region of each type of distributed resource of the network node, and transforms the Minkowski sum process of the polyhedron into a linear superposition by selecting different search directions to obtain the optimal Chino polyhedron approximation feasible region of the network node; Feasible region correction module, which considers the spatial coupling characteristics of the virtual power plant, and takes the maximum sum of the maximum extension interval lengths of the feasible region at the point of common coupling as the goal to correct the optimal Chino polyhedron approximation feasible region to obtain the final aggregated feasible region.

[0014] In some other embodiments, the following technical solutions are adopted: A terminal device, which includes a processor and a memory. The processor is used to implement instructions; the memory is used to store multiple instructions, and the instructions are suitable for being loaded and executed by the processor to perform the above-mentioned virtual power plant aggregation method considering spatio-temporal coupling characteristics.

[0015] In some other embodiments, the following technical solutions are adopted: A computer-readable storage medium, in which multiple instructions are stored, and the instructions are suitable for being loaded and executed by the processor of the terminal device to perform the above-mentioned virtual power plant aggregation method considering spatio-temporal coupling characteristics.

[0016] In some other embodiments, the following technical solutions are adopted: A computer program product, including a computer program / instructions, which when executed by the processor implements the above-mentioned virtual power plant aggregation method considering spatio-temporal coupling characteristics.

[0017] Compared with the prior art, the beneficial effects of the present invention are: (1) According to the linear constraints of each type of distributed resource independently participating in the optimal scheduling of the distribution network, the present invention obtains a vector consisting of a control variable parameter matrix and constraint constants; the overall feasible domain of each type of resource and the overall aggregated feasible domain of a single node are calculated respectively; in view of the difficult problem of Minkowski sum in high-dimensional space caused by the coupling characteristics of power and energy of virtual power plant resources in time series, the present invention proposes a Chino polyhedron aggregation method for searching in the time series coupling direction based on the Chino polyhedron approximation principle; by selecting different search directions, the Minkowski sum process of the polyhedron is transformed into a linear superposition, which provides feasibility for the aggregation of feasible domains in high-dimensional space for massive multivariate distributed resources.

[0018] (2) In order to make the calculated approximate Chino polyhedron as close as possible to the original feasible domain, the present invention aims to maximize the sum of the extension lengths in each dimensional direction, constructs an optimal linear decision model for aggregation parameters, and solves the corresponding linear programming (LP) problem to achieve the optimal aggregation parameters. This solves the complex problem of the spatial dimension of the feasible domain for the aggregation and regulation of various types of resources, and realizes the accurate characterization of the feasible domain of multi-energy heterogeneous resource aggregation taking into account the coupling characteristics.

[0019] (3) Based on the Zonotope efficient solution method of Farkas' lemma, the present invention constructs an optimal linear decision model for aggregation parameters, avoids the nonlinear problem of the Hausdorff distance method, and effectively improves the computational efficiency of the virtual power plant aggregation method.

[0020] (4) The present invention takes into account the spatial coupling characteristics of the virtual power plant, effectively handles the spatial coupling relationship between resources, and corrects the aggregated feasible domain, thereby solving the problem that the obtained approximate feasible domain ignores the impact of the spatial constraints on the distribution system network where the virtual power plant is located, and improving the calculation accuracy of the feasible domain of the common coupling point of the virtual power plant.

[0021] Other features and advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of a virtual power plant aggregation method that takes into account the spatiotemporal coupling characteristics in an embodiment of the present invention. DETAILED DESCRIPTION It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0023] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0024] Terminology explanation: "Aggregation parameters": matrix control variable parameter matrix and the vector of the constrained constants elements.

[0025] "Feasible domain" refers to the dynamic boundary range within which the distributed resources aggregated by the virtual power plant can respond safely, reliably and economically to grid dispatch or market demand under specific operating conditions. It is a set of feasible operations within a multi-dimensional constraint space formed by mathematizing the physical constraints, response characteristics and collaborative capabilities of distributed resources.

[0026] Power-energy characteristic: represents the functional relationship between power and energy.

[0027] Embodiment 1 In one or more embodiments, a virtual power plant aggregation method considering the spatiotemporal coupling characteristics is disclosed, combined with Figure 1 , specifically including the following process: S101: Select a single network node in the virtual power plant area; select a single type of distributed resource connected to the node, and based on the linear constraints of the distributed resources independently participating in the optimal scheduling of the distribution network, obtain a control variable parameter matrix and a vector composed of constraint constants.

[0028] In this embodiment, the virtual power plant aggregated distributed resources considered mainly include air conditioning loads, distributed energy storage, distributed photovoltaics, micro-turbines, etc. The linear constraints of each type of distributed resource independently participating in the optimal dispatch of the distribution network constitute the physical model of its own original feasible domain. The basic mathematical form is as follows: (1) in, is the original feasible domain of distributed resources; It is a vector of decision variables, generally the unit output or load; is the control variable parameter matrix, A vector of constants for the constraints; express N dimensional real vector space, which represents the time dimension in the following.

[0029] The specific distributed resources are as follows: (1) Air conditioning load For the establishment of a single air conditioning load model, the first-order equivalent thermal parameter model is often used to describe the thermal dynamic process inside and outside the room. The model is: (2) (3) in, is the indoor temperature, is the outdoor temperature; is the cooling capacity of the air conditioner; and are thermodynamic parameters, namely the thermal resistance of the indoor room and the heat capacity of the indoor air; is the working efficiency parameter of the air conditioner; is the electric power of the air conditioner. According to formula (2), the dynamic electrical model of the air conditioner is derived as follows: (4) (5) (6) in, for t Indoor temperature during the period; and They are the upper and lower limits of the comfortable temperature in the room; for t Outdoor temperature during the period; for t Electric power of time-slot air conditioner; is the maximum power of the air conditioner; is the decision time step.

[0030] In the process of aggregation and regulation of air conditioning as a flexible resource, since the thermodynamic parameters, efficiency, outdoor temperature, etc. of individual rooms are taken as constants, let , , formula (4) can be simplified as: (7) (2) Distributed energy storage The constraints of distributed energy storage include power constraints and state of charge (SOC) operation constraints. The model is: (8) (9) (10) (11) (12) in, and They are t Discharging and charging power of energy storage at all times; for t The charge and discharge state variable of the energy storage at each moment, 1 for discharge and 0 for charge; and are the maximum charging and discharging power of energy storage respectively; for t The state of charge of the energy storage at all times; The charging and discharging efficiency of energy storage; and They are the minimum and maximum charge states of energy storage respectively; The initial charge state of the energy storage during the dispatch cycle; It is the charge state of the energy storage at the end of the dispatch cycle.

[0031] It is the power of the energy storage at time t. When it is a positive number, it indicates discharge, and when it is a negative number, it indicates charging.

[0032] (3) Micro-turbine The micro gas turbine only considers its output limitation and climbing limitation.

[0033] (13) (14) in, for t The output of the micro-turbine in the time period, for t The running status of the generator at the moment, 1 means running, 0 means shut down; and are the maximum and minimum output of the generator respectively; and They are respectively the maximum upward and downward climbing capabilities of the generator in a single period.

[0034] (4) Distributed photovoltaic (15) (16) in, for t The active power output of distributed photovoltaic at all times, represents the expected value of photovoltaic output, is the maximum output active power of distributed photovoltaics; is a random variable, indicating the uncertainty of distributed photovoltaic output.

[0035] Since the optimization model cannot directly handle constraints containing random variables, this embodiment uses chance constraints to transform constraints containing random variables into deterministic constraints. According to empirical results, the prediction error of distributed photovoltaic output can be well modeled using parameterized normal distribution, that is, random variables ,in for t The covariance matrix of the distributed photovoltaic prediction error at the moment. Therefore, the operation constraints (15) (16) containing random variables can be expressed in the form of chance constraints: (17) in, Indicates the risk level of unit overload, usually a smaller value (such as 5%) is selected.

[0036] Since the chance constraint formula (17) cannot be solved directly, this embodiment uses the moment information of the random variable to transform the chance constraint into a processable linear constraint to facilitate the solution of the model. The specific derivation process is as follows: Formula (17) can be transformed into the following form: (18) set up ,in satisfy , According to the one-sided Chebyshev inequality principle, we can get: (19) (20) The infimum calculation depends on , which is equivalent to the square of the distance from the origin of the hyperplane. for: (twenty one) Then we can get: (twenty two) in, .

[0037] Through the above processing, the complex probabilistic optimization problem can be simplified into a convex optimization model that can be efficiently solved, thereby improving the efficiency.

[0038] In summary, the modeling of the internal resources of the virtual power plant includes constraints (5)~(14) and (22), which respectively constitute the high-dimensional expression of the feasible domain of various types of resources aggregated by the virtual power plant.

[0039] All the above-mentioned resources can be expressed in mathematical form as formula (1).

[0040] S102: Based on the Chino polyhedron approximation principle, a search direction set is selected according to the power-energy characteristics of the distributed resource at different time sequences.

[0041] This embodiment is based on the Chino polyhedron approximation principle and proposes a Chino polyhedron aggregation method for time-series coupled direction search. The basic model is as follows: (twenty three) (twenty four) (25) (26) in, are the coordinates of the center point of the polyhedron; is the search direction matrix, M Direction vector Composition, which means that the polyhedron extends in all directions from the center point; is the extension length vector in each direction, which is composed of the maximum extension length limit; represents the 2-norm operation; is the i-th direction vector, is the Mth extension length vector.

[0042] For selecting the same search direction matrix Since the extension directions of the Chino polyhedra are consistent, the Minkowski sum process of these Chino polyhedra can be simplified to the XOR operation between each Chino polyhedron: (27) (28) in, The number of distributed resource entities on a certain node; is the coordinate of the center point of the aggregated polyhedron; is the coordinate of the sth center point; and For the aggregate polyhedron m The sum of the total extension length and the extension length limit in each search direction, and Before aggregation s The polyhedron in m The sum of the extension length and the extension length limit in each search direction.

[0043] By selecting different search directions, the Minkowski sum process of the polyhedron is transformed into a simple linear superposition, which provides the feasibility for the aggregation of feasible domains in high-dimensional space of massive multivariate distributed resources.

[0044] Faced with the construction of various distributed resource feasible domain models, the selection of the extension direction matrix is ​​crucial to the accuracy of using the Qino polyhedron to approximate the original feasible domain of the resource.

[0045] When constructing the search direction vector, this embodiment extracts the coupling and complementary characteristics of power-energy of various types of resources on a time scale based on the natural properties and regulatory elasticity of the internal resources of the virtual power plant, fully taps the spatiotemporal synergy potential of multiple types of resources in the virtual power plant, and enhances the robustness and economy of the virtual power plant in dealing with complex working conditions.

[0046] (1) Basic timing constraints It includes 2N search directions, each of which represents the power operation boundary of the common coupling point at a certain moment, and is used to approximate the extension direction of the aggregated resources at each decision moment: (29) in, is the Kronecker symbol, when i=j 1 if the value is 0, otherwise it is 0. (2) Time coupling constraints It includes N-1 search directions, each of which can represent the difference between the power of the next time period and the current time period, and is used to approximate the extension direction of the power-time coupling constraint of the aggregated resources: (30) (3) Power-energy coupling constraints It includes N-1 search directions, representing the sum of energy changes in a certain time period, and is expressed as the sum of power values ​​over a certain period of time, which is used to approximate the extension direction of the power-energy coupling constraint of the aggregated resources: (31) For temperature control load constraints such as air conditioning (7), the energy change is not the sum of the power work done in each period. When the power is 0, there is also a linear relationship between the energy variables at adjacent moments. Its mathematical expression is: (32) (33) in, is the energy variable in period t, is the power variable in period t, is the external influencing parameter in period t, and All are proportional parameters.

[0047] According to formula (33), the general form of the power-energy coupling constraint search direction vector can be obtained , specifically expressed as follows: (34) in, , , are all search direction sets, i and j represent the rows and columns in the matrix, N represents the matrix dimension, and h is the introduced auxiliary variable.

[0048] It should be noted that N is determined by the time scale considered by the model. If 1 hour is considered as the decision step, N is 24, and if 15 minutes is considered, N is 96.

[0049] In summary, we can get the search direction matrix of each type of resource, among which the search direction matrix of air conditioning load and distributed energy storage is and The distributed photovoltaic search direction matrix is ​​composed of The micro-turbine search direction matrix is ​​composed of and By constructing a search direction matrix for various resources, the coupling problem of time and space can be solved.

[0050] S103: Taking maximizing the sum of the extension lengths in each dimensional direction as the optimization goal, an optimal linear decision model for aggregation parameters is established; the model is solved to obtain an optimal Chino polyhedron approximate feasible domain.

[0051] For each resource type, the calculated Chino polyhedron Z and the original feasible domain space P All high-dimensional spaces are generated from the same data source. In order to make the calculated approximate Chino polyhedron as close to the original feasible domain as possible, this embodiment adopts a high-dimensional space generation method based on the Farkas condition, and solves the linear programming (LP) problem corresponding to the method to achieve the solution of the optimal aggregation parameters.

[0052] The specific process is as follows: The definition of the Chino polyhedron is that formula (23) can be rewritten as: (33) in, and is an auxiliary variable, which represents the transformed decision coefficient, as follows: (34) in, for dimensional identity matrix.

[0053] In order to , Z The coordinates of any point in x Satisfy the original feasible region P Each constraint , i for P According to the Farkas lemma, there exists a non-negative Farkas multiplier ,make P and Z Satisfy the coefficient matching conditions and right-hand side constraints: (35) (36) Substituting formula (34) into the equation, we get No. j Dimensions: (37) (38) Then we can sort it out: (39) In order to make the calculated approximate Chino polyhedron as close to the original feasible domain as possible, the objective function is set to maximize the sum of the extension lengths in each dimensional direction. .

[0054] In summary, the optimal linear decision model for aggregation parameters is summarized as follows: (40) Where L is the sum of the extension lengths in each dimension, , is the maximum extension length vector in each dimension, is a column vector, each dimension has its own maximum extension length; is the search direction matrix; are the coordinates of the center point of the polyhedron; is a non-negative Farkas multiplier, is the control variable parameter matrix, is a vector of constrained constants, is the original feasible domain of distributed resources, i for P Identification of the constraints in It is j non-negative Farkas multipliers in dimensions, It is the Nth+ j non-negative Farkas multipliers in dimensions, is the control variable parameter matrix j dimensions, is the coordinate of the center point of the polyhedron j dimensions.

[0055] The optimal aggregation parameters are solved by solving the linear programming (LP) problem corresponding to the optimal linear decision model of the aggregation parameters.

[0056] By solving the problem, we can get the optimal center point coordinates of the aggregated polyhedron. , and the total extension length of the aggregated polyhedron in each search direction , aggregate polyhedron m The sum of the extension length limits in the search directions , thus obtaining the optimal Qino polyhedron approximate feasible domain.

[0057] This embodiment establishes an optimal linear decision model for aggregation parameters, and obtains the optimal Chino polyhedron approximate feasible domain through model solution, thereby achieving accurate and efficient solution of the feasible domain of aggregation resources. S104: Calculate the optimal Chino polyhedron approximate feasible domain of each type of resource at the node; perform Minkowski summation to form the Chino polyhedron approximate aggregate feasible domain of each type of resource at the node.

[0058] The same method as in S101-S103 is used to calculate the optimal Chino polyhedron approximate feasible domain of other types of distributed resources of the node; the optimal Chino polyhedron approximate feasible domain of all distributed resources is Minkowski summed to form the Chino polyhedron approximate aggregate feasible domain of each type of resource of the node.

[0059] The method of this embodiment transforms the process of feasible domain aggregation into a simple superposition in several directions, which solves the problem of difficulty in solving the Minkowski sum of polyhedrons and provides feasibility for feasible domain aggregation in high-dimensional space of massive multi-distributed resources.

[0060] S105: Characterization of the output boundary of the common coupling point of the virtual power plant considering the spatial coupling characteristics.

[0061] The external aggregation characteristics of the virtual power plant should project the high-dimensional model with high coupling characteristics as the external power characteristics of the point of common coupling (PCC). However, the approximate feasible domain obtained by the above process ignores the impact of the network space constraints of the distribution system where the virtual power plant is located. In fact, the network flow will produce a large deviation in the calculation of the feasible domain of the common coupling point of the virtual power plant. Therefore, this embodiment, based on the optimal feasible domain obtained by the above solution, considers the spatial coupling characteristics of the virtual power plant and corrects the aggregation feasible domain.

[0062] 1. Constraints (1) Constraints on equivalent adjustable output of internal resources The output upper and lower limits of the virtual power plant's internal air conditioning load, distributed photovoltaics, energy storage, micro-turbines and other resources are calculated. The boundaries are: (41) For nodes n At , the feasible domain of the Chino polyhedron approximation is expressed as follows: (42) in," " represents the parameters corresponding to any type of resources in the virtual power plant, including air conditioning load, distributed photovoltaic, energy storage, and micro-turbines; represents the power of the center point of the feasible region of a certain type of resource, Indicates the extension length of the feasible domain of this type of resource, Indicates the operating power of this type of resource; is the extended length, The upper limit of the extension length.

[0063] (2) Participation factor constraints The external power of PCC is distributed among various types of resource units at each node according to the participation factors. Since the power adjusted by the units needs to match the external power of PCC, the sum of the participation factors is 1.

[0064] (43) in, represents the column vector of participation factors, Represents a column vector whose elements are all 1.

[0065] (3) Constraints on the maximum extension of the feasible domain of the common coupling point (44a) (44b) (45a) (45b) in, and Respectively represent nodes n No. s The physical upper and lower limits of the output of a distributed resource. and Respectively represent nodes n The maximum upward and downward capacity of a certain type of distributed resource connected to the network; and Point of Common Coupling o The maximum extension of the feasible domain; represents the participation factor of a resource, represents the power of the center point of the feasible region of a certain type of resource, Indicates the extension length of the feasible domain of this type of resource, Indicates the number of distributed resource entities on a node.

[0066] (4) Internal network constraints of virtual power plants The network flow constraints of the distribution system where the virtual power plant is located mainly consider line flow constraints, node voltage constraints, node power balance constraints and node power adjustment range constraints.

[0067] (46) (47) (48) (49) (50) in, , , , and The two end nodes are , Active power, reactive power, conductance, susceptance and maximum apparent power of the line; For Node n The square value of the voltage; and They are the upper and lower limits of the voltage square respectively; For Node n The voltage phase angle; , , Node n Active power, reactive power and power factor angle; For Node n The adjustment amount of a certain type of resource output; For Node n The set of upper connected branches; For Node n 1 The square value of the voltage, For Node n 2 The square value of the voltage, For Node n 1 The phase angle, For Node n 2 The phase angle, For Node n On a certain type of resource output, The two end nodes are , Active power of the line.

[0068] (4) Coupling point power constraint (51) (52) in, Common Coupling Point o Output active power; For each node Common Coupling Point o Output active power; For the common coupling point o End nodes of connected lines n Flow to the point of common coupling o The active power.

[0069] 2. Objective Function The goal of this embodiment is to maximize the volume of the feasible domain of the virtual power plant at the common coupling point o in the distribution system. In order to avoid the nonlinear problem caused by the volume formula of the hyperpolyhedron, the objective function adopts the maximum sum of the lengths of the maximum extension intervals of the feasible domain of the common coupling point, which approximately replaces the objective function form of maximizing the volume of the feasible domain in high-dimensional space.

[0070] (54) in, and Point of Common Coupling o The lower and upper limits of the maximum extension of the feasible domain; t refers to a certain time period, and N refers to the number of decision moments for day-ahead scheduling (when one decision moment is one hour, N is 24).

[0071] 3. Model Linearization Formulas (44) and (47) in the above model are nonlinear constraints. This embodiment uses the McCormick envelope method and the inscribed circle linearization method to linearize these two sets of constraints respectively.

[0072] The constraint condition of formula (44) contains bilinear variables, and the bilinear terms are all continuous variables, so the big M method cannot be directly applied to achieve linear processing. It is proposed to relax the constraint (44) by constructing the McCormick envelope. However, the bilinear quantity and There are no physical upper and lower limits, so the McCormick envelope method cannot be directly applied. and The relaxation boundary of .

[0073] (55) in, and Satisfy the following formula: (56) in, Indicates that the node o A collection of network nodes other than .

[0074] Subsequently, formulas (44a) and (44b) are transformed into linear constraint groups (57a) and (57b) respectively: (57a) (57b) For the constraint condition of formula (47), according to the inscribed circle linearization method, this paper converts the outer circumference of the constraint area represented by formula (47) into K Divide equally, connect each point in turn, and obtain the inscribed right K In other words, the inscribed regular K The area enclosed by the polygon approximately replaces the area represented by the original constraint, so formula (47) can be transformed into the following formula.

[0075] (58) in, (59) Through linearization processing, the difficulty of solving the model can be reduced, and the MILP problem that can be directly solved by commercial solvers can be obtained, and then the corrected aggregate feasible domain can be obtained, which improves the solution efficiency.

[0076] Embodiment 2 In one or more embodiments, a virtual power plant aggregation system considering time-space coupling characteristics is disclosed, including: A search direction selection module selects a single type of distributed resource accessed by a node in the virtual power plant area network, and selects a search direction set based on the power-energy characteristics of the distributed resource at different time sequences based on the Chino polyhedron approximation principle; The feasible domain solving module establishes an optimal linear decision model for aggregation parameters with the optimization goal of maximizing the sum of the extension lengths in each dimensional direction; solves the model to obtain the optimal Chino polyhedron approximate feasible domain of the distributed resource; A feasible domain aggregation module calculates the optimal Chino polyhedron approximate feasible domain of each type of distributed resources of the network node, and converts the Minkowski sum process of the polyhedron into a linear superposition by selecting different search directions to obtain the optimal Chino polyhedron approximate feasible domain of the network node; The feasible domain correction module considers the spatial coupling characteristics of the virtual power plant, takes the maximum sum of the lengths of the maximum extension intervals of the feasible domain of the common coupling point as the goal, corrects the optimal Chino polyhedron approximate feasible domain, and obtains the final aggregate feasible domain.

[0077] It should be noted that the specific implementation of the above modules is the same as that in Example 1 and will not be described in detail.

[0078] Embodiment 3 In one or more embodiments, a terminal device is disclosed, which includes a processor and a memory, the processor being used to implement instructions; the memory being used to store multiple instructions, the instructions being suitable for being loaded by the processor and executing the virtual power plant aggregation method considering the time-space coupling characteristics described in Example 1.

[0079] Embodiment 4 In one or more embodiments, a computer-readable storage medium is disclosed, in which a plurality of instructions are stored, wherein the instructions are suitable for being loaded by a processor of a terminal device and executing the virtual power plant aggregation method considering the time-space coupling characteristics described in Example 1.

[0080] Embodiment 5 In one or more embodiments, a computer program product is disclosed, including a computer program / instruction, which, when executed by a processor, implements the virtual power plant aggregation method considering the time-space coupling characteristics described in Example 1.

[0081] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A virtual power plant aggregation method considering time-space coupling characteristics, characterized in that: include: Select a single type of distributed resource accessed by a node in the virtual power plant area network, and select a set of search directions based on the power-energy characteristics of the distributed resource at different time sequences based on the Chino polyhedron approximation principle; Taking maximizing the sum of the extension lengths in each dimensional direction as the optimization goal, an optimal linear decision model for aggregation parameters is established; the model is solved to obtain the optimal Chino polyhedron approximate feasible domain of the distributed resource; Calculating the optimal Chino polyhedron approximate feasible domain of each type of distributed resource of the network node, and converting the Minkowski sum process of the polyhedron into a linear superposition by selecting different search directions to obtain the optimal Chino polyhedron approximate feasible domain of the network node; Considering the spatial coupling characteristics of virtual power plants, the optimal Chino polyhedron approximate feasible domain is modified with the goal of maximizing the sum of the maximum extension interval lengths of the feasible domain of the common coupling point to obtain the final aggregate feasible domain.

2. A virtual power plant aggregation method considering time-space coupling characteristics as claimed in claim 1, characterized in that: The distributed resources include at least: air conditioning load, distributed energy storage, micro-turbines and distributed photovoltaics; for distributed photovoltaics, the constraints containing random variables are converted into deterministic constraints by using chance constraints, and the moment information of random variables is used to convert the chance constraints into processable linear constraints.

3. A virtual power plant aggregation method considering time-space coupling characteristics as claimed in claim 1, characterized in that: Based on the Chino polyhedron approximation principle, a search direction set is selected according to the power-energy characteristics of the distributed resources at different time sequences, specifically: The search direction set of air conditioning load and distributed energy storage is composed of and The distributed photovoltaic search direction set is composed of The search direction set of the micro-turbine is composed of and constitute; in, It includes 2N search directions, each of which represents the power operation boundary of the common coupling point at a certain moment, and is used to approximate the extension direction of the aggregated resources at each decision moment; It includes N-1 search directions, each of which can represent the difference between the power of the next time period and the current time period, and is used to approximate the extension direction of the power-time coupling constraint of the aggregated resources: It includes N-1 search directions, representing the sum of energy changes in a certain time period, and is expressed as the sum of power values ​​over a certain period of time, which is used to approximate the extension direction of the power-energy coupling constraint of the aggregated resources; N is the time scale.

4. A virtual power plant aggregation method considering time-space coupling characteristics as claimed in claim 1, characterized in that: The optimal linear decision model for aggregation parameters is specifically: ； Where L is the sum of the extension lengths in each dimension, , The maximum extension length vector in each dimension direction, is the search direction matrix; are the coordinates of the center point of the polyhedron; is a non-negative Farkas multiplier, is the control variable parameter matrix, is a vector of constrained constants, is the original feasible domain of distributed resources, i for P Identification of the constraints in It is j non-negative Farkas multipliers in dimensions, It is the Nth+ j non-negative Farkas multipliers in dimensions, is the control variable parameter matrix j dimensions, is the coordinate of the center point of the polyhedron j dimensions.

5. A virtual power plant aggregation method considering time-space coupling characteristics as claimed in claim 1, characterized in that: Considering the spatial coupling characteristics of the virtual power plant, the optimal Qino polyhedron approximate feasible domain is modified with the goal of maximizing the sum of the maximum extension interval lengths of the feasible domain of the common coupling point, specifically: The objective function is constructed with the constraints of the equivalent adjustable output of internal resources, participation factor constraints, the maximum extension of the feasible domain of the common coupling point, the internal network constraints of the virtual power plant and the power constraints of the coupling point as the constraints, and the maximum sum of the length of the feasible domain of the common coupling point as the goal. The nonlinear constraints in the constraints are linearized; the optimization model consisting of the objective function and constraints is solved using a commercial solver to obtain the revised aggregate feasible domain.

6. A virtual power plant aggregation method considering time-space coupling characteristics as claimed in claim 5, characterized in that: The objective function is specifically: ； in, and Point of Common Coupling o The lower and upper limits of the maximum extension of the feasible domain; t refers to a certain time period, and N refers to the number of decision moments for day-ahead scheduling.

7. A virtual power plant aggregation system considering time-space coupling characteristics, characterized in that: include: A search direction selection module selects a single type of distributed resource accessed by a node in the virtual power plant area network, and selects a search direction set based on the power-energy characteristics of the distributed resource at different time sequences based on the Chino polyhedron approximation principle; The feasible domain solving module establishes an optimal linear decision model for aggregation parameters with the optimization goal of maximizing the sum of the extension lengths in each dimensional direction; solves the model to obtain the optimal Chino polyhedron approximate feasible domain of the distributed resource; A feasible domain aggregation module calculates the optimal Chino polyhedron approximate feasible domain of each type of distributed resources of the network node, and converts the Minkowski sum process of the polyhedron into a linear superposition by selecting different search directions to obtain the optimal Chino polyhedron approximate feasible domain of the network node; The feasible domain correction module considers the spatial coupling characteristics of the virtual power plant, takes the maximum sum of the lengths of the maximum extension intervals of the feasible domain of the common coupling point as the goal, corrects the optimal Chino polyhedron approximate feasible domain, and obtains the final aggregate feasible domain.

8. A terminal device, comprising a processor and a memory, wherein the processor is used to implement instructions; and the memory is used to store multiple instructions, characterized in that: The instructions are suitable for being loaded by a processor and executing the virtual power plant aggregation method considering the time-space coupling characteristics as described in any one of claims 1-6.

9. A computer-readable storage medium storing a plurality of instructions, characterized in that: The instructions are suitable for being loaded by a processor of a terminal device and executing the virtual power plant aggregation method considering the time-space coupling characteristics as described in any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the virtual power plant aggregation method considering the time-space coupling characteristics as described in any one of claims 1-6 is implemented.

Images