Available transfer capability evaluation method and apparatus for multi-region power system

Abstract

An available transfer capability evaluation method and apparatus for a multi-region power system, belonging to the technical field of new energy grid connection. The method comprises the steps: in view of multi-dimensional uncertainty of new energy output and a load demand, on the basis of a conditional generative adversarial network method, determining a typical daily source-load scenario set; constructing an initial operation point set on the basis of the typical daily source-load scenario set, and determining a limit operation point of a multi-region power system; on the basis of the initial operation point set and the limit operation point, constructing an ATC evaluation model on the basis of safety indexes of multi-region power grid operation; and, on the basis of the ATC evaluation model and the typical daily source-load scenario set, determining available transfer capability probability distribution of the multi-region power system.

Description

Method and device for evaluating available transmission capacity of multi-regional power system

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on March 13, 2024, with application number 202410281521.2, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to a method and device for evaluating the available transmission capacity of a multi-regional power system, belonging to the field of new energy grid connection technology. Background Art

[0003] In the power market environment, achieving economically optimal operation while ensuring system security constraints has become a pressing issue for grid managers and all participants in the power market. Available Transfer Capability (ATC) represents the remaining power transmission capacity required to ensure safe and stable operation of the power system. ATC approximately measures the safety and stability margin of the power grid at its current operating point. Therefore, ATC is not only an important basis for power market participants to trade transmission rights, but also a boundary condition for power system expansion planning. In this context, proposing an ATC assessment method that coordinates safety and economics is of great significance for ensuring the rational operation of the power market and improving power grid utilization efficiency.

[0004] In early research, domestic and foreign scholars proposed various calculation methods for the ATC assessment problem with deterministic source and load parameters, including the linear distribution factor method, the continuous power flow method, the repeated power flow method, and the optimal power flow (OPF) method. The OPF method models the ATC calculation as a mathematical optimization problem with the objective function of maximizing the transmission power of inter-regional channels and the constraints of grid power balance and safety and stability criteria. By solving the OPF optimization problem, the ATC of the system under a certain operating state can be obtained. The new power system with renewable energy as the main body is characterized by high renewable energy penetration and complex load structure. Therefore, the uncertainty of the output of large-scale wind farms and photovoltaic power stations, as well as the uncertainty of load, makes the transmission power between different regions uncertain, increasing the difficulty of ATC assessment. In response to the uncertainty of renewable energy and load, domestic and foreign scholars have proposed methods based on robust optimization, interval optimization, and optimization with chance constraints.

[0005] Affected by the time series uncertainty of new energy and load, the ATC of the new power system actually has the characteristics of time series probability distribution. Therefore, the key difficulty of ATC evaluation lies in the generation of source-load random scenarios and the construction of ATC optimization models. Traditional methods based on robust optimization or interval optimization have a small amount of computation, but only give the distribution range of random variables and fail to fully characterize the probability distribution of ATC. On the other hand, the uncertainty of new energy and load in large power systems has high-dimensional uncertainty characteristics. Although the traditional Monte Carlo simulation method can generate a large number of random scenarios, it has a large amount of computation and is difficult to effectively classify and screen the scenarios, which brings certain difficulties to the probability distribution evaluation of ATC. Therefore, how to comprehensively characterize the probability distribution of ATC is a technical problem that needs to be solved urgently by those skilled in the art.

[0006] Summary of the Invention

[0007] One purpose of the embodiments of the present application is to provide a temperature control device that can quickly adjust the temperature of a component to be temperature-controlled and has high heat exchange efficiency.

[0008] In order to solve the above problems, this application proposes a method and device for evaluating the available transmission capacity of a multi-regional power system, which can accurately and efficiently calculate the ATC of the renewable energy transmission section online, helping to improve the renewable energy absorption capacity.

[0009] The technical solution adopted by this application to solve its technical problems is:

[0010] In a first aspect, an embodiment of the present application provides a method for evaluating available transmission capacity of a multi-regional power system, characterized by comprising the following steps:

[0011] Considering the multi-dimensional uncertainty of renewable energy output and load demand, a set of typical daily source-load scenarios is determined based on the conditional generative adversarial network method;

[0012] Constructing an initial operating point set based on the typical daily source-load scenario set, and determining a limit operating point of the multi-regional power system;

[0013] According to the initial operating point set and the limit operating point, an ATC evaluation model is constructed based on the safety index of multi-regional power grid operation;

[0014] According to the ATC evaluation model and a set of typical daily source-load scenarios, a probability distribution of available transmission capacity of the multi-regional power system is determined.

[0015] As a possible implementation of this embodiment, determining a typical daily source-load scenario set based on a conditional generative adversarial network method includes:

[0016] Based on the conditional generative adversarial network method, a deep neural network model is used to characterize the nonlinear relationship generator and the classification signal discriminator, and conditional information is transmitted as an input layer to the classification signal discriminator and the nonlinear relationship generator. The conditional information includes: historical meteorological data with time attributes, spatial characteristics of the power system, and wind power plant output and load demand characteristics;

[0017] According to the historical data of new energy output and load demand, a typical daily source-load scenario set S is constructed. 0 : S 0 ={S W ,S D},

[0018] Where: represents the wind farm output at access node i, S W Indicates the system A collection of represents the wind farm load at access node i, S D Indicates the system W represents the set of wind farm nodes in the system, and D represents the set of load nodes in the system.

[0019] As a possible implementation of this embodiment, constructing an initial operating point set based on the typical daily source-load scenario set and determining the limit operating point of the multi-region power system includes:

[0020] According to the initial operating point set and based on the safety index of the multi-regional power system operation, the exchange power between the multi-regional power systems is continuously increased by adjusting the power generation and load power, so as to obtain the extreme operating point of the multi-regional power system under extreme operating conditions.

[0021] As a possible implementation of this embodiment, the ATC evaluation model is constructed based on the initial operating point set and the limit operating point and the safety index of the multi-regional power grid operation, including:

[0022] The objective function is established based on minimizing the overall generation cost of the power system and maximizing the ATC between different regions:

[0023] Where, and are the active power output of the thermal generator at node i under the limit state and its value at the initial operating point; G is the set of power generation nodes to be adjusted at the sending end; c i is the unit power generation cost of unit i, α and β are the weight coefficients of multi-objective optimization;

[0024] Constructing a power system benchmark operating state model, wherein the power system benchmark operating state model includes operating parameters of the grid structure, startup mode, load power, power growth mode, and safety constraints;

[0025] By adjusting the power generation and load power, the exchange power between power grid areas is continuously increased until the safety constraint conditions are exceeded. The limit operation point of the power grid is obtained and the ATC evaluation model is constructed.

[0026] As a possible implementation of this embodiment, the security constraint conditions include:

[0027] (1) The active and reactive power balance constraints of nodes under the power system benchmark operation state are:

[0028] Where, P i,t , Q i,t are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, λ i is the ratio of reactive load to active load;

[0029] The linearized power flow constraint under the grid benchmark operation state is:

[0030] The linear equation can be used to simultaneously solve the amplitude and phase angle of the node voltage and the equivalent admittance of the power flow equation. and As shown below:

[0031] The capacity constraint of the line under the grid benchmark operation state is:

[0032] The node voltage constraint under the grid benchmark operation state is: V min ≤V i,t ≤V max ,

[0033] The output constraint of the generator group under the grid benchmark operating state is:

[0034] The above formula indicates that at time t, the active output of the nth generator must be between its upper and lower limits;

[0035] The following formulas respectively represent the increase or decrease in the output of the generator set per unit time:

[0036] Where, δ i,t 、V i,t is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node; and Represent the upper and lower limits of power generation of each unit, and r is the upper limit of ramp rate;

[0037] (2) The active and reactive power balance constraints of the nodes in the power grid under the extreme operation state are:

[0038] The linearized power flow constraint of the power grid under the extreme operating state is:

[0039] The capacity constraint of the power line under the extreme operation state is:

[0040] The node voltage constraint of the power grid under the extreme operation state is:

[0041] The output constraint of the generator group under the extreme operation state of the power grid is:

[0042] Where, are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node; and They represent the upper and lower limits of power generation of each unit respectively, r is the upper limit of ramp rate; SR represents the sending end area in the system, and SK represents the receiving end area in the system.

[0043] As a possible implementation of this embodiment, determining the probability distribution of the available transmission capacity of the multi-regional power system based on the ATC evaluation model and the typical daily source-load scenario set includes:

[0044] The nonlinear relationship generator receives random noise and conditional values ​​of the multi-regional power system as input, and the classification signal discriminator receives a wind power output curve or a load curve and the conditional value as input, and decomposes the wind power output data or load data of each plant into a matrix form;

[0045] The noise, conditional values ​​and real data of the multi-regional power system are respectively input into the classification signal discriminator and the nonlinear relationship generator, and the false sample data generated by the nonlinear relationship generator and the real data are input into the classification signal discriminator for discrimination. The classification signal discriminator outputs the Wasserstein distance as the discrimination result to determine the probability distribution of the available transmission capacity of the multi-regional power system.

[0046] As a possible implementation of this embodiment, the mean value of the probability distribution of the available transmission capacity of the multi-regional power system is:

[0047] Where ATC s is the daily available transmission capacity of the power system under the typical daily scenario s, t is the daily time period, T is the set of daily time periods, ATC t,s is the available transmission capacity of the system at time t in a typical day scenario s.

[0048] In a second aspect, an embodiment of the present application provides a device for evaluating available transmission capacity of a multi-regional power system, comprising:

[0049] The source-load scenario set determination module is used to consider the multi-dimensional uncertainty of renewable energy output and load demand, and determine the typical daily source-load scenario set based on the conditional generative adversarial network method;

[0050] A limit operating point determination module is used to construct an initial operating point set based on the typical daily source-load scenario set and determine the limit operating point of the multi-region power system;

[0051] An evaluation model building module is used to build an ATC evaluation model based on the initial operating point set and the limit operating point and the safety index of the multi-region power grid operation;

[0052] The available transmission capacity evaluation module is used to determine the probability distribution of the available transmission capacity of the multi-regional power system based on the ATC evaluation model and a set of typical daily source-load scenarios.

[0053] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a program stored in the memory and executable on the processor. When the electronic device is running, the processor executes the program to implement any of the steps of the above-mentioned method for evaluating the available transmission capacity of a multi-regional power system.

[0054] In a fourth aspect, an embodiment of the present application provides a storage medium having a program stored thereon, which, when executed by a processor, executes the steps of any of the above-mentioned methods for evaluating the available transmission capacity of a multi-regional power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] FIG1 is a flow chart showing a method for evaluating available transmission capacity of a multi-regional power system according to an exemplary embodiment;

[0056] FIG2 is a block diagram of an apparatus for evaluating available transmission capacity of a multi-regional power system according to an exemplary embodiment;

[0057] FIG3 is a schematic diagram showing the structure of a conditional adversarial generative network according to an exemplary embodiment. DETAILED DESCRIPTION

[0058] The present application will be further described below with reference to the accompanying drawings and embodiments:

[0059] In order to clearly illustrate the technical features of the present solution, the present application is described in detail below through specific implementation methods and in conjunction with the accompanying drawings. The disclosure below provides many different embodiments or examples for realizing different structures of the present application. In order to simplify the disclosure of the present application, the components and settings of specific examples are described below. In addition, the present application may repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplicity and clarity and does not itself indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the accompanying drawings are not necessarily drawn to scale. The present application omits descriptions of well-known components and processing technologies and processes to avoid unnecessarily limiting the present application.

[0060] As shown in FIG1 , an embodiment of the present application provides a method for evaluating available transmission capacity of a multi-regional power system, comprising the following steps:

[0061] Considering the multi-dimensional uncertainty of renewable energy output and load demand, a set of typical daily source-load scenarios is determined based on the conditional generative adversarial network method;

[0062] Constructing an initial operating point set based on the typical daily source-load scenario set, and determining a limit operating point of the multi-regional power system;

[0063] According to the initial operating point set and the limit operating point, an ATC evaluation model is constructed based on the safety index of multi-regional power grid operation;

[0064] According to the ATC evaluation model and a set of typical daily source-load scenarios, a probability distribution of available transmission capacity of the multi-regional power system is determined.

[0065] As a possible implementation of this embodiment, determining a typical daily source-load scenario set based on a conditional generative adversarial network method includes:

[0066] Based on the conditional generative adversarial network method, a deep neural network model is used to characterize the nonlinear relationship generator and the classification signal discriminator, and conditional information is transmitted as an input layer to the classification signal discriminator and the nonlinear relationship generator. The conditional information includes: historical meteorological data with time attributes, spatial characteristics of the power system, and wind power plant output and load demand characteristics;

[0067] According to the historical data of new energy output and load demand, a typical daily source-load scenario set S is constructed. 0 : S 0 ={S W ,S D},

[0068] Where: represents the wind farm output at access node i, S W Indicates the system A collection of represents the wind farm load at access node i, S D Indicates the system W represents the set of wind farm nodes in the system, and D represents the set of load nodes in the system.

[0069] As a possible implementation of this embodiment, constructing an initial operating point set based on the typical daily source-load scenario set and determining the limit operating point of the multi-region power system includes:

[0070] According to the initial operating point set and based on the safety index of the multi-regional power system operation, the exchange power between the multi-regional power systems is continuously increased by adjusting the power generation and load power, so as to obtain the extreme operating point of the multi-regional power system under extreme operating conditions.

[0071] As a possible implementation of this embodiment, the ATC evaluation model is constructed based on the initial operating point set and the limit operating point and the safety index of the multi-regional power grid operation, including:

[0072] The objective function is established based on minimizing the overall generation cost of the power system and maximizing the ATC between different regions:

[0073] Where, and are the active power output of the thermal generator at node i under the limit state and its value at the initial operating point; G is the set of power generation nodes to be adjusted at the sending end; c i is the unit power generation cost of unit i, α and β are the weight coefficients of multi-objective optimization;

[0074] Constructing a power system benchmark operating state model, wherein the power system benchmark operating state model includes operating parameters of the grid structure, startup mode, load power, power growth mode, and safety constraints;

[0075] By adjusting the power generation and load power, the exchange power between power grid areas is continuously increased until the safety constraint conditions are exceeded. The limit operation point of the power grid is obtained and the ATC evaluation model is constructed.

[0076] As a possible implementation of this embodiment, the security constraint conditions include:

[0077] (1) The active and reactive power balance constraints of nodes under the power system benchmark operation state are:

[0078] Where, P i,t , Q i,t are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, λ i is the ratio of reactive load to active load;

[0079] The linearized power flow constraint under the grid benchmark operation state is:

[0080] The linear equation can be used to simultaneously solve the amplitude and phase angle of the node voltage and the equivalent admittance of the power flow equation. and As shown below:

[0081] The capacity constraint of the line under the grid benchmark operation state is:

[0082] The node voltage constraint under the grid benchmark operation state is: V min≤V i,t ≤V max ,

[0083] The output constraint of the generator group under the grid benchmark operating state is:

[0084] The above formula indicates that at time t, the active output of the nth generator must be between its upper and lower limits;

[0085] The following formulas respectively represent the increase or decrease in the output of the generator set per unit time:

[0086] Where, δ i,t 、V i,t is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node; and Represent the upper and lower limits of power generation of each unit, and r is the upper limit of ramp rate;

[0087] (2) The active and reactive power balance constraints of the nodes in the power grid under the extreme operation state are:

[0088] The linearized power flow constraint of the power grid under the extreme operating state is:

[0089] The capacity constraint of the power line under the extreme operation state is:

[0090] The node voltage constraint of the power grid under the extreme operation state is:

[0091] The output constraint of the generator group under the extreme operation state of the power grid is:

[0092] Where, are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node; and They represent the upper and lower limits of power generation of each unit respectively, r is the upper limit of ramp rate; SR represents the sending end area in the system, and SK represents the receiving end area in the system.

[0093] As a possible implementation of this embodiment, determining the probability distribution of the available transmission capacity of the multi-regional power system based on the ATC evaluation model and the typical daily source-load scenario set includes:

[0094] The nonlinear relationship generator receives random noise and conditional values ​​of the multi-regional power system as input, and the classification signal discriminator receives a wind power output curve or a load curve and the conditional value as input, and decomposes the wind power output data or load data of each plant into a matrix form;

[0095] The noise, conditional values ​​and real data of the multi-regional power system are respectively input into the classification signal discriminator and the nonlinear relationship generator, and the false sample data generated by the nonlinear relationship generator and the real data are input into the classification signal discriminator for discrimination. The classification signal discriminator outputs the Wasserstein distance as the discrimination result to determine the probability distribution of the available transmission capacity of the multi-regional power system.

[0096] As a possible implementation of this embodiment, the mean value of the probability distribution of the available transmission capacity of the multi-regional power system is:

[0097] Where ATC s is the daily available transmission capacity of the power system under the typical daily scenario s, t is the daily time period, T is the set of daily time periods, ATC t,s is the available transmission capacity of the system at time t in a typical day scenario s.

[0098] As shown in FIG2 , an embodiment of the present application provides a device for evaluating available transmission capacity of a multi-regional power system, including:

[0099] The source-load scenario set determination module is set to consider the multi-dimensional uncertainty of renewable energy output and load demand, and determine the typical daily source-load scenario set based on the conditional generative adversarial network method;

[0100] a limit operating point determination module, configured to construct an initial operating point set based on the typical daily source-load scenario set and determine the limit operating point of the multi-region power system;

[0101] An evaluation model construction module is configured to construct an ATC evaluation model based on the initial operating point set and the limit operating point and on the safety indicators of the multi-regional power grid operation;

[0102] The available transmission capacity evaluation module is configured to determine the probability distribution of the available transmission capacity of the multi-regional power system based on the ATC evaluation model and a typical daily source-load scenario set.

[0103] The specific process of using this application to evaluate the available transmission capacity of a multi-regional power system is as follows.

[0104] Step S1: Based on the multi-dimensional uncertainty of renewable energy output and load demand, a typical daily source-load scenario set is determined through the conditional generative adversarial network method; the typical daily source-load scenario set is a scenario set of load and power source output that is highly representative of historical data and simulates multiple factors of renewable energy output and load demand using the conditional generative adversarial network model.

[0105] The conditional generative adversarial network (CGN) method uses a deep neural network model to represent complex nonlinear relationships. It determines the generator and discriminator for classifying complex signals. Conditional information is fed as input to the discriminator and generator. Both the generator and discriminator in the CGN are deep neural network models. The generator takes as input a random noise vector and conditional information, and then, through a series of nonlinear transformations and mappings, produces output samples that meet the conditions. The generator's goal is to generate data samples as realistic as possible to deceive the discriminator. The discriminator takes as input real historical data samples and conditional information, and through a series of nonlinear transformations and classification operations, determines whether the input data samples are real or fake samples generated by the generator. The discriminator's goal is to correctly distinguish between real and generated samples. Conditional information mainly comes in three categories: 1) historical meteorological data with temporal attributes; 2) spatial features such as site location, terrain, and landforms; and 3) wind farm output and load demand characteristics. Conditional information is fed as input to the discriminator and generator.

[0106] The grid structure (including component parameters and disconnection status), initial startup method, and load power constitute a balanced power flow solution point, known as the initial operating point. The initial operating point of the ATC model is a set of typical scenarios for a series of parameters. A comprehensive ATC evaluation requires constructing a set of typical scenarios for both renewable energy output and load demand based on extensive historical data on renewable energy output and load.

[0107] Based on the historical data of renewable energy output and load demand, a typical daily source-load scenario set is constructed as follows: S 0 ={S W ,S D},

[0108] Where: represents the wind farm output at access node i, S W Indicates the system A collection of represents the wind farm load at access node i, S D Indicates the system W represents the set of wind farm nodes in the system, and D represents the set of load nodes in the system.

[0109] Step S2: constructing an initial operating point set based on the typical daily source-load scenario set, and determining the limit operating point of the power system.

[0110] The difference between the inter-regional exchange power at the initial operating point and the extreme operating point is the ATC. Based on the initial operating point set and the safety indicators of multi-regional power system operation, the exchange power between the multi-regional power systems is continuously increased by adjusting generation and load power to obtain the extreme operating point of the power system under extreme operating conditions.

[0111] In a specific embodiment of the present application, the power system satisfies node balance constraints, power flow constraints, node voltage constraints and thermal power unit output constraints under extreme operating conditions.

[0112] Optionally, the active and reactive power balance constraints of nodes under the power system benchmark operation state are:

[0113] Where, P i,t , Q i,t are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, λ i It is the ratio of reactive load to active load.

[0114] The linearized power flow constraint under the grid benchmark operation state is:

[0115] The amplitude and phase angle of the node voltage can be solved simultaneously through the linear equation. The equivalent admittance of the power flow equation is as follows:

[0116] The capacity constraint of the line under the grid benchmark operation state is:

[0117] The node voltage constraint under the grid benchmark operation state is: V min ≤V i,t ≤V max ,

[0118] The output constraint of the generator group under the grid benchmark operating state is:

[0119] Among them, this formula indicates that the active output of the nth generator must be between its upper and lower limits at time t. The following formulas respectively represent the increase or decrease in the output of the generator set per unit time:

[0120] Where, δ i,t 、V i,t is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node. and They represent the upper and lower limits of power generation of each unit respectively, and r is the upper limit of the ramp rate.

[0121] The power system still meets the node balance constraints, power flow constraints, node voltage constraints and thermal power unit output constraints under extreme operating conditions.

[0122] The active and reactive power balance constraints of the nodes in the power grid under the extreme operation state are:

[0123] The linearized power flow constraint of the power grid under the extreme operating state is:

[0124] The capacity constraint of the power line under the extreme operation state is:

[0125] The node voltage constraint of the power grid under the extreme operation state is:

[0126] The output constraint of the generator group under the extreme operation state of the power grid is:

[0127] Where, are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, λi is the ratio of reactive load to active load, is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min The upper and lower limits of the voltage of each PQ node. and Represent the upper and lower limits of the power generation of each unit, and r is the upper limit of the ramp rate. SR represents the sending end area in the system, and SK represents the receiving end area in the system. The output constraint formula of the generator unit under the extreme operating state of the power grid represents the changing relationship between the generator and the load under the base state and the extreme state. Since ATC usually evaluates the potential for increasing transmission capacity between regions, it is assumed that the incremental power generation output between the extreme operating point and the basic operating point is entirely provided by the thermal power units at the sending end, and the output of the thermal power units at the receiving end does not change. On the other hand, the load increment between the extreme operating point and the basic operating point is entirely caused by the load at the receiving end, while the load at each node at the sending end remains unchanged.

[0128] Step S3: constructing an ATC evaluation model based on the initial operating point set and the limit operating point and on the safety index of the multi-regional power grid operation.

[0129] Step S31: establishing an objective function based on minimizing the overall generation cost of the power system and maximizing the ATC between different regions;

[0130] The objective function under the extreme operating conditions is:

[0131] Where, and are the active power output of the thermal generator at node i under the limit state and its value at the initial operating point; G is the set of power generation nodes to be adjusted at the sending end; c i is the unit power generation cost of unit i, α and β are the weight coefficients of multi-objective optimization.

[0132] Step S32: constructing a power system benchmark operating state model, including grid structure, startup mode, load power, operating parameters of power growth mode and safety constraints.

[0133] Step S33: Continuously increase the exchange power between grid areas by adjusting the power generation and load power until a safety constraint condition is exceeded, obtain the limit operation point of the grid, and construct an ATC evaluation model.

[0134] Step S4: Determine the probability distribution of available transmission capacity of the multi-regional power system based on the ATC evaluation model and the typical daily source-load scenario set.

[0135] According to the typical daily source-load scenario set, the formula for calculating the mean probability distribution of the available transmission capacity of the multi-regional power system is:

[0136] Where ATC s is the available transmission capacity of the power grid in a typical day scenario s, t is the daily time period, T is the set of daily time periods, ATC t,s is the available transmission capacity of the system at time t in a typical day scenario s.

[0137] The generator receives random noise and conditional values ​​of the multi-regional power system as input, while the discriminator receives the wind power output curve or load curve and conditional values ​​as input, and decomposes the wind power output data or load data of each plant into an N×T matrix.

[0138] One-hot encoding is used to convert discrete data in multi-regional power systems into binary vectors, and conditional information is introduced into the training process of the generator and discriminator to ensure that both the generator and the discriminator can obtain conditional information during training to generate and discriminate sample data.

[0139] The noise, conditional values ​​and real data of the multi-regional power system are input into the generator and discriminator respectively. The fake sample data generated by the generator and the real data are input into the discriminator for discrimination. The discriminator outputs the size of the Wasserstein distance as the discrimination result to determine the probability distribution of the available transmission capacity of the multi-regional power system.

[0140] As shown in Figure 3, the present application provides a data-driven scenario controllable generation method based on the Conditional Generative Adversarial Network (CGAN). It aims to set up a two-person zero-sum game under the minimax theorem between the generator neural network and the discriminator neural network. During each training process, the generator continuously updates its weights to generate "false" samples in an attempt to "cheat" the discriminator network, while the discriminator tries to distinguish between real historical samples and generated samples. This training process will continue until the discriminator can no longer determine whether the output of the generator is true. CGAN is based on the generative adversarial network and transmits additional conditional information to the discriminant model and the generative model as part of the input layer.

[0141] Assume that the observed value of renewable energy in time t∈T Used for each power plant. Real historical data distribution express, It is unknown and difficult to model and solve. Assume that a set of known distributions can be obtained. The noise vector input Z (which can be obtained by the joint Gaussian method) is denoted as Convert from The sample Z drawn from the (Historical data) distribution needs to be achieved by training the generator network and the discriminator network at the same time. Let G represent the distribution of θ (G) The parameterized generating function is denoted as G(Z;θ (G) ); Let D represent the (D) The parameterized generating function, which we write as D(x; θ (D) ). Here, θ (G) and θ (D) are the weights of the two neural networks respectively.

[0142] Generator: The generator is trained to take a batch of random variables Z as input and then output the real scene through a series of upsampling operations. Assume that Z is a distributed random variables, then G(Z;θ (G) ) is a new random variable, and we express its distribution as

[0143] Discriminator: The discriminator takes samples from real historical data and performs a series of downsampling operations using another deep neural network. It outputs a continuous value p real , used to measure the input sample and The discriminator can be expressed as: p real =D(x;θ(D) ),

[0144] Where: D(x;θ (D) ) represents the (D) Parameterized generating function; x is taken from historical data The discriminator is continuously trained to distinguish as well as and maximize (real data) and (Generate data) differences.

[0145] After defining the discriminator and generator, we need to formulate loss functions for the generator and discriminator (represented by L G and L D ). Among them, L G The smaller the value of , the more realistic the samples generated by the generator are from the perspective of the discriminator, and L D The smaller the value of L, the stronger the discriminator's ability to distinguish between generated scenes and historical scenes. G and L D for:

[0146] Where: represents the mathematical expectation.

[0147] In order to set up a game between the generator and the discriminator, the minima and maxima of the game are described using the Wasserstein distance. and And make them close to each other, the Wasserstein distance is defined as:

[0148] Where: y is different types of conditions, For random variables expected value, For random variables Expected value, class labels are assigned based on user-defined classification metrics. Class labels are simply representations of sample events reflected by the distribution of generated data. CGAN should be able to learn conditional distributions and generate samples based on any given meaningful conditional metric.

[0149] The specific steps for conditional scenario generation are as follows:

[0150] The generator accepts random noise and a conditional value as input, while the discriminator accepts a wind power output curve or load curve and the conditional value as input. The wind power output or load data for each plant is decomposed into an N × T matrix. For wind power and load, five and three numbers are selected as classification labels, respectively. These discrete classification labels are converted into binary vectors using one-hot encoding. This method effectively incorporates conditional information into the training process of the generator and discriminator, ensuring that both generator and discriminator have access to conditional information during training, enabling better generation and discrimination of sample data.

[0151] Subsequently, these label values ​​are horizontally spliced ​​into the historical data matrix to input the real data and conditional values ​​as the training set into the discriminator for training. Similarly, the Gaussian noise randomly sampled from the normal distribution is set to have the same dimension, namely N×T, and is horizontally spliced ​​with the label values ​​after one-hot encoding, and fed into the generator as input for training. In each training, an appropriate number of batches are selected to contain noise, conditional values, and real data and conditional values, and they are input into the generator and discriminator respectively. The fake sample data generated by the generator is input into the discriminator together with the real data for discrimination. The discriminator outputs the size of the Wasserstein distance as its discrimination result. In order to improve the accuracy of the discriminator network and reduce the number of updates to the network parameters, thereby making the training process more stable, we choose to train the generator once after every four training of the discriminator during the training process. This alternating update method can balance the training of the generator and discriminator and ensure that they influence each other to achieve better training results. Through continuous iterative training, the Wasserstein distance will gradually approach 0. At this time, the generator can more accurately generate wind power output or load scenarios under different conditions.

[0152] This application addresses the multi-dimensional uncertainty of renewable energy output and load demand by using a conditional generative adversarial network method to generate and screen a set of typical daily scenarios, thereby constructing a set of initial operating points for the system. A method for calculating the system's extreme operating points that considers safety indicators for multi-regional power grid operation is proposed, and an ATC evaluation model is constructed. Based on the generated set of scenarios, a probability distribution of available transmission capacity is obtained, providing an important basis for new power system expansion planning and power market trading mechanisms.

[0153] This application proposes a method for evaluating the available transmission capacity of a multi-regional power system based on the multi-dimensional uncertainty of renewable energy output and load demand caused by large-scale renewable energy grid connection; the typical daily source-load scenario set established by the conditional generative adversarial network can cover various possible operating scenarios, truly reflect the operating conditions of the power system, and improve the accuracy of the evaluation of the available transmission capacity of the multi-regional power system; at the same time, the ability to accurately and efficiently calculate the ATC of the renewable energy transmission section online can improve the renewable energy absorption capacity while maintaining the safe and stable operation of the system; the ATC evaluation model uses the minimization of the difference between the cost of the thermal power plant and the ATC of the interconnection line as a multi-objective function, and is carried out under a set of random renewable energy output and load demand scenarios. Under the condition of operational uncertainty, the transmission margin of the connecting line can be fully evaluated, overcoming the problem of being overly conservative in the traditional available transmission capacity evaluation method.

[0154] An embodiment of the present application provides an electronic device, including a processor, a memory, and a program stored in the memory and executable on the processor. When the electronic device is running, the processor executes the program to implement any of the steps of the above-mentioned method for evaluating the available transmission capacity of a multi-regional power system.

[0155] Optionally, the above-mentioned memory and processor can be general-purpose memory and processor, which are not specifically limited here. When the processor runs the computer program stored in the memory, it can execute the above-mentioned method for evaluating the available transmission capacity of a multi-regional power system.

[0156] Corresponding to the method for starting the above-mentioned application, an embodiment of the present application also provides a storage medium on which a program is stored. When the program is run by a processor, the steps of any of the above-mentioned methods for evaluating the available transmission capacity of a multi-regional power system are executed.

[0157] The startup device of the application provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in the embodiment of the present application are the same as those of the aforementioned method embodiment. For the sake of brief description, for any part not mentioned in the device embodiment, reference can be made to the corresponding content in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0158] The embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, read-only memory (Compact Disc Read-Only Memory, CD-ROM), optical storage, etc.) that contain computer-usable program code.

[0159] The above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present application can still be modified or replaced by equivalents. Any modifications or equivalent replacements that do not depart from the spirit and scope of the present application should be included in the scope of protection of the claims of the present application.

Claims

1. A method for assessing available transmission capacity of a multi-regional power system, comprising: Considering the multi-dimensional uncertainty of renewable energy output and load demand, a set of typical daily source-load scenarios is determined based on the conditional generative adversarial network method; Constructing an initial operating point set based on the typical daily source-load scenario set, and determining a limit operating point of the multi-regional power system; According to the initial operating point set and the limit operating point, an available transmission capacity (ATC) evaluation model is constructed based on the security index of multi-regional power grid operation; Determine the probability distribution of available transmission capacity of the multi-regional power system based on the ATC evaluation model and a set of typical daily source-load scenarios.

2. The method according to claim 1, wherein The method of determining a typical daily source-load scenario set based on a conditional generative adversarial network method includes: Based on the conditional generative adversarial network method, a deep neural network model is used to characterize a nonlinear relationship generator and a classification signal discriminator, and conditional information is transmitted as an input layer to the classification signal discriminator and the nonlinear relationship generator. The conditional information includes: historical meteorological data with time attributes, spatial characteristics of the power system, and wind power plant output and load demand characteristics; According to the historical data of new energy output and load demand, a typical daily source-load scenario set S is constructed. 0 : S 0 ={S W ,S D }, S W ={P1 W ,P2 W ,L,P i W }, S D ={P1 D ,P2 D ,L,P i D }, Where: P i W represents the wind farm output at access node i, S W Indicates that P in the system i W The collection of P i D represents the wind farm load at access node i, S D Indicates that P in the system i D W represents the set of wind farm nodes in the system, and D represents the set of load nodes in the system.

3. The method according to claim 1, wherein The constructing an initial operating point set according to the typical daily source-load scenario set and determining the limit operating point of the multi-region power system includes: According to the initial operating point set and based on the safety index of the multi-regional power system operation, the exchange power between the multi-regional power systems is continuously increased by adjusting the power generation and load power, so as to obtain the extreme operating point of the multi-regional power system under extreme operating conditions.

4. The method according to claim 1, wherein The ATC evaluation model is constructed based on the initial operating point set and the limit operating point and the safety index of the multi-regional power grid operation, including: The objective function is established based on minimizing the overall generation cost of the power system and maximizing the ATC between different regions: Where, and are the active power output of the thermal generator at node i under the limit state and its value at the initial operating point; G is the set of power generation nodes to be adjusted at the sending end; c i is the unit power generation cost of unit i, α and β are the weight coefficients of multi-objective optimization; Constructing a power system benchmark operating state model, wherein the power system benchmark operating state model includes operating parameters of the grid structure, startup mode, load power, power growth mode, and safety constraints; By adjusting the power generation and load power, the exchange power between power grid areas is continuously increased until the safety constraint conditions are exceeded. The limit operation point of the power grid is obtained and the ATC evaluation model is constructed.

5. The method according to claim 4, wherein The security constraints include: (1) The active and reactive power balance constraints of nodes under the power system benchmark operation state are: Where, P i,t , Q i,t are the active and reactive injected powers of node i at time t, For the festival The active and reactive outputs of the thermal power generating unit at point i at time t are: is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, λ i is the ratio of reactive load to active load; The linearized power flow constraint under the grid benchmark operation state is: The linear equation can be used to simultaneously solve the amplitude and phase angle of the node voltage and the equivalent admittance of the power flow equation. and As shown below: The capacity constraint of the line under the grid benchmark operation state is: The node voltage constraint under the grid benchmark operation state is: In min ≤V i,t ≤V max , The output constraint of the generator group under the grid benchmark operating state is: The above formula indicates that at time t, the active output of the nth generator must be between its upper and lower limits; The following formulas respectively represent the increase or decrease in the output of the generator set per unit time: Where, δ i,t 、V i,t is the phase angle and amplitude of the voltage at node i at time t; r ij 、x ij is the power of line (i, j) resistance and reactance; is the upper capacity limit of line (i, j); V max 、V min is the upper and lower limits of each PQ node voltage; P i Gmax and P i Gmin Represent the upper and lower limits of power generation of each unit, and r is the upper limit of ramp rate; (2) The active and reactive power balance constraints of the nodes in the power grid under the extreme operation state are: The linearized power flow constraint of the power grid under the extreme operating state is: The capacity constraint of the power line under the extreme operation state is: The node voltage constraint of the power grid under the extreme operation state is: The output constraint of the generator group under the extreme operation state of the power grid is: Where, are the active and reactive injected powers of node i at time t, is the active and reactive output of the thermal power generating unit at node i at time t, is the active power output of the wind farm at node i at time t; is the active and reactive load of node i at time t, For node i The phase angle and amplitude of the voltage at time t; r ij 、x ij are the resistance and reactance of line (i, j); is the upper capacity limit of line (i, j); V max 、V min is the upper and lower limits of each PQ node voltage; P i Gmax and P i Gmin They represent the upper and lower limits of power generation of each unit respectively, r is the upper limit of ramp rate; SR represents the sending end area in the system, and SK represents the receiving end area in the system.

6. The method according to claim 2, wherein: Determining the probability distribution of available transmission capacity of the multi-regional power system based on the ATC evaluation model and the typical daily source-load scenario set includes: The nonlinear relationship generator receives random noise and conditional values ​​of the multi-regional power system as input, and the classification signal discriminator receives wind power output curve or load curve and conditional values ​​as input, and decomposes wind power output data or load data of each plant into a matrix form; The noise, conditional values ​​and real data of the multi-regional power system are respectively input into the classification signal discriminator and the nonlinear relationship generator, and the false sample data generated by the nonlinear relationship generator and the real data are input into the classification signal discriminator for discrimination. The classification signal discriminator outputs the Wasserstein distance as the discrimination result to determine the probability distribution of the available transmission capacity of the multi-regional power system.

7. The method according to any one of claims 1 to 6, wherein: The mean of the probability distribution of the available transmission capacity of the multi-regional power system is: Where ATC s is the daily available transmission capacity of the power system under the typical daily scenario s, t is the daily time period, T is the set of daily time periods, ATC t,s is the available transmission capacity of the system at time t in a typical day scenario s.

8. A device for evaluating available transmission capacity of a multi-regional power system, comprising: The source-load scenario set determination module is set to consider the multi-dimensional uncertainty of renewable energy output and load demand, and determine the typical daily source-load scenario set based on the conditional generative adversarial network method; a limit operating point determination module, configured to construct an initial operating point set based on the typical daily source-load scenario set and determine the limit operating point of the multi-region power system; An evaluation model construction module is configured to construct an available transmission capacity (ATC) evaluation model based on the initial operating point set and the limit operating point and the security index of the multi-regional power grid operation; The available transmission capacity evaluation module is configured to determine the probability distribution of the available transmission capacity of the multi-regional power system based on the ATC evaluation model and a typical daily source-load scenario set.

9. An electronic device comprising a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the electronic device is running, the processor executes the program to implement the method for evaluating the available transmission capacity of a multi-regional power system as described in any one of claims 1 to 7.

10. A storage medium having a program stored thereon, wherein when a processor executes the program, the method for evaluating available transmission capacity for a multi-regional power system according to any one of claims 1 to 7 is executed.

Images