Power distribution network node optical charge typical scene extraction method and device

Abstract

The invention relates to the technical field of power distribution network operation control, and particularly provides a power distribution network node optical charge typical scene extraction method and device, and the method comprises the steps: carrying out the dimension reduction of a power distribution network node optical charge scene description matrix, and obtaining a dimension reduction matrix; substituting the dimension reduction matrix corresponding to each time period into a pre-fitted probability distribution function, and selecting a power distribution network node optical load scene description matrix corresponding to the dimension reduction matrix of which the probability is greater than a threshold value as a candidate power distribution network node optical load typical scene; and obtaining similarity indexes among the candidate power distribution network node optical charge typical scenes, and screening the power distribution network node optical charge typical scenes based on the similarity indexes among the candidate power distribution network node optical charge typical scenes. According to the technical scheme provided by the invention, the optical charge space-time distribution characteristics of multiple nodes in the power distribution network are comprehensively captured, and detailed data support is provided for analysis and optimization of a power system.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network operation control, and in particular to a method and device for extracting typical light-load scenarios of distribution network nodes. Background Art

[0002] In recent years, with the large-scale access of renewable energy and the development of smart grid technology, the operating characteristics of distribution networks have become more complex and changeable. The intermittent and uncertain nature of photovoltaic power generation and the dynamic changes in load demand have posed new challenges to the planning, operation and management of distribution networks. Therefore, how to accurately extract and analyze the typical output scenarios of multi-node systems with regard to spatiotemporal distribution characteristics has become a key issue that needs to be urgently addressed in the field of distribution networks.

[0003] Scholars have conducted a lot of research on typical output scenarios. Traditional methods for analyzing photovoltaic output scenarios in distribution networks mainly focus on single nodes or independent systems. Typical daily output scenarios of different nodes may not occur at the same time. These methods often ignore the mutual influence and spatiotemporal distribution characteristics between nodes in the distribution network. In addition, existing technologies mostly use static or simplified models to predict and analyze photovoltaic output. These models cannot accurately capture the volatility of photovoltaic power generation and the variability of load, resulting in limited accuracy and practicality of scenario analysis. Although existing studies have attempted to improve the effect of scenario extraction by introducing probability models, cluster analysis and other methods, these methods still face problems of low efficiency and insufficient accuracy when processing large-scale, high-dimensional photovoltaic data. In addition, existing technologies have not yet formed a mature and systematic methodology in comprehensively considering the spatiotemporal correlation and dynamic characteristics between multiple nodes in the power grid. Summary of the invention

[0004] In order to overcome the above defects, the present invention proposes a method and device for extracting typical scenarios of light loads at distribution network nodes.

[0005] In a first aspect, a method for extracting typical scenarios of light loads at distribution network nodes is provided, the method comprising:

[0006] The distribution network node light load scenario description matrix is ​​reduced in dimension to obtain a reduced-dimension matrix;

[0007] Substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with probability greater than the threshold as the candidate distribution network node optical load typical scenario;

[0008] The similarity indexes between the typical light-load scenarios of the candidate distribution network nodes are obtained, and the typical light-load scenarios of the distribution network nodes are screened based on the similarity indexes between the typical light-load scenarios of the candidate distribution network nodes.

[0009] Preferably, the distribution network node light load scenario description matrix is ​​as follows:

[0010]

[0011] In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

[0012] Preferably, the dimension reduction matrix is ​​as follows:

[0013] Y′=YV

[0014] In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

[0015] Furthermore, the initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows:

[0016] Y=XU

[0017] The covariance matrix of Y is as follows:

[0018]

[0019] In the above formula, X is the distribution network node optical load scenario description matrix, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the distribution network node optical load scenario description matrix, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

[0020] Preferably, the pre-fitted probability distribution function is as follows:

[0021]

[0022] In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π k is the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

[0023] Furthermore, the process of obtaining the pre-fitted probability distribution function includes:

[0024] The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

[0025] Preferably, the similarity index θ between the typical light load scenarios of the candidate distribution network nodes is as follows:

[0026] θ=p1d E,pv (S a , S b )+p2d E,load (S a , S b )+p3d C (S a , S b )

[0027] In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

[0028] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows:

[0029]

[0030] In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0031] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows:

[0032]

[0033] In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix of the matrix composed of odd columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0034] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows:

[0035]

[0036] In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

[0037] Preferably, the screening of typical light load scenarios of distribution network nodes based on similarity indicators between typical light load scenarios of candidate distribution network nodes includes:

[0038] When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted;

[0039] The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

[0040] In a second aspect, a device for extracting typical light load scenarios of distribution network nodes is provided, and the device for extracting typical light load scenarios of distribution network nodes comprises:

[0041] A dimension reduction module is used to reduce the dimension of the light load scenario description matrix of the distribution network node to obtain a reduced dimension matrix;

[0042] An analysis module is used to substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with a probability greater than a threshold as the candidate distribution network node optical load typical scenario;

[0043] The screening module is used to obtain similarity indicators between typical light-load scenarios of candidate distribution network nodes, and screen typical light-load scenarios of distribution network nodes based on the similarity indicators between typical light-load scenarios of candidate distribution network nodes.

[0044] Preferably, the distribution network node light load scenario description matrix is ​​as follows:

[0045]

[0046] In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

[0047] Preferably, the dimension reduction matrix is ​​as follows:

[0048] Y′=YV

[0049] In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

[0050] Furthermore, the initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows:

[0051] Y=XU

[0052] The covariance matrix of Y is as follows:

[0053]

[0054] In the above formula, X is the distribution network node optical load scenario description matrix, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the distribution network node optical load scenario description matrix, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

[0055] Preferably, the pre-fitted probability distribution function is as follows:

[0056]

[0057] In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π kis the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

[0058] Furthermore, the process of obtaining the pre-fitted probability distribution function includes:

[0059] The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

[0060] Preferably, the similarity index θ between the typical light load scenarios of the candidate distribution network nodes is as follows:

[0061] θ=p1d E,pv (S a , S b )+p2d E,load (S a , S b )+p3d C (S a , S b )

[0062] In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

[0063] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows:

[0064]

[0065] In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0066] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows:

[0067]

[0068] In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix of the matrix composed of odd columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0069] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows:

[0070]

[0071] In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

[0072] Preferably, the screening of typical light load scenarios of distribution network nodes based on similarity indicators between typical light load scenarios of candidate distribution network nodes includes:

[0073] When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted;

[0074] The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

[0075] In a third aspect, a computer device is provided, comprising: one or more processors;

[0076] The processor is configured to execute one or more programs;

[0077] When the one or more programs are executed by the one or more processors, the method for extracting typical scenarios of light loads at distribution network nodes is implemented.

[0078] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed, the method for extracting typical scenarios of light loads at distribution network nodes is implemented.

[0079] The above one or more technical solutions of the present invention have at least one or more of the following beneficial effects:

[0080] The present invention provides a method and device for extracting typical light-load scenarios of distribution network nodes, including: reducing the dimension of the light-load scenario description matrix of distribution network nodes to obtain a reduced-dimensional matrix; substituting the reduced-dimensional matrix corresponding to each time period into a pre-fitted probability distribution function, and selecting the light-load scenario description matrix of distribution network nodes corresponding to the reduced-dimensional matrix with a probability greater than a threshold as a candidate light-load scenario of distribution network nodes; obtaining similarity indicators between each candidate light-load typical scenario of distribution network nodes, and screening the light-load typical scenario of distribution network nodes based on the similarity indicators between each candidate light-load typical scenario of distribution network nodes. The technical solution provided by the present invention comprehensively captures the spatiotemporal distribution characteristics of light loads of multiple nodes in the distribution network, and provides detailed data support for the analysis and optimization of the power system, specifically:

[0081] The dimensionality reduction matrix of the present invention not only retains the core features of the data, but also simplifies the complexity of subsequent processing. It also effectively identifies and extracts typical scenarios representing different operating states through a Gaussian mixture model, providing an accurate reference for the formulation of power grid operation strategies. The constructed comprehensive similarity index between scenarios comprehensively considers multiple factors such as the Euclidean distance between photovoltaic loads and the correlation between light and load, thereby improving the effect and efficiency of scene clustering. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a flow chart of the main steps of the method for extracting typical light load scenarios of distribution network nodes according to an embodiment of the present invention. DETAILED DESCRIPTION

[0083] The specific implementation modes of the present invention will be further described in detail below in conjunction with the accompanying drawings.

[0084] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0085] As disclosed in the background technology, in recent years, with the large-scale access of renewable energy and the development of smart grid technology, the operating characteristics of the distribution network have become more complex and changeable. The intermittent and uncertain nature of photovoltaic power generation and the dynamic changes in load demand have posed new challenges to the planning, operation and management of the distribution network. Therefore, how to accurately extract and analyze the typical output scenarios of multi-node systems with light loads that take into account the temporal and spatial distribution characteristics has become a key issue that needs to be urgently addressed in the field of distribution networks.

[0086] Scholars have conducted a lot of research on typical output scenarios. Traditional methods for analyzing photovoltaic output scenarios in distribution networks mainly focus on single nodes or independent systems. Typical daily output scenarios of different nodes may not occur at the same time. These methods often ignore the mutual influence and spatiotemporal distribution characteristics between nodes in the distribution network. In addition, existing technologies mostly use static or simplified models to predict and analyze photovoltaic output. These models cannot accurately capture the volatility of photovoltaic power generation and the variability of load, resulting in limited accuracy and practicality of scenario analysis. Although existing studies have attempted to improve the effect of scenario extraction by introducing probability models, cluster analysis and other methods, these methods still face problems of low efficiency and insufficient accuracy when processing large-scale, high-dimensional photovoltaic data. In addition, existing technologies have not yet formed a mature and systematic methodology in comprehensively considering the spatiotemporal correlation and dynamic characteristics between multiple nodes in the power grid.

[0087] In order to improve the above-mentioned problems, the present invention provides a method and device for extracting typical light load scenarios of distribution network nodes, including: reducing the dimension of the light load scenario description matrix of the distribution network nodes to obtain a reduced dimension matrix; substituting the reduced dimension matrix corresponding to each time period into a pre-fitted probability distribution function, and selecting the light load scenario description matrix of the distribution network nodes corresponding to the reduced dimension matrix with a probability greater than a threshold as the candidate light load typical scenario of the distribution network nodes; obtaining the similarity index between the candidate light load typical scenarios of the distribution network nodes, and screening the light load typical scenarios of the distribution network nodes based on the similarity index between the candidate light load typical scenarios of the distribution network nodes. The technical solution provided by the present invention comprehensively captures the spatiotemporal distribution characteristics of light loads of multiple nodes in the distribution network, and provides detailed data support for the analysis and optimization of the power system, specifically:

[0088] The dimensionality reduction matrix of the present invention not only retains the core features of the data, but also simplifies the complexity of subsequent processing. It also effectively identifies and extracts typical scenarios representing different operating states through a Gaussian mixture model, providing an accurate reference for the formulation of power grid operation strategies. The constructed comprehensive similarity index between scenarios comprehensively considers multiple factors such as the Euclidean distance between photovoltaic loads and the correlation between light and load, thereby improving the effect and efficiency of scene clustering.

[0089] The above scheme is described in detail below.

[0090] Example 1

[0091] See attached Figure 1 , Figure 1 FIG. 1 is a flow chart showing the main steps of a method for extracting typical light load scenarios of distribution network nodes according to an embodiment of the present invention. Figure 1 As shown, the method for extracting typical light load scenarios of distribution network nodes in the embodiment of the present invention mainly includes the following steps:

[0092] Step S101: reducing the dimension of the light load scenario description matrix of the distribution network node to obtain a reduced dimension matrix;

[0093] Step S102: Substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with a probability greater than a threshold as the candidate distribution network node optical load typical scenario;

[0094] Step S103: obtaining similarity indices between typical light load scenarios of candidate distribution network nodes, and screening typical light load scenarios of distribution network nodes based on the similarity indices between typical light load scenarios of candidate distribution network nodes.

[0095] In this embodiment, the distribution network node light load scenario description matrix is ​​as follows:

[0096]

[0097] In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

[0098] In this embodiment, the dimension reduction matrix is ​​as follows:

[0099] Y′=YV

[0100] In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

[0101] In one embodiment, the initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows:

[0102] Y=XU

[0103] The covariance matrix of Y is as follows:

[0104]

[0105] In the above formula, X is the distribution network node optical load scenario description matrix, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the distribution network node optical load scenario description matrix, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

[0106] In this embodiment, the pre-fitted probability distribution function is as follows:

[0107]

[0108] In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π k is the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

[0109] In one embodiment, the process of obtaining the pre-fitted probability distribution function includes:

[0110] The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

[0111] In this embodiment, in order to effectively organize and summarize the extracted typical scenarios, the present invention constructs a similarity index between scenarios to screen and optimize the candidate typical output scenarios. This step ensures that the typical scenarios finally selected are not only economical and efficient in quantity, but also can accurately reflect the operating characteristics of the system in terms of quality. The similarity index θ between the typical light load scenarios of the candidate distribution network nodes is as follows:

[0112] θ=p1d E,pv (S a , S b)+p2d E,load (S a , S b )+p3d C (S a , S b )

[0113] In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

[0114] In one embodiment, the typical light load scenario S of the candidate distribution network node a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows:

[0115]

[0116] In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0117] In one embodiment, the typical light load scenario S of the candidate distribution network node a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows:

[0118]

[0119] In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node Sa The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix of the matrix composed of odd columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0120] In one embodiment, the typical light load scenario S of the candidate distribution network node a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows:

[0121]

[0122] In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

[0123] In this embodiment, the screening of typical light load scenarios of distribution network nodes based on similarity indicators between typical light load scenarios of candidate distribution network nodes includes:

[0124] When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted;

[0125] The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

[0126] Example 2

[0127] Based on the same inventive concept, the present invention also provides a typical scene extraction device for light load at a distribution network node, the typical scene extraction device for light load at a distribution network node comprising:

[0128] A dimension reduction module is used to reduce the dimension of the light load scenario description matrix of the distribution network node to obtain a reduced dimension matrix;

[0129] An analysis module is used to substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with a probability greater than a threshold as the candidate distribution network node optical load typical scenario;

[0130] The screening module is used to obtain similarity indicators between typical light-load scenarios of candidate distribution network nodes, and screen typical light-load scenarios of distribution network nodes based on the similarity indicators between typical light-load scenarios of candidate distribution network nodes.

[0131] Preferably, the distribution network node light load scenario description matrix is ​​as follows:

[0132]

[0133] In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

[0134] Preferably, the dimension reduction matrix is ​​as follows:

[0135] Y′=YV

[0136] In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

[0137] Furthermore, the initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows:

[0138] Y=XU

[0139] The covariance matrix of Y is as follows:

[0140]

[0141] In the above formula, X is the distribution network node optical load scenario description matrix, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the distribution network node optical load scenario description matrix, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

[0142] Preferably, the pre-fitted probability distribution function is as follows:

[0143]

[0144] In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π k is the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

[0145] Furthermore, the process of obtaining the pre-fitted probability distribution function includes:

[0146] The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

[0147] Preferably, the similarity index θ between the typical light load scenarios of the candidate distribution network nodes is as follows:

[0148] θ=p1d E,pv (S a , S b )+p2d E,load (S a , S b )+p3d C (S a , S b )

[0149] In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

[0150] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows:

[0151]

[0152] In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0153] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows:

[0154]

[0155] In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix of the matrix composed of odd columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

[0156] Furthermore, the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows:

[0157]

[0158] In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

[0159] Preferably, the screening of typical light load scenarios of distribution network nodes based on similarity indicators between typical light load scenarios of candidate distribution network nodes includes:

[0160] When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted;

[0161] The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

[0162] Example 3

[0163] Based on the same inventive concept, the present invention also provides a computer device, which includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of a typical scene extraction method of light load of a distribution network node in the above embodiment.

[0164] Example 4

[0165] Based on the same inventive concept, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It can be understood that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of a typical scene extraction method for light loads of distribution network nodes in the above embodiment.

[0166] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0167] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0168] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0169] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention 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 invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for extracting typical scenarios of light loads at distribution network nodes, characterized in that: The method comprises: The distribution network node light load scenario description matrix is ​​reduced in dimension to obtain a reduced-dimension matrix; Substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with probability greater than the threshold as the candidate distribution network node optical load typical scenario; The similarity indexes between the typical light-load scenarios of the candidate distribution network nodes are obtained, and the typical light-load scenarios of the distribution network nodes are screened based on the similarity indexes between the typical light-load scenarios of the candidate distribution network nodes.

2. The method according to claim 1, characterized in that The distribution network node light load scenario description matrix is ​​as follows: In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

3. The method according to claim 1, characterized in that The dimension reduction matrix is ​​as follows: Y′=YV In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

4. The method according to claim 3, characterized in that The initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows: Y=XU The covariance matrix of Y is as follows: In the above formula, X is the distribution network node optical load scenario description matrix, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the distribution network node optical load scenario description matrix, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

5. The method according to claim 1, characterized in that The pre-fitted probability distribution function is as follows: In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π k is the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

6. The method according to claim 5, characterized in that The process of obtaining the pre-fitted probability distribution function includes: The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

7. The method according to claim 1, characterized in that The similarity index θ between the typical scenarios of light loads of the candidate distribution network nodes is as follows: θ=p1d E,pv (S a ,S b )+p2d E,load (S a ,S b )+p3d C (S a ,S b ) In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

8. The method according to claim 7, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows: In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

9. The method according to claim 8, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows: In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix composed of the odd number of columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

10. The method according to claim 8, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows: In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

11. The method according to claim 1, characterized in that The method of screening the typical light load scenarios of distribution network nodes based on the similarity index between the typical light load scenarios of candidate distribution network nodes includes: When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted; The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

12. A typical scene extraction device for light load of distribution network nodes, characterized in that: The device comprises: A dimension reduction module is used to reduce the dimension of the light load scenario description matrix of the distribution network node to obtain a reduced dimension matrix; An analysis module is used to substitute the dimension reduction matrix corresponding to each time period into the pre-fitted probability distribution function, and select the distribution network node optical load scenario description matrix corresponding to the dimension reduction matrix with a probability greater than a threshold as the candidate distribution network node optical load typical scenario; The screening module is used to obtain similarity indicators between typical light-load scenarios of candidate distribution network nodes, and screen typical light-load scenarios of distribution network nodes based on the similarity indicators between typical light-load scenarios of candidate distribution network nodes.

13. The device according to claim 12, characterized in that The distribution network node light load scenario description matrix is ​​as follows: In the above formula, P t,i is the load power of node i at time t, p t,i is the photovoltaic power of node i at time t, P t,n is the load power of node n at time t, p t,n is the photovoltaic power of node n at time t, t is the total number of time points, and n is the total number of nodes.

14. The device according to claim 12, characterized in that The dimension reduction matrix is ​​as follows: Y′=YV In the above formula, Y′ is the dimension reduction matrix, Y is the initial dimension reduction matrix corresponding to the light load scenario description matrix of the distribution network node, V is the eigenvector corresponding to the first x largest eigenvalues ​​of the covariance matrix of Y, and x is the preset dimension.

15. The device according to claim 14, characterized in that The initial dimension reduction matrix corresponding to the distribution network node light load scenario description matrix is ​​as follows: Y=XU The covariance matrix of Y is as follows: In the above formula, X is the optical load scenario description matrix of the distribution network node, U is the first x smallest non-trivial eigenvectors of the Laplace matrix corresponding to the optical load scenario description matrix of the distribution network node, C is the covariance matrix of Y, T is the transpose sign, and n is the total number of nodes.

16. The device according to claim 12, characterized in that The pre-fitted probability distribution function is as follows: In the above formula, p is the probability of the reduced dimension matrix, K is the number of Gaussian components, and π k is the weight of the kth Gaussian component, μ k is the sample mean, ∑k is the sample covariance matrix, Ν(Y′|μ k ,∑k) is a multivariate Gaussian distribution with sample mean and sample covariance matrix as parameters, and Y′ is a dimensionality reduction matrix.

17. The device according to claim 16, characterized in that The process of obtaining the pre-fitted probability distribution function includes: The dimension reduction matrix of the historical period is used as a sample, and the expectation maximization algorithm is used to iteratively learn the parameters of the probability distribution function to obtain the pre-fitted probability distribution function, wherein the parameters include: the weight of the Gaussian component, the sample mean and the sample covariance matrix.

18. The device according to claim 12, characterized in that The similarity index θ between the typical scenarios of light loads of the candidate distribution network nodes is as follows: θ=p1d E,pv (S a ,S b )+p2d E,load (S a ,S b )+p3d C (S a ,S b ) In the above formula, p1, p2, and p3 are the first, second, and third weights respectively, and d E,pv (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between the photovoltaic power, d E,load (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance between load powers, d C (S a , S b ) is the typical scenario of the light load of the candidate distribution network node S a Typical scenario of light load of candidate distribution network nodes S b The correlation distance between them.

19. The device according to claim 18, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The Euclidean distance of photovoltaic power between them is as follows: In the above formula, D pv (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of even columns and the typical scenario of the optical load of the candidate distribution network node S b The element value of the cth row and mth column of the difference matrix of the matrix composed of even columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

20. The device according to claim 19, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The load power Euclidean distance between them is as follows: In the above formula, D E,load (c,m) is the typical scenario of the light load of the candidate distribution network node S a The matrix composed of odd columns and typical scenarios of optical loads of candidate distribution network nodes S b The element value of the cth row and mth column of the difference matrix composed of the odd number of columns in , where C is the number of rows of the difference matrix and M is the number of columns of the difference matrix.

21. The device according to claim 19, characterized in that The typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The correlation distances between them are as follows: In the above formula, D c (i, j) is the typical scenario S of the candidate distribution network node light load a Typical scenario of light load of candidate distribution network nodes S b The element value of the i-th row and j-th column of the difference matrix, t is the total number of time points, and n is the total number of nodes.

22. The device according to claim 12, characterized in that The screening of typical light load scenarios of distribution network nodes based on similarity indicators between typical light load scenarios of candidate distribution network nodes includes: When the typical scenario of the candidate distribution network node light load S a Typical scenario of light load of candidate distribution network nodes S b When the similarity index between them is less than the preset value, the candidate distribution network node light load typical scenario with small probability is deleted; The remaining candidate distribution network node typical light load scenarios are taken as the distribution network node typical light load scenarios.

23. A computer device, characterized in that: include: one or more processors; The processor is configured to execute one or more programs; When the one or more programs are executed by the one or more processors, the method for extracting typical scenarios of light loads at distribution network nodes as described in any one of claims 1 to 11 is implemented.

24. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the method for extracting typical scenarios of light loads at distribution network nodes as described in any one of claims 1 to 11 is implemented.

Images