Power system high-dimensional uncertainty aggregation modeling method based on Gaussian mixture model

Abstract

The invention discloses an electric power system high-dimensional uncertainty aggregation modeling method based on a Gaussian mixture model, which comprises two stages of original uncertainty distribution representation and GMM aggregation modeling, and the GMM aggregation modeling specifically comprises three links of aggregation variable dimension determination, aggregation parameter calculation and aggregation result integration. Therefore, probability security aggregation of the high-dimensional uncertainty variables of the power system is realized under the condition of greatly reducing the demand quantity of samples, and the problem that the quantity of required samples is large when high-dimensional data is processed by a Monte Carlo sampling method is effectively solved. The method can be applied to uncertainty power system high-dimensional variable dimension reduction aggregation under high-proportion new energy access, sample space reduction is achieved under the condition that probability features are not changed, and the method is further applied to power system risk assessment, reliability analysis and operation planning.

Description

Technical Field

[0001] The present invention belongs to the field of power system new energy power data aggregation, and specifically relates to a high-dimensional uncertainty aggregation modeling method for power systems based on a Gaussian mixture model. Background Art

[0002] With the rapid development of new energy technologies, the penetration rate of new energy such as wind power and photovoltaics has continued to increase, and the randomness and volatility of their output have had a great impact on the safe and stable operation of the power grid. At the same time, due to the continuous increase in the number of new energy power plants and units, the dimension of their output data has been greatly increased, and the resulting "high-dimensional uncertainty" problem has brought certain challenges to power system analysis.

[0003] The increase in the dimension of the uncertain output and load data of new energy sources will lead to problems such as "dimensionality curse", which greatly increases the difficulty of data processing; and high-dimensional data has poor readability, which is not conducive to analysis and identification, so it is usually necessary to reduce the dimension of the data. The traditional Monte Carlo sampling method can reduce the dimension of variables, but the Monte Carlo method is prone to uneven sampling in high-dimensional space, and usually requires a large number of sampling simulations to ensure the accuracy of the results, and the computational efficiency is relatively low. Summary of the invention

[0004] Purpose of the invention: The purpose of the present invention is to provide a high-dimensional uncertainty aggregation modeling method for power systems based on a Gaussian mixture model, which is used to perform dimensionality reduction and aggregation processing on high-dimensional uncertainty variables.

[0005] Technical solution: The Gaussian mixture model-based high-dimensional uncertainty aggregation modeling method for power systems described in the present invention is implemented in two stages: original uncertainty distribution characterization and GMM-based aggregation. The original uncertainty distribution characterization uses power system data as original high-dimensional random variables, and GMM is used to fit the original high-dimensional random variables of the power system. The original high-dimensional random variables of the power system are represented in a specific analytical form to obtain a fitted GMM model; the GMM-based aggregation uses the fitted GMM model parameters to calculate the aggregation model parameters based on the required aggregation variable dimension, and integrates them into the aggregation model result.

[0006] Furthermore, GMM is used to fit the original high-dimensional random variables of the power system, and the original high-dimensional random variables of the power system are represented by specific analytical forms, including:

[0007] Power system data includes renewable energy output or load;

[0008] The original high-dimensional random variable data set X of new energy output or load is X=(x1, x2, ..., x d ) TGMM is used for fitting, and the fitting is a GMM model; x i is the original high-dimensional random variable of the power system of the i-th dimension, i = 1, 2, ..., d, d is the total dimension of the original high-dimensional random variable of the power system, T is the transpose of the matrix; the GMM model parameters obtained by fitting: the number of Gaussian components k, the weight coefficient of each component ω k , mean matrix μ k and the covariance matrix Σ k , the uncertainty distribution of the original high-dimensional random variable data set of renewable energy output or load is characterized by a Gaussian mixture model:

[0009]

[0010] Among them, P(X) is the GMM probability density function of the original high-dimensional random variable data set of new energy output or load, K is the total number of Gaussian components of the GMM model, N k is the probability density of the kth Gaussian component in the GMM model.

[0011] Furthermore, parameter estimation of the GMM model of the original high-dimensional random variable of the renewable energy output or load is achieved through the expectation-maximization algorithm, and the optimal number of Gaussian components of the GMM model is determined by the elbow method, the silhouette coefficient method or the Bayesian information criterion.

[0012] Furthermore, GMM-based variable aggregation includes:

[0013] (1) Determine the required dimension n of the power system aggregate variables and the original high-dimensional random variables contained in each aggregate variable;

[0014] (2) According to whether the aggregated variable of renewable energy output or load in actual power system analysis needs to obtain a one-dimensional or n-dimensional result, the mean matrix μ of each Gaussian component parameter in the GMM model fitted with the original high-dimensional random variable of renewable energy output or load is calculated. k and the covariance matrix Σ k Transformed into the aggregated result μ by analytical calculation k,Y and σ k,Y Or μ k,Y and Σ k,Y ;μ k,Y is the one-dimensional mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, is the variance of the kth Gaussian component of the original high-dimensional random variable data set of renewable energy output or load after aggregation; μ k,Y Represents the mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, Σ k,Y is the covariance matrix of the kth Gaussian component after aggregation;

[0015] (3) The mean and covariance results μ after aggregation are obtained k,Y and σ k,Y Or μ k,Y and Σ k,Y , and the joint probability distribution model of the aggregated variables of the power system after dimensionality reduction and aggregation is obtained.

[0016] Furthermore, in step (2), if the dimension of the required power system aggregate variable is one-dimensional, for the kth Gaussian component of the aggregate variable, its corresponding mean and variance are expressed as:

[0017]

[0018] Among them, μ k,Y is the one-dimensional mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μ k,i is the kth Gaussian component x i The mean value of x i is the original high-dimensional random variable of the power system of the i-th dimension, d is the total dimension of the original high-dimensional random variables of the power system; Σ k,ij is the kth Gaussian component x i With x j The covariance of x j is the j-th dimension of the original high-dimensional random variable of the power system. If i=j, then x i Its own variance, that is, ∑ k,ii is the kth Gaussian component x i Its own variance, i and j are the indices of the dimensions of the original high-dimensional random variables of the power system; It is the variance of the kth Gaussian component after the original high-dimensional random variable data set of renewable energy output or load is aggregated.

[0019] Furthermore, in step (2), if the dimension of the required power system aggregate variable is n, for the kth Gaussian component of the aggregate variable, each element in its mean column vector is the sum of the means of several random variables corresponding to the aggregate variable; the calculation of the covariance matrix includes the following steps:

[0020] (21) Solving the mean matrix: Each element in the mean matrix is ​​the sum of the means of several original high-dimensional random variables corresponding to the aggregate variable;

[0021] If the m-dimensional power system data is evenly distributed into n-dimensional aggregate variables, its mean matrix is ​​expressed as:

[0022]

[0023] Among them, μ k,YRepresents the mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μ k,i is the original variable x in the kth Gaussian component i The mean of

[0024] (22) Solution for the variance of the aggregate variable y: For the kth Gaussian component, the i-th aggregate variable y i The variance of is expressed as:

[0025]

[0026] in, is the i-th aggregate variable y i The variance of the kth Gaussian component, c ij is a 0-1 variable, which means: if the aggregate variable y i contains random variables x j , then the y i The following x j The corresponding coefficient c ij The value is 1, otherwise it is 0. il Similarly; Cov(x k,j ,x k,l ) represents the original high-dimensional random variable x in the kth Gaussian component j With x l The covariance between them, j and l are the indices of the dimensions of the original high-dimensional random variables of the power system;

[0027] (23) The correlation coefficient ρ between each two aggregated variables in the kth Gaussian component k,ab for:

[0028]

[0029] Among them, ρ k,ab represents the aggregate variable y in the kth Gaussian component a With y b The correlation coefficient between k,a ,y k,b ) is the aggregate variable y in the kth Gaussian component a With y b The covariance between is the ath aggregate variable y a The variance of the kth Gaussian component in , is the bth aggregate variable y b The variance of the kth Gaussian component in aj With c bl Both are 0-1 variables, with the same meaning as c ij Similarly;

[0030] (24) Solving the covariance matrix: Combine the variance and correlation coefficient obtained in the above steps to obtain the covariance matrix corresponding to the kth Gaussian component:

[0031]

[0032] Among them, Σ k,Y is the covariance matrix of the kth Gaussian component after aggregation, is the variance of the kth Gaussian element in the first aggregate variable y1, is the variance of the kth Gaussian element in the second aggregate variable y2, is the nth aggregate variable y n The variance of the kth Gaussian element in k,1 is the standard deviation of the kth Gaussian element in the first aggregate variable y1, σ k,2 is the standard deviation of the kth Gaussian element in the second aggregate variable y2, σ k,n is the nth aggregate variable y n The standard deviation of the kth Gaussian element in k,12 is the correlation coefficient between the aggregate variables y1 and y2 in the kth Gaussian component, ρ k,1n is the aggregate variable y1 and y in the kth Gaussian component n The correlation coefficient between .

[0033] Furthermore, the joint probability distribution model of the aggregated variables of the power system after dimensionality reduction and aggregation is:

[0034]

[0035] Among them, P(Y) is the GMM probability density function of the power system data aggregation variable after dimensionality reduction aggregation.

[0036] The system corresponding to the above method includes two stages: original uncertainty distribution representation and GMM-based aggregation, where:

[0037] The original uncertainty distribution characterization unit is used to fit the original high-dimensional random variables of the power system using GMM, and to characterize the original high-dimensional random variables of the power system in a specific analytical form to obtain a fitted GMM model;

[0038] The GMM-based aggregation unit is used to calculate the aggregation model parameters based on the required aggregation variable dimensions using the fitted GMM model parameters, and integrate them into the aggregation model results.

[0039] An electronic device for storing and executing the method, the device comprising:

[0040] A memory storing executable program code;

[0041] a processor coupled to the memory;

[0042] The processor calls the executable program code stored in the memory to execute the steps of the high-dimensional uncertainty aggregation modeling method for the power system based on the Gaussian mixture model.

[0043] A computer-readable storage medium for storing and executing the method, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the steps of high-dimensional uncertainty aggregation modeling of power systems based on Gaussian mixture models.

[0044] Beneficial effects: Compared with the prior art, the significant technical effect of the present invention is that the dimensionality reduction aggregation of the original high-dimensional random variables is realized through the two stages of original uncertainty distribution characterization and GMM-based aggregation, which has higher accuracy and efficiency than the traditional Monte Carlo simulation method, thereby achieving the purpose of saving the number of samples and computing resources in the dimensionality reduction of high-dimensional uncertainty probability preservation. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 Two marginal distribution comparison diagrams of the aggregation method proposed in the embodiment of the present invention and the results obtained by the Monte Carlo method with different sampling times; wherein (a) and (b) are respectively the comparison of the marginal distribution y1 of the aggregation results obtained by the GMM aggregation method and the Monte Carlo method with sampling 100 and 1000 times; (c) and (d) are respectively the comparison of the marginal distribution y2 of the aggregation results obtained by the GMM aggregation method and the Monte Carlo method with sampling 100 and 1000 times;

[0047] Figure 3 The diagram is a comparison diagram of the voltage amplitude and phase angle of a node obtained by substituting the aggregation result obtained by the aggregation method proposed in the embodiment of the present invention and the original data into the probability flow calculation; wherein, (a) is the voltage amplitude comparison diagram of the node, and (b) is the voltage phase angle comparison diagram of the node. DETAILED DESCRIPTION

[0048] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.

[0049] In order to solve the problem of dimensionality reduction and aggregation of high-dimensional data in uncertain power systems, the present invention proposes a high-dimensional uncertainty aggregation modeling method for power systems based on Gaussian mixture models, which realizes the probabilistic dimensionality preservation of high-dimensional variables while ensuring a certain degree of accuracy and efficiency. It is implemented in two stages: original uncertainty distribution characterization and GMM-based aggregation. The original uncertainty distribution characterization uses GMM to fit the original high-dimensional random variables of the power system such as new energy output and load, and characterizes the random variables in the form of specific analytical expressions; the GMM-based aggregation uses the fitted GMM model parameters to calculate the aggregation model parameters based on the required aggregation variable dimension, and integrates them into the aggregation model results. Figure 1 As shown, it mainly includes the following steps:

[0050] S1. First, characterize the uncertainty distribution of the original data of the power system. The power system data is the original high-dimensional random variable, including: new energy output and load, etc. The original high-dimensional random variable data set of new energy output or load is X=(x1,x2,...,x d ) T GMM is used for fitting, and the fitting is a GMM model; x i is the original high-dimensional random variable of the power system of the i-th dimension, that is, the original high-dimensional random variable of the new energy output or load of the i-th dimension, i = 1, 2, ..., d, d is the total dimension of the original high-dimensional random variable of the power system, T is the transpose of the matrix; the GMM model parameters obtained by fitting: the number of Gaussian components k, the weight coefficient ω of each Gaussian component k , mean matrix μ k and the covariance matrix Σ k , the uncertainty distribution of the original high-dimensional random variable data set of renewable energy output or load is characterized by a Gaussian mixture model:

[0051]

[0052] Among them, P(X) is the GMM probability density function of the original high-dimensional random variable data set of new energy output or load, K is the total number of Gaussian components of the GMM model, N k is the probability density of the kth Gaussian component in the GMM model.

[0053] The GMM model parameter estimation of the original high-dimensional random variable data set of the power system such as renewable energy output and load can be achieved through the expectation maximization algorithm, while the optimal number of Gaussian components can be determined by the elbow method, silhouette coefficient method or Bayesian information criterion.

[0054] S2. According to the network architecture, unit location, variable correlation and other factors of the actual power system, combined with the actual power system analysis needs, determine the required aggregate variable dimension n and the original high-dimensional random variables contained in each aggregate variable.

[0055] S3. Depending on whether the power system data aggregation variables need to obtain one-dimensional or n-dimensional results, for each Gaussian component, the original GMM model parameter μ k and Σ k Transformed into the aggregated result μ by analytical calculation k,Y and σ k,Y Or μ k,Y and Σ k,Y The specific steps include:

[0056] S31. If the dimension of the required aggregate variable is one-dimensional, for the kth Gaussian component, its corresponding mean and variance are expressed as:

[0057]

[0058] Among them, μ k,Y is the one-dimensional mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μ k,i is the kth Gaussian component x i The mean of k,ij is the kth Gaussian component x i With x j The covariance of x j is the j-th dimension of the original high-dimensional random variable of the power system. If i=j, then x i Its own variance, that is, ∑ k,ii is the kth Gaussian component x i Its own variance, i and j are the indices of the dimensions of the original high-dimensional random variables of the power system; It is the variance of the kth Gaussian component after the original high-dimensional random variable data set of renewable energy output or load is aggregated.

[0059] S32. If the required power system data aggregation variable dimension is n, for the kth Gaussian component, obtaining the mean matrix and the covariance matrix includes the following steps:

[0060] S321. Solving the mean matrix: Each element in the mean matrix is ​​the sum of the means of several original high-dimensional random variables corresponding to the aggregate variable;

[0061] If the m-dimensional power system data (new energy output or load) is evenly distributed to the n-dimensional aggregate variables, its mean matrix is ​​expressed as:

[0062]

[0063] Among them, μ k,Y Represents the mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μk,i is the original variable x in the kth Gaussian component i The mean of .

[0064] In this embodiment, 9-dimensional wind farm output data is taken as an example. If the 9-dimensional wind farm output data is evenly distributed into 3-dimensional aggregate variables, its mean matrix is ​​expressed as:

[0065]

[0066] Among them, μ k,Y，9-3 represents the 3D mean matrix of the kth Gaussian component after the 9-dimensional wind farm output data is aggregated into 3 dimensions, μ k,i is the original variable x in the kth Gaussian component i The mean of .

[0067] S322, variance σ of aggregate variable y 2 Solution: For the kth Gaussian component, the i-th aggregate variable y i The variance of can be expressed as:

[0068]

[0069] in, is the i-th aggregate variable y i The variance of the kth Gaussian component, c ij is a 0-1 variable, which means: if the aggregate variable y i contains random variables x j , then the y i The following x j The corresponding coefficient c ij The value is 1, otherwise it is 0. il Similarly; Cov(x k,j ,x k,l ) represents the original high-dimensional random variable x in the kth Gaussian component j With x l The covariance between them, j and l are the indices of the dimensions of the original high-dimensional random variables of the power system.

[0070] S323, the correlation coefficient ρ between each two aggregated variables in the kth Gaussian component k,ab for:

[0071]

[0072] Among them, ρ k,ab represents the aggregate variable y in the kth Gaussian component a With y b The correlation coefficient between k,a ,y k,b ) is the aggregate variable y in the kth Gaussian componenta With y b The covariance between is the ath aggregate variable y a The variance of the kth Gaussian component in , is the bth aggregate variable y b The variance of the kth Gaussian component in aj With c bl Both are 0-1 variables, with the same meaning as c ij Same reason.

[0073] S324, solving the covariance matrix: combining the variance and correlation coefficient obtained in the above steps to obtain the covariance matrix corresponding to the k-th Gaussian component:

[0074]

[0075] Among them, Σ k,Y is the covariance matrix of the kth Gaussian component after aggregation, is the variance of the kth Gaussian element in the first aggregate variable y1, is the variance of the kth Gaussian element in the second aggregate variable y2, is the nth aggregate variable y n The variance of the kth Gaussian element in k,1 is the standard deviation of the kth Gaussian element in the first aggregate variable y1, σ k,2 is the standard deviation of the kth Gaussian element in the second aggregate variable y2, σ k,n is the nth aggregate variable y n The standard deviation of the kth Gaussian element in k,12 is the correlation coefficient between the aggregate variables y1 and y2 in the kth Gaussian component, ρ k,1n is the aggregate variable y1 and y in the kth Gaussian component n The correlation coefficient between .

[0076] S4, the mean and covariance results μ after aggregation k,Y and Σ k,Y , we can get the aggregate variable Y=(y1,y2,...,y n ) T The joint probability distribution model is:

[0077]

[0078] Among them, P(Y) is the GMM probability density function of the aggregated variable of the original high-dimensional random variable of the new energy output or load after dimensionality reduction aggregation.

[0079] The present invention also provides a high-dimensional uncertainty aggregation modeling system for power systems based on a Gaussian mixture model, which includes two stages: original uncertainty distribution characterization and GMM-based aggregation, wherein:

[0080] The original uncertainty distribution characterization unit is used to fit the original high-dimensional random variables of the power system using GMM, and to characterize the original high-dimensional random variables of the power system in a specific analytical form to obtain a fitted GMM model;

[0081] The GMM-based aggregation unit is used to calculate the aggregation model parameters based on the required aggregation variable dimensions using the fitted GMM model parameters, and integrate them into the aggregation model results.

[0082] The present invention also provides an electronic device for storing and executing the method, the device comprising:

[0083] A memory storing executable program code;

[0084] a processor coupled to the memory;

[0085] The processor calls the executable program code stored in the memory to execute the steps of the high-dimensional uncertainty aggregation modeling method for the power system based on the Gaussian mixture model.

[0086] The present invention also provides a computer-readable storage medium for storing and executing the method, wherein the computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the steps of high-dimensional uncertainty aggregation modeling of the power system based on the Gaussian mixture model.

[0087] Taking the IEEE 39-bus system of the power system as an example, five traditional units are replaced with new energy units, and the 20-dimensional wind farm measured output data is reduced and aggregated into 5-dimensional data and substituted into five new energy units, and the system probabilistic power flow analysis is performed. First, the GMM aggregation results are compared with the Monte Carlo method to verify the accuracy of the GMM method. Figure 2 (a) to (d) show the comparison of the probability density curves of the GMM aggregation results and the Monte Carlo method MC sampling results of different times. It can be found that the Monte Carlo method is getting closer and closer to the results obtained by the GMM method as the number of sampling increases, so the accuracy and precision of the GMM method are guaranteed. The GMM aggregation results are then substituted into the system probability power flow calculation and compared with the power flow results obtained from the original data. Figure 3 (a) and (b) show the probability results of node voltage and phase angle of a certain node under two data sources. It can be found that the GMM aggregation method also has extremely high accuracy in system probabilistic power flow analysis.

[0088] The technical means disclosed in the scheme of the present invention are not limited to the technical means disclosed in the above-mentioned implementation mode, but also include technical schemes composed of any combination of the above-mentioned technical features. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications are also regarded as the protection scope of the present invention.

Claims

1. A high-dimensional uncertainty aggregation modeling method for power system based on Gaussian mixture model, characterized in that: The method is implemented in two stages: original uncertainty distribution characterization and GMM-based aggregation. The original uncertainty distribution characterization is that the power system data is the original high-dimensional random variable. The GMM is used to fit the original high-dimensional random variables of the power system. The original high-dimensional random variables of the power system are represented in a specific analytical form to obtain a fitted GMM model. The GMM-based aggregation uses the fitted GMM model parameters to calculate the aggregation model parameters based on the required aggregation variable dimension, and integrates them into the aggregation model result.

2. The high-dimensional uncertainty aggregation modeling method for power system based on Gaussian mixture model according to claim 1 is characterized in that: GMM is used to fit the original high-dimensional random variables of the power system, and the original high-dimensional random variables of the power system are represented in a specific analytical form, including: Power system data includes renewable energy output or load; The original high-dimensional random variable data set X of new energy output or load is X=(x1, x2, ..., x d ) T GMM is used for fitting, and the fitting is a GMM model; x i is the original high-dimensional random variable of the power system of the i-th dimension, i = 1, 2, ..., d, d is the total dimension of the original high-dimensional random variable of the power system, T is the transpose of the matrix; the GMM model parameters obtained by fitting: the number of Gaussian components k, the weight coefficient of each component ω k , mean matrix μ k and the covariance matrix Σ k , the uncertainty distribution of the original high-dimensional random variable data set of renewable energy output or load is characterized by a Gaussian mixture model: Among them, P(X) is the GMM probability density function of the original high-dimensional random variable data set of new energy output or load, K is the total number of Gaussian components of the GMM model, N k is the probability density of the kth Gaussian component in the GMM model.

3. The method for high-dimensional uncertainty aggregation modeling of power system based on Gaussian mixture model according to claim 2 is characterized in that: The parameter estimation of the GMM model of the original high-dimensional random variable of renewable energy output or load is achieved through the expectation maximization algorithm, and the optimal number of Gaussian components of the GMM model is determined by the elbow method, the silhouette coefficient method or the Bayesian information criterion.

4. The method for high-dimensional uncertainty aggregation modeling of power system based on Gaussian mixture model according to claim 1 is characterized in that: GMM-based variable aggregation includes: (1) Determine the required dimension n of the power system aggregate variables and the original high-dimensional random variables contained in each aggregate variable; (2) According to whether the aggregated variable of renewable energy output or load in actual power system analysis needs to obtain a one-dimensional or n-dimensional result, the mean matrix μ of each Gaussian component parameter in the GMM model fitted with the original high-dimensional random variable of renewable energy output or load is calculated. k and the covariance matrix Σ k Transformed into the aggregated result μ by analytical calculation k,Y and σ k,Y Or μ k,Y and Σ k,Y ;μ k,Y is the one-dimensional mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, is the variance of the kth Gaussian component of the original high-dimensional random variable data set of renewable energy output or load after aggregation; μ k,Y Represents the mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, Σ k,Y is the covariance matrix of the kth Gaussian component after aggregation; (3) The mean and covariance results μ after aggregation are obtained k,Y and σ k,Y Or μ k,Y and Σ k,Y , and the joint probability distribution model of the aggregated variables of the power system after dimensionality reduction and aggregation is obtained.

5. The method for high-dimensional uncertainty aggregation modeling of power system based on Gaussian mixture model according to claim 4 is characterized in that: In step (2), if the dimension of the required power system aggregate variable is one-dimensional, for the kth Gaussian component of the aggregate variable, its corresponding mean and variance are expressed as: Among them, μ k,Y is the one-dimensional mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μ k,i is the kth Gaussian component x i The mean value of x i is the original high-dimensional random variable of the power system of the i-th dimension, d is the total dimension of the original high-dimensional random variables of the power system; Σ k,ij is the kth Gaussian component x i With x j The covariance of x j is the j-th dimension of the original high-dimensional random variable of the power system. If i=j, then x i Its own variance, that is, ∑ k,ii is the kth Gaussian component x i Its own variance, i and j are the indices of the dimensions of the original high-dimensional random variables of the power system; It is the variance of the kth Gaussian component after the original high-dimensional random variable data set of renewable energy output or load is aggregated.

6. The method for high-dimensional uncertainty aggregation modeling of power system based on Gaussian mixture model according to claim 1 is characterized in that: In step (2), if the dimension of the required power system aggregate variable is n, for the kth Gaussian component of the aggregate variable, each element in its mean column vector is the sum of the means of several random variables corresponding to the aggregate variable; the calculation of the covariance matrix includes the following steps: (21) Solving the mean matrix: Each element in the mean matrix is ​​the sum of the means of several original high-dimensional random variables corresponding to the aggregate variable; If the m-dimensional power system data is evenly distributed into n-dimensional aggregate variables, its mean matrix is ​​expressed as: Among them, μ k,Y Represents the mean matrix of the kth Gaussian component after the aggregation of the original high-dimensional random variable data set of renewable energy output or load, μ k,i is the original variable x in the kth Gaussian component i The mean of (22) Solution for the variance of the aggregate variable y: For the kth Gaussian component, the i-th aggregate variable y i The variance of is expressed as: in, is the i-th aggregate variable y i The variance of the kth Gaussian component, c ij is a 0-1 variable, which means: if the aggregate variable y i contains random variables x j , then the y i The following x j The corresponding coefficient c ij The value is 1, otherwise it is 0. il Similarly; Cov(x k,j ,x k,l ) represents the original high-dimensional random variable x in the kth Gaussian component j With x l The covariance between them, j and l are the indices of the dimensions of the original high-dimensional random variables of the power system; (23) The correlation coefficient ρ between each two aggregated variables in the kth Gaussian component k,ab for: Among them, ρ k,ab represents the aggregate variable y in the kth Gaussian component a With y b The correlation coefficient between k,a ,y k,b ) is the aggregate variable y in the kth Gaussian component a With y b The covariance between is the a-th aggregate variable y a The variance of the kth Gaussian component in , is the bth aggregate variable y b The variance of the kth Gaussian component in aj With c bl Both are 0-1 variables, with the same meaning as c ij Similarly; (24) Solving the covariance matrix: Combine the variance and correlation coefficient obtained in the above steps to obtain the covariance matrix corresponding to the kth Gaussian component: Among them, Σ k,Y is the covariance matrix of the kth Gaussian component after aggregation, is the variance of the kth Gaussian element in the first aggregate variable y1, is the variance of the kth Gaussian element in the second aggregate variable y2, is the nth aggregate variable y n The variance of the kth Gaussian element in k,1 is the standard deviation of the kth Gaussian element in the first aggregate variable y1, σ k,2 is the standard deviation of the kth Gaussian element in the second aggregate variable y2, σ k,n is the nth aggregate variable y n The standard deviation of the kth Gaussian element in k,12 is the correlation coefficient between the aggregate variables y1 and y2 in the kth Gaussian component, ρ k,1n is the aggregate variable y1 and y in the kth Gaussian component n The correlation coefficient between .

7. The method for high-dimensional uncertainty aggregation modeling of power system based on Gaussian mixture model according to claim 1 is characterized in that: The joint probability distribution model of power system aggregate variables after dimensionality reduction and aggregation is: Among them, P(Y) is the GMM probability density function of the power system data aggregation variable after dimensionality reduction aggregation.

8. A high-dimensional uncertainty aggregation modeling system for power systems based on Gaussian mixture models, characterized in that: It includes two stages: original uncertainty distribution representation and GMM-based aggregation, where: The original uncertainty distribution characterization unit is used to fit the original high-dimensional random variables of the power system using GMM, and to characterize the original high-dimensional random variables of the power system in a specific analytical form to obtain a fitted GMM model; The GMM-based aggregation unit is used to calculate the aggregation model parameters based on the required aggregation variable dimensions using the fitted GMM model parameters, and integrate them into the aggregation model results.

9. An electronic device, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps of the high-dimensional uncertainty aggregation modeling method for power systems based on Gaussian mixture models as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, which, when called, are used to execute the steps of high-dimensional uncertainty aggregation modeling of power systems based on Gaussian mixture models as described in any one of claims 1-7.

Images