Regional wind power modeling method, device, storage medium and equipment

Abstract

The present application provides a regional wind power modeling method, including: obtaining the historical power generation of several wind farms in a set area, performing linear correlation analysis and clustering processing on the wind farms in the area based on each of the historical power generation, and determining multiple equivalent wind farms in the set area; according to each of the equivalent wind farms, using the orthogonal series method, establishing the edge density function of each equivalent wind farm; according to the edge density function of each of the equivalent wind farms, using the Pair‑Copula function method, establishing a wind power joint probability model for the set area. Through the above method, the number of wind farms in the region is reduced, the modeling dimension is reduced, and the influence of the boundary effect of random variables is avoided, thereby improving the accuracy of the probability modeling of the correlation of wind farm power generation.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a method, device, storage medium and equipment for regional wind power modeling. Background Art

[0002] Wind power generation is characterized by volatility and intermittency. Therefore, the power generated by wind farms is highly random, intermittent, and has no power regulation capability, which results in wind power not being able to be controlled at will. With the continuous growth of wind power installed capacity and grid-connected scale, the random volatility of wind farm power generation has a huge impact on the power system. In addition, there are usually multiple wind farms connected to the power system in a region. When analyzing the regional power system, people are often only concerned with the overall changes in the power generation of wind farms in the region, and simply equate all wind farms to a random variable, ignoring the impact between wind farms.

[0003] With the large-scale development and utilization of wind power generation, the integration of wind farms into the power system is a subsequent development trend. In order to more clearly understand the impact of wind farm power generation in the region on the power system, it is a research trend for the large-scale application of wind power generation to establish a probability distribution model of the power generation of multiple wind farms in the region, predict the power generation of wind farms in a short period of time based on the model, generate corresponding wind power scenarios, and analyze the impact of wind farm integration into the power system on power generation. Therefore, how to establish a model of the power generation of multiple wind farms in the region is crucial to multiple application links of wind farm integration into the power system.

[0004] At present, the probabilistic modeling of the correlation between the power generation of multiple wind farms in a region mainly adopts the Copula function method combined with the one-dimensional kernel density estimation method and the multidimensional kernel density estimation method. Due to the large number of random variables, the above-mentioned multidimensional kernel density estimation method increases the complexity of the solution and reduces the accuracy of the results, and it is easy to fall into the curse of dimensionality in the case of high-dimensional operations. Furthermore, the above-mentioned kernel density estimation method will also be affected by the boundary effect of random variables, and the choice of window width has a great influence on the accuracy of the estimation results, and there is currently no unified method for selecting the window width. The defects of the above-mentioned prior art result in the low accuracy of the current probabilistic modeling of the correlation between the power generation of wind farms.

[0005] Therefore, how to improve the accuracy of probabilistic modeling of power generation correlation between wind farms is a technical problem that technical personnel in this field need to solve urgently. Summary of the invention

[0006] The present invention provides a regional wind power modeling method, device, storage medium and equipment, which can improve the accuracy of probabilistic modeling of power correlation between wind farms. The wind power model can be used to predict the power generation of wind farms in a short time, generate wind power scenarios, and analyze the impact of regional wind farms being incorporated into the power system on power generation and perform control adjustments.

[0007] In a first aspect, the present invention provides a regional wind power modeling method, comprising:

[0008] Acquire historical power generation of several wind farms in a set area, perform linear correlation analysis and clustering processing on the wind farms based on the historical power generation, and determine a plurality of equivalent wind farms in the set area, wherein the number of the equivalent wind farms is less than the number of the wind farms;

[0009] According to each of the equivalent wind farms, using the orthogonal series method, establishing the edge density function of each equivalent wind farm;

[0010] According to the edge density function of each of the equivalent wind farms, a wind power joint probability model of a set area is established by using a Pair-Copula function method.

[0011] In a second aspect, the present invention provides a regional wind power modeling device, comprising:

[0012] An equivalent wind farm establishment module is used to obtain wind power history samples of several wind farms in a set area, perform linear correlation analysis and clustering processing on the wind farms based on the historical power generation, and determine a plurality of equivalent wind farms in the set area, wherein the number of the equivalent wind farms is less than the number of the wind farms;

[0013] An edge density function establishment module is used to establish edge density functions of each equivalent wind farm using a non-parametric orthogonal series method according to each of the equivalent wind farms;

[0014] The wind power joint probability model establishment module is used to establish a wind power joint probability model of a set area according to the edge density function of each of the equivalent wind farms using the Pair-Copula function method.

[0015] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a method for regional wind power modeling according to the first aspect are implemented.

[0016] In a fourth aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes a regional wind power modeling method according to the first aspect when executing the computer program.

[0017] The beneficial effects of adopting the above technical solution are as follows: a regional wind power modeling method of the present invention obtains equivalent wind farms by linear correlation analysis and clustering processing of wind farms in the region, establishes the edge density function of equivalent wind farms by using the orthogonal series method, and then establishes the wind power joint probability model of the region by using the Pair-Copula function method. Through the above method, the number of wind farms in the region is reduced, the modeling dimension is reduced, and the influence of the boundary effect of random variables is avoided, thereby improving the accuracy of the probability modeling of the correlation of wind farm power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0019] Figure 1 A flowchart of the regional wind power modeling method is shown;

[0020] Figure 2 A schematic diagram of the process of correlation analysis and clustering of wind farms in the regional wind power modeling method;

[0021] Figure 3 It is a flowchart of the Pair-Copula function of the D-vine structure;

[0022] Figure 4 A schematic diagram of the modules of the device for modeling regional wind power;

[0023] Figure 5 Schematic diagram of the modules for regional wind power modeling device. DETAILED DESCRIPTION

[0024] The technical solutions 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 only 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. In order to explain the present invention in more detail, a regional wind power modeling method, device, storage medium and equipment provided by the present invention are specifically described below in conjunction with the drawings.

[0025] With the large-scale development and utilization of new energy sources such as wind power generation, the integration of wind farms into the power system is a future development trend. In order to more clearly understand the impact of wind farm power generation in the region on the power system, a model of power generation of multiple wind farms in the region is established. According to the model, the power generation of wind farms in a short period of time is predicted, the corresponding wind power scenarios are generated, and the impact of wind farm integration into the power system on power generation is analyzed and adjusted. This is a research trend for large-scale application of wind power generation. Therefore, how to establish a model of power generation of multiple wind farms in the region plays an important role in multiple application links of wind farm integration into the power system.

[0026] In order to more clearly illustrate the regional wind power modeling method, the embodiment of the present application provides a specific application scenario of the regional wind power modeling method. The application scenario includes the terminal device provided in the embodiment of the present application, and the terminal device can be various electronic devices with a display screen, including but not limited to smart phones and computer devices, wherein the computer device can be at least one of a desktop computer, a portable computer, a laptop computer, a tablet computer, and the like. The user operates the terminal device and issues a regional wind power modeling operation instruction, and the terminal device executes the regional wind power modeling method of the present invention. For the specific process, please refer to the embodiment of the regional wind power modeling method.

[0027] Based on this, a regional wind power modeling method is provided in the embodiment of the present application, and the method is applied to a terminal device as an example for explanation. Figure 1 The flowchart of the regional wind power modeling method shown in FIG. 1 includes the following steps:

[0028] Step S11: Obtain historical power generation of several wind farms in a set area, perform linear correlation analysis and clustering processing on the wind farms based on the historical power generation, and determine equivalent wind farms in the set area, wherein the number of equivalent wind farms is less than the number of wind farms.

[0029] There are usually multiple wind farms connected to the power system in a region. Due to the location of the wind farms and the influence of surrounding environmental factors, the power conditions of multiple wind farms in the region are not the same. Since there are a large number of wind farms in the region, if the edge density function is established for all wind farms in the region to obtain the regional wind power probability distribution model, it will lead to excessive computation and fall into the problem of dimensionality curse. Therefore, the present invention first clusters the wind farms in the region to obtain equivalent wind farms with fewer dimensions. The specific process of the clustering process is as follows:

[0030] As attached Figure 2 As shown, the specific steps of performing correlation analysis and clustering processing on wind farms provided by the embodiment of the present invention are as follows:

[0031] First, a linear correlation analysis is performed on all wind farms in the region. The specific steps of the linear correlation analysis include:

[0032] Step S1101: Based on the historical power generation of wind farms in a set area, traverse and calculate to obtain the correlation coefficient of power generation of any two wind farms in the area. For ease of understanding, specific parameters are cited for explanation. It is assumed that there are N wind farms in the area, which are recorded as W1, W2, ..., W N . The wind farm W i The power generated in the set period is normalized and recorded as P i = {P i,1 ,P i,2 ,…,P i,T}, where i=1, 2, ..., N, and T is a set period. The normalized power generation of the wind farm is the ratio of the actual power generation of the wind farm to the rated power of the wind farm.

[0033] Any two wind farms W in the region i and W j The expression of the power generation correlation coefficient is

[0034]

[0035] Among them, r ij For wind farm W i and W j Correlation coefficient of power generation, r ij The value range is [-1,1], For wind farm W i The average power generated during the T period, P i,k For wind farm W i The power generation in the kth period is: For wind farm W j The average power generated during the T period, P j,k For wind farm W j The power generated in the kth time period.

[0036] Step S1102: Calculate the correlation distance between any two wind farms according to the correlation coefficient of the power generation of the wind farms.

[0037] The expression of the correlation distance is specifically: R ij =1-|r ij |. Among them, R ij For wind farm W i and wind farm W j The correlation distance is in the range of [0,1]. ij The smaller the value, the greater the wind farm Wi and wind farm W j The stronger the linear correlation.

[0038] Then, according to the correlation distance R obtained by the linear correlation analysis of the wind farm ij The clustering process is continued. The clustering method used in this embodiment is the K-medoids clustering method, which is specifically:

[0039] Step S1111: obtaining a density index of the wind farms according to the correlation distance calculation between the wind farms.

[0040] The specific expression of the wind farm density index is: Where R kj For wind farm W k and wind farm W j The correlation distance.

[0041] Step S1112: Sort the wind farms according to their density indexes from small to large, select the first K wind farms with the smallest density index as initial clustering center points, and the remaining wind farms as initial clustering non-center points.

[0042] Specifically, assuming that the number of equivalent wind farms in the region is K, the wind farms in the region can be divided into K categories, which are recorded as Where 1≤K≤N. The wind farms are sorted according to their density index, and the first K wind farms with the smallest density index are selected as the initial center points of K clusters, denoted as O1, O2, ..., O K , the remaining wind farms are taken as the initial non-central points, denoted as S1, S2, …, S N-K .

[0043] Step S1113: For each wind farm corresponding to an initial non-central point, calculate the distance index between it and each central point, and add the initial non-central point to a category with the smallest distance index.

[0044] Among them, the specific expression of the distance index is: The smaller the distance index EC is, the better the clustering result is. Indicates The wind farm S corresponding to the initial non-central point in the class i The wind farm O corresponding to the center point i The size of the distance indicator.

[0045] Step S1114: adjusting the cluster center point to obtain a final clustering result, and obtaining an equivalent wind farm according to the clustering result.

[0046] Specifically, for clustering The initial center point O in iWith the initial non-central point S i , swap their positions, that is, let S i become The new center point of the cluster is calculated, and the new overall clustering evaluation index is obtained. The difference ΔEC between the new overall clustering evaluation index and the original overall clustering evaluation index is obtained. ΔEC is called the wind farm pair (S i ,O i ) The exchange cost of all wind farm pairs in each cluster is calculated. If all exchange costs are not less than 0, the clustering process ends and the clustering result at this time is the final clustering result. Otherwise, the center point of the cluster is repeatedly adjusted.

[0047] For the final K clusters Each cluster can be treated as an equivalent wind farm, and the equivalent wind farm is recorded as W′1, W′2, …, W′ K The power generation capacity of an equivalent wind farm is the sum of the power generation capacity of all wind farms included in the cluster, and the rated power of an equivalent wind farm is the sum of the rated power of all wind farms included in the cluster.

[0048] In this embodiment, equivalent wind farms are obtained through the above-mentioned correlation analysis and clustering processing, which greatly reduces the dimension of wind farms used in the subsequent modeling process, shortens the time used for subsequent modeling, and avoids the problem of dimensionality disaster. The method combining the above-mentioned correlation analysis and clustering processing retains the data of the power generation of all wind farms in the region, and to a great extent ensures that the subsequent model prediction or data analysis is of reference value. It should be noted here that the equivalent wind farm obtained by the correlation analysis and clustering processing is only a better equivalent processing method provided in this embodiment, and the equivalent processing method may also include selecting typical wind farms from wind farms in the region.

[0049] After performing equivalent processing on the wind farms in the region, the edge density function of the wind farms in the region can be obtained according to the equivalent wind farms, specifically:

[0050] Step S12: According to each of the equivalent wind farms, a marginal density function of each equivalent wind farm is established by using a non-parametric orthogonal series method.

[0051] Specifically, the equivalent wind farm W′ k The power generated in the set period is normalized and recorded as P' k = {P′ k,1 ,P′ k,2 ,…,P′ k,T}, wherein k=1, 2, ..., K, and T is a set period. The normalization process of the power generation of the equivalent wind farm is to obtain the ratio of the actual power generation of the equivalent wind farm to the rated power of the equivalent wind farm.

[0052] The specific expression of the edge density function of the equivalent wind farm obtained by the non-parametric orthogonal series method is:

[0053]

[0054] Among them, W′ k is the kth equivalent wind farm, P′ k is the power generation of the kth equivalent wind farm, φ j (P′ k ) is the Fourier basis of the kth equivalent wind farm, β j are the coefficients of the Fourier basis, and J is the cut-off point.

[0055] The specific expressions of the above parameters are:

[0056]

[0057]

[0058] T' is greater than The smallest integer such that when J=J0, δ(J0) is the smallest.

[0059] Thus, the marginal density functions of all equivalent wind farms in the set area can be obtained, and the joint probability distribution model of wind farm power generation can be obtained by further calculating according to the marginal density function.

[0060] Step S13: According to the edge density function of each of the equivalent wind farms, a joint probability distribution model of the power generation in the set area is established by using the Pair-Copula function method.

[0061] Specifically, the expression of the power generation joint probability model is:

[0062]

[0063] Among them, W′ k is the kth equivalent wind farm, P′ k is the power of the kth equivalent wind farm, C j,j+i|j+1,…,j+i-1 P′ is the j-th binary Copula function of the i-1th layer obtained by fitting. j is the power generation of the jth equivalent wind farm, is the marginal density function of the equivalent wind farm.

[0064] The specific expressions of the above parameters are:

[0065] Among them U k (P′ k) is the distribution function of the wind power of the kth equivalent wind farm, k = 1, 2, ..., K; In addition, U k (P′ k ) can also be obtained based on the sample approximation, the specific expression is I(P′ k,i ≤P′ k ) is an indicative function. When the condition P′ is satisfied k,i ≤P′ k , I=1, otherwise I=0.

[0066] Where j = 1, 2, ..., K-1;

[0067] , where j=1, 2, …, Ki, i=3, 4, …, K-1.

[0068] The Pair-Copula function method described in this embodiment is specifically a D-vine structure Pair-Copula function method. Figure 3 , the D-vine structure Pair-Copula function method is specifically as follows: the first layer is set as the marginal distribution function of each equivalent wind farm, and the adjacent marginal distribution functions are combined in pairs to obtain the joint probability density function of the next layer; then the obtained joint probability density functions are combined in pairs, layer by layer, until only one optimal joint probability density function is generated after the joint probability density functions are combined. In order to understand it more clearly, the D-vine structure Pair-Copula function is illustrated as follows: the first layer is set with the marginal distribution functions of K equivalent wind farms, which are respectively recorded as U1, U2, ..., U K , the adjacent marginal distribution functions are combined to obtain the joint probability density function of the second layer, such as U1 and U2 are combined to obtain the joint probability density function of the second layer C 1,2 , U2 and U3 are combined to obtain the joint probability density function C of the second layer 2,3 , and so on, U K-1 and U K Combined to get the joint probability density function C of the second layer K-1,K ; Recursively layer by layer, the joint probability density functions of the second layer are combined to obtain the joint probability density function of the third layer, such as C 1,2 and C 2,3 Combined to get the joint probability density function C of the third layer 1,3|2 , and so on, C K-2,K-1 and C K-1,K Combined to get the joint probability density function C of the third layer K-2,K|K-1 ; Continue recursively layer by layer until the optimal joint probability density function C is obtained at the K-1th layer 1,K|2,3,…,K-1 .

[0069] First, calculate the empirical Copula function C n (u,v), the specific expression is, in is an indicative function. When the condition H(x i )≤u or G(y i )≤v, I=1, otherwise I=0; (x i ,y i ) is a sample of the variable (X,Y) from which the Copula function is constructed, u,v∈[0,1], H(x i ) is x i The distribution function of G(y i ) is y i The distribution function of , n is the number of samples.

[0070] Then, two types of Copula functions, the commonly used elliptic Copula and Archimedean Copula, are selected, including but not limited to normal-Copula, t-Copula, Gumbel Copula, Clayton Copula, and Frank Copula. The one with the smallest Euclidean distance between the Copula function to be fitted and the empirical Copula function is selected as the optimal Pair-Copula function.

[0071] Among them, the fitted Copula function C p (u,v) and the empirical Copula function C n The Euclidean distance between (u,v) is specifically expressed as,

[0072] According to the above method, the binary Copula function of each layer can be fitted, and then the joint probability model of power generation considering the correlation between regions f(P′1,P′2,…,P′ K ).

[0073] It should be noted that the D-vine structure Pair-Copula function method used in this embodiment is a better one of the Copula function methods. The Copula function method can also use the C-vine structure Pair-Copula function method.

[0074] It should be understood that although Figure 1 and Figure 2 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0075] The above-mentioned embodiment disclosed in the present application describes in detail a regional wind power modeling method. The above-mentioned method disclosed in the present application can be implemented by various forms of equipment. Therefore, the present application also discloses a wind power modeling device corresponding to the above-mentioned method. The specific embodiments are given below for detailed description.

[0076] As the instruction manual Figure 4 As shown, the present invention provides a regional wind power modeling device, comprising:

[0077] The equivalent wind farm establishment module 201 is used to obtain the historical power generation of several wind farms in a set area, perform equivalent processing on the wind farms based on the historical power generation, and determine multiple equivalent wind farms in the set area, wherein the number of the equivalent wind farms is less than the number of the wind farms.

[0078] Specifically, as attached Figure 5 As shown, the equivalent wind farm establishing module 201 includes a linear correlation analysis module 2011 and a clustering module 2022 .

[0079] The linear correlation analysis module 2011 is used to traverse and calculate the correlation coefficient of the power generation of any two wind farms in a set area according to the historical power generation of wind farms in the area, and calculate the correlation distance between any two wind farms according to the correlation coefficient of the power generation of the wind farms.

[0080] The specific operation inside the linear correlation analysis module 2011 is:

[0081] There are N wind farms in the assumed area, denoted as W1, W2, ..., W N . The wind farm W i The power generated in the set period is normalized and recorded as P i = {P i,1 ,P i,2 ,…,P i,T}, where i=1, 2, ..., N, and T is a set period. The normalized power generation of the wind farm is the ratio of the actual power generation of the wind farm to the rated power of the wind farm.

[0082] Any two wind farms W in the region i and W jThe expression of the power generation correlation coefficient is

[0083]

[0084] Among them, r ij For wind farm W i and W j Correlation coefficient of power generation, r ij The value range is [-1,1], For wind farm W i The average power generated during the T period, P i,k For wind farm W i The power generation in the kth period is: For wind farm W j The average power generated during the T period, P j,k For wind farm W j The power generated in the kth time period.

[0085] The expression of the correlation distance is specifically: R ij =1-|r ij |. Among them, R ij For wind farm W i and wind farm W j The correlation distance is in the range of [0,1]. ij The smaller the value, the greater the wind farm W i and wind farm W j The stronger the linear correlation.

[0086] The clustering module 2022 is used to calculate the density index of the wind farms according to the correlation distance between the wind farms; sort the wind farms from small to large according to their density index, select the first K wind farms with the smallest density index as the initial center points of the cluster, and the remaining wind farms as the initial non-center points of the cluster; perform clustering iteration processing according to the distance between the initial non-center points of the cluster and the initial center point of the cluster, and obtain K equivalent wind farms.

[0087] The clustering module 2022 specifically adopts K-medoids clustering processing, and its internal operation is specifically as follows:

[0088] The density index of the wind farm is calculated based on the correlation distance between the wind farms. The specific expression of the density index of the wind farm is: Where R kj For wind farm W k and wind farm W j The correlation distance.

[0089] Assuming the number of equivalent wind farms in the region is K, the wind farms in the region can be divided into K categories, denoted as Where 1≤K≤N. The wind farms are sorted according to their density index, and the first K wind farms with the smallest density index are selected as the initial center points of K clusters, denoted as O1, O2, ..., O K , the remaining wind farms are taken as the initial non-central points, denoted as S1, S2, …, S N-K .

[0090] For each wind farm corresponding to the initial non-center point, calculate the distance index between it and each center point, and add the initial non-center point to the category with the smallest distance index. The specific expression of the distance index is: The smaller the distance index EC is, the better the clustering result is. Indicates The wind farm S corresponding to the initial non-central point in the class i The wind farm O corresponding to the center point i The size of the distance indicator.

[0091] The cluster center point is adjusted to obtain the final clustering result, and the equivalent wind farm is calculated based on the clustering result. The initial center point O in i With the initial non-central point S i , swap their positions, that is, let S i become The new center point of the cluster is calculated, and the new overall clustering evaluation index is obtained. The difference ΔEC between the new overall clustering evaluation index and the original overall clustering evaluation index is obtained. ΔEC is called the wind farm pair (S i ,O i ) The exchange cost of all wind farm pairs in each cluster is calculated. If all exchange costs are not less than 0, the clustering process ends and the clustering result at this time is the final clustering result. Otherwise, the center point of the cluster is repeatedly adjusted.

[0092] For the final K clusters Each cluster can be treated as an equivalent wind farm, and the equivalent wind farm is recorded as W′1, W′2, …, W′ K The power generation capacity of an equivalent wind farm is the sum of the power generation capacity of the wind farms included in the cluster, and the rated power of an equivalent wind farm is the sum of the rated power of the wind farms included in the cluster.

[0093] The edge density function establishment module 202 is used to establish the edge density function of each equivalent wind farm according to each of the equivalent wind farms by using a non-parametric orthogonal series method.

[0094] Specifically, the edge density function establishment module 202 performs the following internal operations:

[0095] The equivalent wind farm W′ k The power generated in the set period is normalized and recorded as P' k = {P′ k,1 ,P′ k,2 ,…,P′ k,T}, where k = 1, 2, ..., K, and T is a set period. The power generation normalization of the equivalent wind farm is to obtain the ratio of the actual power generation of the equivalent wind farm to the rated power of the equivalent wind farm. The specific expression of the edge density function of the equivalent wind farm obtained by the non-parametric orthogonal series method is:

[0096]

[0097] Among them, W′ k is the kth equivalent wind farm, P′ k is the power generation of the kth equivalent wind farm, φ j (P′ k ) is the Fourier basis of the kth equivalent wind farm, β j are the coefficients of the Fourier basis, and J is the cut-off point.

[0098] The specific expressions of the above parameters are:

[0099]

[0100]

[0101] T' is greater than The smallest integer such that when J=J0, δ(J0) is the smallest.

[0102] Thus, the marginal density functions of all equivalent wind farms in the set area can be obtained, and the joint probability distribution model of wind farm power generation can be obtained by further calculating according to the marginal density function.

[0103] The wind power joint probability model building module 203 is used to build a wind power joint probability model of a set area according to the edge density function of each equivalent wind farm by using the Pair-Copula function method.

[0104] Specifically, the wind power joint probability model building module 203 performs the following operations:

[0105] The expression of the power generation joint probability model is:

[0106]

[0107] W′ k is the kth equivalent wind farm, P′k is the power of the kth equivalent wind farm, C j,j+i|j+1,…,j+i-1 P′ is the j-th binary Copula function of the i-1th layer obtained by fitting. j is the power generation of the jth equivalent wind farm, is the marginal density function of the equivalent wind farm.

[0108] The specific expressions of the above parameters are:

[0109] Among them U k (P′ k ) is the distribution function of the wind power of the kth equivalent wind farm, k = 1, 2, ..., K; In addition, U k (P′ k ) can also be obtained based on the sample approximation, the specific expression is I(P′ k,i ≤P′ k ) is an indicative function. When the condition P′ is satisfied k,i ≤P′ k , I=1, otherwise I=0.

[0110] Where j = 1, 2, ..., K-1;

[0111] , where j=1, 2, …, Ki, i=3, 4, …, K-1.

[0112] The Copula function method is specifically a D-vine structure Pair-Copula function method. Starting from the first layer, the optimal Pair-Copula function type and parameters between multidimensional variables are determined layer by layer.

[0113] Calculate the empirical Copula function C n (u,v), the specific expression is, in is an indicative function. When the condition H(x i )≤u or G(y i )≤v, I=1, otherwise I=0; (x i ,y i ) is a sample of the variable (X,Y) from which the Copula function is constructed, u,v∈[0,1], H(x i ) is x i The distribution function of G(y i ) is y idistribution function, n is the number of samples. Two types of Copula functions, the commonly used elliptic Copula and Archimedean Copula, are selected, including normal-Copula, t-Copula, Gumbel Copula, Clayton Copula, and Frank Copula. The one with the smallest Euclidean distance between the Copula function to be fitted and the empirical Copula function is selected as the optimal Pair-Copula function. Among them, the fitted Copula function C p (u,v) and the empirical Copula function C n The Euclidean distance between (u,v) is specifically expressed as,

[0114] According to the above method, the binary Copula function of each layer can be fitted, and then the joint probability model of power generation considering the correlation between regions f(P′1,P′2,…,P′ K ).

[0115] In addition, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method steps in the above-mentioned regional wind power modeling method embodiment are implemented.

[0116] The computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM (erasable programmable read-only memory), a hard disk or a ROM. Optionally, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program codes for executing any of the method steps in the above method. These program codes may be read from or written into one or more computer program products, and the program codes may be compressed in an appropriate form.

[0117] Furthermore, the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor executes the method steps in the embodiment of the regional wind power modeling method when executing the computer program.

[0118] The computer device includes a memory, a processor, and one or more computer programs, wherein the one or more computer programs can be stored in the memory and configured to be executed by one or more processors, and the one or more application programs are configured to execute the above-mentioned wind power modeling method.

[0119] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect the various parts of the entire computer device, and executes various functions of the computer device and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor can integrate one or more combinations of a central processing unit (CPU), a reporting verifier (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor, but may be implemented separately through a communication chip.

[0120] The memory may include a random access memory (RAM) or a read-only memory (ROM). The memory may be used to store instructions, programs, codes, code sets or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may also store data created by the terminal device during use, etc.

[0121] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A regional wind power modeling method, characterized in that: include: Acquire historical power generation of several wind farms in a set area, perform linear correlation analysis and clustering processing on the wind farms based on the historical power generation, and obtain a plurality of equivalent wind farms in the set area, wherein the number of the equivalent wind farms is less than the number of the wind farms; According to each of the equivalent wind farms, using a non-parametric orthogonal series method, an edge density function of each equivalent wind farm is established; According to the edge density function of each of the equivalent wind farms, a wind power joint probability distribution model of the set area is established by using a Pair-Copula function method; The performing linear correlation analysis on the wind farm comprises: According to the historical power generation of wind farms in the set area, the power generation correlation coefficient of any two wind farms in the set area is obtained by traversal calculation; Calculating the correlation distance between any two wind farms according to the power generation correlation coefficients of the wind farms; The clustering process is a K-medoids clustering process, including: Obtaining a density index of the wind farms according to the correlation distance between the wind farms; Sorting the wind farms according to their density index from small to large, selecting the first K wind farms with the smallest density index as the initial center points of the cluster, and the remaining wind farms as the initial non-center points of the cluster; Performing clustering iteration processing according to the distance between the initial non-center point of the cluster and the initial center point of the cluster, to obtain K equivalent wind farms; wherein K is the number of equivalent wind farms; The establishing of the edge density function of the equivalent wind farm by using a non-parametric orthogonal series method according to the equivalent wind farm comprises: The marginal density function expression of the equivalent wind farm is: ； in, is the kth equivalent wind farm, is the power generation of the kth equivalent wind farm, is the Fourier basis of the kth equivalent wind farm, are the coefficients of the Fourier basis, and J is the cut-off point.

2. The regional wind power modeling method according to claim 1, characterized in that: The expression of the wind power joint probability model is: ； in, is the kth equivalent wind farm, is the power of the kth equivalent wind farm, is the power of the jth equivalent wind farm, To fit the j-th binary Copula function of the i-1th layer, is the marginal density function of the equivalent wind farm.

3. A regional wind power modeling device, characterized in that: include: An equivalent wind farm establishment module is used to obtain historical power generation of several wind farms in a set area, perform linear correlation analysis and clustering processing on the wind farms based on each of the historical power generation, and determine a plurality of equivalent wind farms in the set area, wherein the number of the equivalent wind farms is less than the number of the wind farms; An edge density function establishment module is used to establish edge density functions of each equivalent wind farm using a non-parametric orthogonal series method according to each of the equivalent wind farms; A wind power joint probability model establishment module is used to establish a wind power joint probability model of a set area according to the edge density function of each equivalent wind farm using a Pair-Copula function method; The equivalent wind farm establishment module includes a linear correlation analysis module, which is used to traverse and calculate the correlation coefficient of the power generation of any two wind farms in the region according to the historical power generation of the wind farms in the set area, and calculate the correlation distance between any two wind farms according to the correlation coefficient; The equivalent wind farm establishment module also includes a clustering module, which is used to calculate the density index of the wind farm according to the correlation distance between the wind farms, sort the wind farms from small to large according to the density index, select the first K wind farms with the smallest density index as the initial center points of the cluster, and the remaining wind farms as the initial non-center points of the cluster, and perform clustering iteration processing according to the distance between the initial non-center points of the cluster and the initial center point of the cluster, so as to obtain K equivalent wind farms, wherein K is the number of equivalent wind farms; The establishing of the edge density function of the equivalent wind farm by using a non-parametric orthogonal series method according to the equivalent wind farm comprises: The marginal density function expression of the equivalent wind farm is: ； in, is the kth equivalent wind farm, is the power generation of the kth equivalent wind farm, is the Fourier basis of the kth equivalent wind farm, are the coefficients of the Fourier basis, and J is the cut-off point.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the regional wind power modeling method according to any one of claims 1 to 2 are implemented.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the regional wind power modeling method according to any one of claims 1 to 2 is performed.

Images