A coupling mechanism analysis method and device for master-slave coordination integration planning

Abstract

The present application relates to the technical field of main distribution network planning, and more particularly to a coupling mechanism analysis method and device for main-distribution collaborative integrated planning, which comprises the following steps: obtaining the edge distribution function and the probability density function of wind power plants, photovoltaic power plants and load power; calculating the Kendall-Tau coupling coefficient between each sample variable, and selecting a vine node; establishing a C-vine-Copula model; solving the sampling probability distribution value; converting the random number into an actual sample value; obtaining a sampling scenario set based on the C-vine-Copula coupling mechanism model; judging whether the data process continues; taking the minimum difference between the wind power plant output curve, the photovoltaic power plant output curve and the load curve as the objective function, using the K-means clustering algorithm and the silhouette coefficient as the clustering effect index to describe the similarity and closeness within the cluster and the dispersion degree between the clusters. The present application effectively handles the coupling problem between the main distribution network, and deeply mines and analyzes the coupling characteristics between the main network and the distribution network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of main distribution network planning, and particularly relates to a coupling mechanism analysis method and device for main-distribution collaborative integration planning. BACKGROUND

[0002] Under the "double carbon" target, the proportion of new energy will gradually increase, and its randomness and volatility will impact the operation of the power grid, which brings additional burden to the matching of the boundary power between the main grid and the distribution network. In the process of power grid construction in China, there are still many loopholes in the coordination planning of the main grid and the distribution network, which seriously restricts the further development of the power industry in China.

[0003] In related technologies, the grid planning of the distribution network focuses on the power balance and grid optimization of the lines within the grid, and does not consider the high-voltage main grid in depth, resulting in disconnection between main-distribution network planning and insufficient active support of the distribution network to the main network. Without global overall planning and target constraints, the rationality and long-term effectiveness of the planning scheme cannot be ensured.

[0004] In the above technology, due to the geographical proximity of distributed new energy and load areas and the high coupling of environmental factors, the coupling characteristics between the main grid and the distribution network are difficult to accurately identify and analyze, and this coupling problem cannot be effectively handled, so it is necessary to further explore and analyze the coupling characteristics between the main grid and the distribution network. SUMMARY

[0005] In view of at least one of the above technical problems, the present application provides a coupling mechanism analysis method and device for main-distribution collaborative integration planning, which establishes a coupling mechanism model and selects coupling scenarios by using a sampling method to analyze the coupling mechanism.

[0006] According to a first aspect of the present application, a coupling mechanism analysis method for main-distribution collaborative integration planning is provided, comprising the following steps:

[0007] Obtain the edge distribution function and the probability density function of the wind power plant, the photovoltaic power plant and the load power;

[0008] Calculate the Kendall-Tau coupling coefficient between each sample variable, and select the vine node;

[0009] Establish a C-vine-Copula model by calculating the first Euclidean distance between the fitted Copula function and the empirical Copula function;

[0010] Solve the sampling probability distribution value;

[0011] Convert the random number to the actual sample value by inverting the cumulative distribution function;

[0012] The probability distribution values ​​obtained by sampling based on the coupling mechanism model are transformed into the corresponding wind and solar load power, thus obtaining the sampling scene set based on the C-vine-Copula coupling mechanism model;

[0013] Using the historical output data of the master-supplier collaborative operation model, the second Euclidean distance from each data point to the initial cluster center is calculated. The data is then assigned to the cluster with the smallest second Euclidean distance according to the principle of minimizing the second Euclidean distance.

[0014] Determine whether to continue the data processing based on changes in cluster centers;

[0015] Using the minimum difference between the output curves and load curves of wind power plants and photovoltaic power plants as the objective function, the K-means clustering algorithm and silhouette coefficient are used as indicators to judge the clustering effect, describing the similarity and density within clusters and the dispersion between clusters.

[0016] This also includes establishing the objective function that minimizes the difference between the main distribution network output curve and the load curve, which is:

[0017]

[0018] In the formula, α k and α L k These represent the normalized rates of change in power generation and load of photovoltaic power plants, wind farms, and substations, respectively, δ. i I represents the change in the system's total historical power generation. T Total power generation of main and distribution networks, I C For load power;

[0019] The silhouette coefficient is used as an indicator to judge the clustering effect. The silhouette coefficient combines the clustering degree and the separation degree, and can well describe the similarity and compactness within clusters and the dispersion between clusters. The specific calculation formula is as follows:

[0020]

[0021] In the formula, SC is the overall silhouette coefficient of the clustered data, and a i Let b be the average distance between the i-th data point and other elements in the same cluster. i Let M be the average distance between the i-th data point and all data points in the nearest cluster to the next centroid, and M be the number of samples in the dataset.

[0022] In some embodiments of the present invention, kernel density estimation is used to obtain the edge distribution functions of wind power plants, photovoltaic power plants, and load power:

[0023]

[0024] where v n is the rated wind speed of the wind turbine, v in is the cut-in wind speed of the wind turbine, P w is the output power of the wind turbine, P nw is the rated power of the wind turbine, λ and c are shape parameters of the Weibull distribution, s is the light intensity, s max is the maximum light intensity, σ and β are shape parameters of the Beta distribution, and Γ(σ) is the gamma function.

[0025] In some embodiments of the present application, the probability density of the random variable sample is:

[0026]

[0027] where P w is the output power of the wind turbine, P s is the output power of the photovoltaic, P L is the load power.

[0028] In some embodiments of the present application, when the vine root node is selected, the Kendall-τ coupling coefficient between each sample two-variable is calculated:

[0029]

[0030] where n is the total number of data points, C is the number of point pairs with consistent order, and D is the number of point pairs with inconsistent order.

[0031] In some embodiments of the present application, the C-vine-Copula model is constructed by the following steps:

[0032] Determine the load distribution information of the region to be planned and the voltage level capacity ratio data of the region to be planned;

[0033] Based on the original network structure, fully considering the network and feeder transmission constraints, according to the load distribution characteristics and load development demand, a multi-voltage level distribution network planning model is constructed;

[0034] Calculate the first Euclidean distance between the fitted Copula function and the empirical Copula function;

[0035] Select the fitted Copula function with the smallest first Euclidean distance as the optimal Copula function, and calculate the new random variable sample layer by layer to establish the main-distribution integrated coupling mechanism model.

[0036] In some embodiments of the present application, the total capacity of the substation is also included, and the calculation method is as follows:

[0037] n G S g ≥RG,min ∑P,

[0038] In the formula, n G is the number of substations in the region, S g is the capacity of a single substation, R G,min is the minimum operating rate of the substation, and ∑P is the total power demand of all load points.

[0039] The distribution network planning model is as follows:

[0040]

[0041] In the formula, Z is the distribution network planning cost, L ij is the network construction cost coefficient from node i to node j, x ij is a binary decision variable, indicating the connection relationship between node i and node j.

[0042] The first Euclidean distance between the fitted Copula function and the empirical Copula function is fitted:

[0043]

[0044] In the formula, d is the first Euclidean distance of random variables X1 and X2, C(X1,X2) is the fitted Copula function, and C n (X1,X2) is the empirical Copula function.

[0045] In some embodiments of the present application, the sampling probability distribution value solving includes the following steps:

[0046] The sampling probability distribution value is solved based on the random variable {z1,z2,z3} matrix:

[0047]

[0048] In the formula, z i is a random variable, u i is the sampling probability distribution value corresponding to the random variable.

[0049] The known quantities z2 and z3 are substituted into the above formula, and the differential equation is solved, so that the value of the sampling point is obtained.

[0050] Main distribution field scenario set sampling:

[0051] The random number is converted into an actual sample value by inverting the cumulative distribution function:

[0052] x=F -1 (U),

[0053] In the formula, U=F(x) is the cumulative distribution function of variable x, U∈[0,1], and F -1It is the inverse function of the cumulative distribution function;

[0054] The probability distribution values ​​obtained by sampling based on the coupling mechanism model are transformed into the corresponding wind and solar load power, resulting in a sampling scene set based on the C-vine-Copula coupling mechanism model.

[0055] In some embodiments of the present invention, the historical output data of the main-supporting coordinated operation model is a data sample matrix X with a sampling interval of t, a sampling number of N, and a sampling object number of M. M×N :

[0056]

[0057] In the formula, V M,N This represents the Nth historical power output data of the Mth main distribution network;

[0058] Calculate the second Euclidean distance from each data point to the initial cluster center. Then, assign the data to the cluster with the smallest second Euclidean distance according to the principle of minimizing this distance. Finally, calculate the number of samples C in each new cluster. i ;

[0059] Based on the cluster center update formula, the cluster center point o in each new cluster is recalculated. i The cluster center update expression is:

[0060]

[0061] In the formula, x is the new cluster C i The sample data in the sample.

[0062] In some embodiments of the present invention, the decision to continue the data process is made based on the changes in the cluster centers. When the changes in the clusters are small enough to meet the set conditions or the number of iterations reaches the upper limit, the clustering operation can be terminated.

[0063] According to a second aspect of the invention, an apparatus is also provided, on which a computer program is stored, the computer program being capable of implementing the method as described above.

[0064] The beneficial effects of this invention are as follows: The invention constructs a master-distributor integrated coupling mechanism model based on the C-vine-copula theory, and uses K-means clustering and silhouette coefficient evaluation to screen out typical master-distributor collaborative operation scenarios. Furthermore, it uses historical output data to extract the output fluctuation characteristics of master-distributor collaborative integration, deeply explores the coupling characteristics, generates joint scenarios with master-distributor integrated coupling characteristics, and performs coupling mechanism analysis. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is a flowchart illustrating the steps of the coupling mechanism analysis method for the master-distributor coordinated integrated planning in this embodiment of the invention.

[0067] Figure 2 This is a flowchart illustrating the steps involved in constructing the C-vine-Copula model in an embodiment of the present invention.

[0068] Figure 3 This is a diagram illustrating the coupling characteristics of the scaled-down scenario 1 in the verification test of this invention.

[0069] Figure 4 This is a diagram illustrating the coupling characteristics of the scaled-down scenario 2 in the verification test of this invention.

[0070] Figure 5 This diagram illustrates the coupling characteristics of the scaled-down scenario 3 in the verification test of this invention. Detailed Implementation

[0071] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0072] It should be noted that when an element is referred to as being "fixed to" another element, it can be directly attached to the other element or there may be an intervening element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or there may be an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementation.

[0073] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0074] like Figure 1 The coupling mechanism analysis method for the master-distributor synergistic integrated planning shown includes:

[0075] S10: Obtain the edge distribution function and the probability density function of the wind power plant, the photovoltaic power plant and the load power;

[0076] S20: Calculate the Kendall-Tau coupling coefficient between each pair of variables of each sample, and select the tree root node;

[0077] S30: Establish the C-Fig-Copula model by calculating the first Euclidean distance between the fitted Copula function and the empirical Copula function;

[0078] S40: Solve the sampling probability distribution value;

[0079] S50: Convert the random number into the actual sample value by inverting the cumulative distribution function;

[0080] S60: Convert the probability distribution value obtained based on the coupling mechanism model sampling into the corresponding wind, light and load power, and obtain the sampling scenario set based on the C-Fig-Copula coupling mechanism model;

[0081] S70: Calculate the second Euclidean distance of each data to the initial clustering center point using the historical output data of the main and distribution collaborative operation model, and perform nearest allocation according to the principle of minimum second Euclidean distance to correspondingly allocate the data to the cluster with the minimum second Euclidean distance;

[0082] S80: Determine whether to continue the data process according to the change of the clustering center;

[0083] S90: Take the minimum difference value of the wind power plant, the photovoltaic power plant output curve and the load curve as the objective function, use the K-means clustering algorithm and the silhouette coefficient as the judgment clustering effect index to describe the similar tightness within the cluster and the dispersion degree between the clusters.

[0084] Further, the establishment of the objective function of the minimum difference value of the main and distribution network output curve and the load curve is as follows:

[0085]

[0086] In the formula, α k and α L k are the change rates of the normalized photovoltaic power plant, wind power plant and substation power and the load change rate, δ i is the change amount of the total historical power of the system, I T is the total power of the main and distribution network, I C is the load power;

[0087] The silhouette coefficient is used as a judgment clustering effect index, the silhouette coefficient combines the cohesion and separation of clustering, and can well describe the similar tightness in the cluster and the dispersion degree between clusters, and the specific calculation formula is as follows:

[0088]

[0089] In the formula, SC is the overall silhouette coefficient of the clustering data, a i is the average distance of the ith data from other elements in the same cluster, b i is the average distance of the ith data from all data in the nearest cluster of the next centroid, and M is the number of samples in the data set.a i The smaller, the closer the sample point i is to other points in the cluster to which it belongs; b i Reflects the separation degree between clusters, b i The greater, the farther the sample point i is from other clusters; the overall silhouette coefficient SC is in the range of [-1, 1], when SC is close to 1, it indicates that the clustering effect of the sample point i is very good, that is, the sample point i is very close to the points in the cluster to which it belongs, and the distance from other clusters is very far; when SC is close to-1, it indicates that the clustering effect of the sample point i is very poor, that is, the sample point i should be assigned to other clusters; when SC is close to 0, it indicates that the sample point i is at the edge of the cluster, and the clustering effect may not be obvious.

[0090] In some embodiments of the application, the kernel density estimation method is used to obtain the marginal distribution function of the wind power plant, the photovoltaic power plant and the load power:

[0091]

[0092] In the formula, v n is the rated wind speed of the wind turbine, v in is the cut-in wind speed of the wind turbine, P w is the output power of the wind turbine, P nw is the rated power of the wind turbine, λ and c are the shape parameters of the Weibull distribution, s is the illumination intensity, s max is the maximum illumination intensity, σ and β are the shape parameters of the Beta distribution, and Γ(σ) is the gamma function. The smooth kernel function is used to estimate the probability density function of the data, and the specific application mode is as follows: first, collect the historical data of the wind power plant, the photovoltaic power plant and the load power, calculate the probability density function of the data using the above kernel function, and obtain the marginal distribution function, which can be used for subsequent analysis, such as constructing a Copula model and generating a sampling scenario.

[0093] Further, the probability density of the random variable sample:

[0094]

[0095] where P w is the output power of the wind turbine, P s is the output power of the photovoltaic, P L is the load power.

[0096] In some embodiments of the present application, when selecting the vine root node, the Kendall-Tau coupling coefficient between each pair of variables of the samples is calculated:

[0097]

[0098] where n is the total number of data points, C is the number of point pairs consistent in order, and D is the number of point pairs inconsistent in order. The Kendall-Tau coupling coefficient is used to determine the strongest dependence between variables, thereby serving as the starting point for constructing the vine structure, selecting the variable pair with the highest or lowest coupling coefficient as the root node, and then continuing to construct the remaining part of the vine structure by iteratively selecting the next variable with the strongest dependence on the current structure.

[0099] In some embodiments of the present application, as shown in Figure 2 constructing a C-vine-Copula model includes the following steps:

[0100] S31: Determine the load distribution information of the to-be-planned area and the capacity-load ratio data of each voltage level of the to-be-planned area; collecting the load distribution information of the to-be-planned area includes collecting historical load data, including time series load data, load peak, load growth rate, etc., and then analyzing the statistical characteristics of the load data, such as mean, variance, skewness, kurtosis, etc., to understand the distribution form of the load; determining the capacity-load ratio data of each voltage level of the to-be-planned area includes collecting the substation capacity, feeder capacity and actual load data of each voltage level, calculating the capacity-load ratio of each voltage level, i.e. the ratio of substation or feeder capacity to actual load, and then analyzing the time variation trend and spatial distribution characteristics of the capacity-load ratio;

[0101] S32: Based on the original grid structure, fully considering the network and feeder transmission constraints, according to the load distribution characteristics and load development demand, a multi-voltage level distribution network planning model is constructed; based on the original grid structure, fully considering the network and feeder transmission constraints, i.e. based on the original grid structure, analyzing the topology structure, voltage level, substation location, feeder layout, etc. of the existing power grid, considering the safety and stability operation constraints of the power grid, such as power flow constraint, voltage constraint, short-circuit current constraint, etc.; when constructing the planning model, according to the load distribution characteristics and load development demand, predicting future load changes, considering the network and feeder transmission constraints, constructing a mathematical model of multi-voltage level distribution network planning, the constructed model contains objective functions, such as minimizing construction cost, maximizing power supply reliability, etc., and the model contains constraint conditions, such as load balance, voltage constraint, etc.;

[0102] S33: Calculate the first Euclidean distance between the fitted Copula function and the empirical Copula function; in the present application, according to the statistical characteristics of the load and the capacity ratio data, a suitable Copula function family is selected, such as Archimedean Copula, extreme value Copula, etc., and a specific Copula function form is selected within the function family, such as normal Copula, t-Copula, Clayton Copula, etc.; the first Euclidean distance between the fitted Copula function and the empirical Copula function is calculated to quantify the fitting effect, and the smaller the first Euclidean distance, the better the fitting effect;

[0103] S34: Select the fitted Copula function with the smallest first Euclidean distance as the optimal Copula function, and calculate the new random variable sample layer by layer to establish the main distribution integrated coupling mechanism model. If there are multiple candidate functions and the first Euclidean distance is similar, use other evaluation indicators such as Kendall's tau, Spearman's rho, etc. for further comparison; use the optimal Copula function to generate new load and capacity ratio random variable samples, and consider the correlation between the load and the capacity ratio, calculate the load and the capacity ratio of each voltage level layer by layer, input the generated load and capacity ratio sample into the distribution network planning model, solve the planning model to obtain the optimal power grid planning scheme, analyze the power grid structure, voltage level, load distribution, etc. under the optimal scheme, and establish the main distribution integrated coupling mechanism model.

[0104] Further, it also includes the total capacity of the substation, which is calculated as follows:

[0105] n G S g ≥R G,min ∑P,

[0106] In the formula, n G is the number of substations in the region, S g is the capacity of a single substation, R G,min is the minimum operating rate of the substation, and ∑P is the total power demand of all load points;

[0107] The distribution network planning model is as follows:

[0108]

[0109] In the formula, Z is the distribution network planning cost, L ij is the network construction cost coefficient from node i to node j, x ij is a binary decision variable representing the connection relationship between nodes i and j;

[0110] The first Euclidean distance between the fitted Copula function and the empirical Copula function:

[0111]

[0112] where d is the first Euclidean distance of random variables X1 and X2, C(X1, X2) is the fitted Copula function, C n (X1, X2) is the empirical Copula function. The Copula function is a function used to describe the dependence structure between multiple random variables, and the empirical Copula function is a Copula function estimated based on sample data, C n (X1, X2) is an estimate of the joint distribution function of X1 and X2 based on observed data.

[0113] In some embodiments of the application, solving the sampling probability distribution value comprises the following steps:

[0114] Solving the sampling probability distribution value based on the matrix of random variables {z1, z2, z3}:

[0115]

[0116] where z i is a random variable, u i is the sampling probability distribution value corresponding to the random variable;

[0117] Substitute the known quantities z2 and z3 into the above equation and solve the differential equation to obtain the value of the sampling point;

[0118] Main matching scenario set sampling:

[0119] Convert the random number to the actual sample value by inverting the cumulative distribution function:

[0120] x = F -1 (U),

[0121] where U = F(x) is the cumulative distribution function of the variable x, U ∈ [0, 1], F -1 is the inverse function of the cumulative distribution function;

[0122] Convert the probability distribution value obtained based on the coupling mechanism model sampling into the corresponding wind and light load power to obtain the sampling scenario set based on the C-Copula coupling mechanism model.

[0123] In addition, the historical output data of the main matching collaborative operation model is a data sample matrix X M×N :

[0124]

[0125] In the formula, V M,N represents the Mth main distribution network Nth historical output data;

[0126] The second Euclidean distance of each data to the initial cluster center point is calculated, and the data is allocated to the cluster with the minimum second Euclidean distance according to the principle of minimum second Euclidean distance, and the number of samples contained in the new cluster C is calculated i ;

[0127] According to the cluster center updating formula, the cluster center point o i of each new cluster is recalculated, and the cluster center updating expression is as follows:

[0128]

[0129] In the formula, x is the sample data in the new cluster C i In the K-means algorithm, the updating of the cluster center is a key step, and once the data points are allocated to the clusters, the cluster center of each new cluster needs to be calculated for use in subsequent iterations.

[0130] Further, whether the data process continues or not is judged according to the change of the cluster center, and when the change is small and meets the set condition or the number of iterations reaches the upper limit, the clustering operation can be terminated. When the K-means algorithm is used, in the iteration process, the distance or difference between the new and old cluster centers can be calculated after the cluster center is updated each time, and if the distance or difference is smaller than a preset threshold, that is, the change is small, it is considered that the cluster center has been stabilized, and the iteration can be terminated. At the same time, in order to prevent the algorithm from iterating infinitely, a maximum number of iterations is usually set. If the algorithm does not meet other termination conditions such as small change of the cluster center when the maximum number of iterations is reached, the iteration is also terminated.

[0131] The verification test of the coupling mechanism analysis method of the main-distribution coordination integration planning provided in the embodiment of the application is provided, the technical effects used in the method are verified and explained, and the test results are compared by scientific means to verify the real effects of the method.

[0132] In the embodiment of the application, the reduced scenarios are three scenarios as shown in Figures 3 to 5 The probabilities are 0.24385, 0.12603 and 0.12329 respectively. The pre-optimization results under each scenario are compared with the post-optimization results proposed in the application, and the comparison and analysis results of the coupling after optimization are as follows: as shown in Figure 3 Scenarios 1 and 2 have the minimum coupling value, and both are as shown in Figure 4 Scenario 2, and both are as shown in Figure 5In the third scenario, the coupling characteristics between random variables are different, some of which are positive coupling characteristics and some of which are negative coupling characteristics. At the same time, the fluctuation range of coupling characteristics in scenario 1 is [-0.6503, 0.2917], which is the largest; the fluctuation range of coupling characteristics in scenario 2 is [0.4082, 0.8065], which is the smallest; and the fluctuation range of coupling characteristics in scenario 3 is [-0.4965, 0.3351], which is in the middle. It is illustrated that the uncertainty scenario analysis method considering coupling characteristics proposed in the present application describes the coupling characteristics in several cases, so that the obtained scenario is more representative.

[0133] Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A coupling mechanism analysis method for master-distributor coordinated integrated planning, characterized in that, Includes the following steps: Obtain the edge distribution function and probability density function of wind power plants, photovoltaic power plants, and load power; Calculate the Kendall-τ coupling coefficients between pairwise variables for each sample and select root nodes; A Copula-Copula model is established by calculating the first Euclidean distance between the fitted Copula function and the empirical Copula function. Solve for the sampling probability distribution value; By inverting the cumulative distribution function, random numbers can be converted into actual sample values; The probability distribution values ​​obtained by sampling based on the coupling mechanism model are transformed into the corresponding wind and solar load power, thus obtaining the sampling scene set based on the C-vine-Copula coupling mechanism model; Using the historical output data of the master-supplier collaborative operation model, the second Euclidean distance from each data point to the initial cluster center is calculated. The data is then assigned to the cluster with the smallest second Euclidean distance according to the principle of minimizing the second Euclidean distance. Determine whether to continue the data processing based on changes in cluster centers; Using the minimum difference between the output curves and load curves of wind power plants and photovoltaic power plants as the objective function, the K-means clustering algorithm and silhouette coefficient are used as indicators to judge the clustering effect, describing the similarity and density within clusters and the dispersion between clusters. This also includes establishing the objective function that minimizes the difference between the main distribution network output curve and the load curve, which is: In the formula, α k and α L k These represent the normalized rates of change in power generation and load of photovoltaic power plants, wind farms, and substations, respectively, δ. i I represents the change in the system's total historical power generation. T Total power generation of main and distribution networks, I C For load power; The silhouette coefficient is used as an indicator to judge the clustering effect. The silhouette coefficient combines the clustering degree and the separation degree, and can well describe the similarity and compactness within clusters and the dispersion between clusters. The specific calculation formula is as follows: In the formula, SC is the overall silhouette coefficient of the clustered data, and a i Let b be the average distance between the i-th data point and other elements in the same cluster. i Let M be the average distance between the i-th data point and all data points in the nearest cluster to the next centroid, and M be the number of samples in the dataset.

2. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 1, characterized in that, The kernel density estimation method is used to obtain the marginal distribution functions of wind power plants, photovoltaic power plants, and load power. In the formula, v n It is the rated wind speed of the fan, v in It is the cut-in wind speed of the fan, P w It is the output power of the fan, P nw λ is the rated power of the wind turbine, λ and c are the shape parameters of the Weibull distribution, and s is the light intensity. max Γ(σ) is the maximum light intensity, σ and β are the shape parameters of the Beta distribution, and Γ(σ) is the gamma function.

3. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 2, characterized in that, Probability density of a random variable sample: In the formula, P w It is the output power of the fan, P s It is the output power of photovoltaics, P L It is the load power.

4. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 3, characterized in that, When selecting the root nodes, calculate the Kendall-τ coupling coefficients between pairwise variables for each sample: In the formula, n is the total number of data points, C is the number of pairs of points with the same order, and D is the number of pairs of points with different orders.

5. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 1, characterized in that, Constructing a C-vine-Copula model involves the following steps: Determine the load distribution information of the area to be planned and the capacity-to-load ratio data of each voltage level in the area to be planned; Based on the existing network structure, and taking into full account the network and feeder transmission constraints, a multi-voltage-level distribution network planning model is constructed according to the load distribution characteristics and load development needs. Calculate the first Euclidean distance between the fitted Copula function and the empirical Copula function; The Copula function with the smallest first Euclidean distance is selected as the optimal Copula function, and new random variable samples are obtained by calculating and solving layer by layer to establish a principal-partition integrated coupling mechanism model.

6. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 5, characterized in that, This also includes the total capacity of the substations, which is calculated as follows: n G S g ≥R G,min ∑P, In the formula, n G S is the number of substations in the area. g It is the capacity of a single substation, R G,min ∑P is the minimum operating rate of the substation, and ∑P is the total power demand of all load points. The power distribution network planning model is as follows: In the formula, Z is the distribution network planning cost, and L... ij x is the network construction cost coefficient from node i to node j. ij It is a binary decision variable, representing the connection relationship between node i and node j; The first Euclidean distance between the fitted Copula function and the empirical Copula function: In the formula, d is the first Euclidean distance between random variables X1 and X2, C(X1,X2) is the fitted Copula function, and C n (X1,X2) is an empirical Copula function.

7. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 1, characterized in that, Solving for the sampled probability distribution value includes the following steps: Solving for the sampling probability distribution based on the random variable matrix {z1,z2,z3}: In the formula, z i It is a random variable, u i It is the sampled probability distribution value of the corresponding random variable; Substituting the known quantities z2 and z3 into the above equation and solving the differential equation will yield the values ​​of the sampling points. Sampling of main and supporting scene sets: Convert random numbers into actual sample values ​​by inverting the cumulative distribution function: x=F -1 (U), In the formula, U = F(x) is the cumulative distribution function of the variable x, U ∈ [0,1], F -1 It is the inverse function of the cumulative distribution function; The probability distribution values ​​obtained by sampling based on the coupling mechanism model are transformed into the corresponding wind and solar load power, resulting in a sampling scene set based on the C-vine-Copula coupling mechanism model.

8. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 1, characterized in that, The historical output data of the main and auxiliary power supply coordinated operation model is a data sample matrix X with a sampling interval of t, a sampling number of N, and a sampling object number of M. M×N : In the formula, V M,N This represents the Nth historical power output data of the Mth main distribution network; Calculate the second Euclidean distance from each data point to the initial cluster center. Then, assign the data to the cluster with the smallest second Euclidean distance according to the principle of minimizing this distance. Finally, calculate the number of samples C in each new cluster. i ; Based on the cluster center update formula, the cluster center point o in each new cluster is recalculated. i The cluster center update expression is: In the formula, x is the new cluster C i The sample data in the sample.

9. The coupling mechanism analysis method for master-distributor coordinated integrated planning according to claim 8, characterized in that, The decision to continue the data processing is based on changes in the cluster centers. The clustering operation can be terminated when the changes in the clusters are small enough to meet the set conditions or when the number of iterations reaches the upper limit.

10. An apparatus having a computer program stored thereon, characterized in that, The computer program can implement the method as described in any one of claims 1-9.

Images