New energy unit clustering method based on modified node admittance matrix

Abstract

The application provides a new energy unit clustering method based on a modified node admittance matrix, comprising the following steps: (1) reading power grid topology data and element parameters to obtain a branch table of the power grid; (2) constructing a modified node admittance matrix according to the branch table of the power grid; (3) calculating eigenvalues and eigenvectors of the modified node admittance matrix; (4) sorting n eigenvalues in ascending order to obtain new eigenvalues and corresponding eigenvector sequences; and (5) based on the new eigenvalues and corresponding eigenvector sequences obtained in step (4), using a mean value division method based on Laplacian spectrum to divide nodes in the power grid. The network topology is described by the constructed modified node admittance matrix, and the nodes are divided by the mean value division method based on the Laplacian spectrum, so that the new energy units can be accurately clustered. The application is suitable for grouping of various new energy stations.

Description

TECHNICAL FIELD

[0001] The application relates to a new energy station unit clustering method, in particular to a new energy unit clustering method based on a modified node admittance matrix. BACKGROUND

[0002] When a new energy station is modeled, the units in the station are often equalized to reduce the calculation amount of simulation. Before equalization, the units need to be divided into several relatively independent groups according to the electrical connection relationship.

[0003] The existing method mainly uses the k-means grouping method, which is difficult to accurately identify the electrical connection strength between the units, resulting in inaccurate grouping results. SUMMARY

[0004] In view of the above problems existing in the prior art, the application provides a new energy unit clustering method based on a modified node admittance matrix.

[0005] The technical scheme adopted by the application is as follows:

[0006] A new energy unit clustering method based on a modified node admittance matrix comprises the following steps:

[0007] (1) Read the power grid topology data and element parameters to obtain the branch table of the power grid:

[0008] T l ={N i ,N j ,X ij},i=1,2,…,n,j=1,2,…,n;

[0009] Wherein T l is the branch table of the power grid, N i and N j are the nodes on both sides of the branch, X ij is the branch reactance between the nodes N i and N j , and n is the number of network nodes.

[0010] (2) According to the branch table of the power grid, a modified node admittance matrix B is constructed as follows:

[0011]

[0012] (3) Calculate the eigenvalue D and eigenvector V of the modified node admittance matrix B:

[0013] D=[λ1,λ2,…λ n ],V=[v1,v2,…v n ]

[0014] where λ1, λ2, … λn are the eigenvalues of the node admittance matrix. n to modify the eigenvalues of the node admittance matrix, v1, v2, … vn are the eigenvectors of the node admittance matrix. n to modify the eigenvectors of the node admittance matrix.

[0015] (4) Sort the n eigenvalues in ascending order to obtain a new sequence of eigenvalues and corresponding eigenvectors:

[0016] D S = [λ S1 , λ S2 , … λ Sn ], V S = [v S1 , v S2 , … v Sn ]

[0017] (5) Based on the new sequence of eigenvalues and corresponding eigenvectors obtained in step (4), use the mean value partition method based on Laplacian spectrum to partition the nodes in the power grid.

[0018] Further, step (5) specifically includes:

[0019] Step 5.1: Arrange the second to m+1 sorted eigenvectors into an n x m matrix:

[0020] V MS = [v S2 v S2 … v S(m+1) ]

[0021] where m is the partition coefficient.

[0022] Step 5.2: According to the relationship between the value of V MS of each node and the average value of each column element, the nodes are partitioned, and the steps are as follows:

[0023] 1) Read the column data V MS of V MSj , calculate the average value of each column:

[0024]

[0025] where is the average value of the jth column vector.

[0026] 2) Define a new allocation matrix C, and calculate each element in the allocation matrix C as follows:

[0027]

[0028] 3) Replace V MSThe elements in each column are calculated as follows to obtain V MS partition vector Q D :

[0029]

[0030] where Q Di is the i-th element of Q D .

[0031] 4) The non-repeated elements in the partition vector Q D are sequentially re-encoded in ascending order:

[0032] Q Di =n, n is Q Di The order of the non-repeated elements in Q D ,

[0033] The node N i is divided into the block Q Di .

[0034] Advantages of the present application:

[0035] 1. The modified admittance matrix is used to describe the topological relationship between new energy units, which can better reflect the influence of the actual unit impedance on the grouping.

[0036] 2. The Laplace-based mean partition method has high discrimination and can accurately divide the units into different groups. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 is one of the embodiments of the new energy unit clustering method based on the modified node admittance matrix of the present application test data diagram;

[0038] Figure 2 is the grouping result calculated by the present application. DETAILED DESCRIPTION

[0039] The technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application.

[0040] Figure 1 The schematic diagram of the new energy unit clustering method based on the modified node admittance matrix of the present application is shown in the figure, and the method comprises the following steps:

[0041] (1) Read the power grid topology data and element parameters to obtain the branch table of the power grid:

[0042] T l ={N i ,N j ,X ij}, i = 1, 2, …, n, j = 1, 2, …, n;

[0043] where T l is the power grid branch table, N i , N j are the nodes on both sides of the branch, X ij is the resistance and reactance of the branch between nodes i and j, and n is the number of network nodes.

[0044] (2) According to the power grid branch table, construct the modified node admittance matrix B as follows:

[0045]

[0046] (3) Calculate the eigenvalues and eigenvectors of the modified node admittance matrix B:

[0047] D = [λ1, λ2, … λ n ], V = [v1, v2, … v n ]

[0048] (4) Sort the n eigenvalues in ascending order to obtain the new eigenvalues and corresponding eigenvector sequences:

[0049] D S = [λ S1 , λ S2 , … λ Sn ], V S = [v S1 , v S2 , … v Sn ]

[0050] (5) Based on the new eigenvalues and corresponding eigenvector sequences obtained in step (4), use the mean value division method based on Laplacian spectrum to divide the network.

[0051] Step (5) specifically includes:

[0052] Step 5.1: Arrange the second to m+1 sorted eigenvectors into an n x m matrix:

[0053] V MS = [v S2 v S2 …v S(m+1) ]

[0054] where m is the division coefficient.

[0055] Step 5.2: According to the relationship between the V MS value corresponding to each node and the average value of each column element, divide the nodes, the steps are as follows:

[0056] 1) Read VMS column data V MSj , the average value of each column is calculated:

[0057]

[0058] wherein is the average value of the jth column vector.

[0059] 2) define a new distribution matrix C, each element of C is calculated as follows:

[0060]

[0061] 3) the elements in each column of V MS are calculated as follows to obtain the partition vector Q MS of V D :

[0062]

[0063] wherein Q Di is the i-th element of Q D .

[0064] 4) the non-repeated elements in the vector Q D are sequentially re-encoded in ascending order:

[0065] Q Di = n, n is the order of the non-repeated elements in Q Di . D

[0066] 5) the nodes N i are partitioned into blocks Q Di .

[0067] The method will be further described in combination with the accompanying drawings and a specific example:

[0068] A new energy photovoltaic station contains 30 inverters, 30 box transformers, 1 main transformer, and 32 power collection lines. The electrical wiring diagram is shown in Figure 1 . The partition coefficient m is taken as 3. The parameters of each element of the power grid are shown in Table 1 and Table 2:

[0069] Table 1 Branch table

[0070] Line Name

[0002] N i ]] <![CDATA[N j ]]> X ij ]]> AC_20 #8 Box Transformer 35kV BUS 0.003985 AC_38 35kV BUS #20 Box Transformer 0.029717 AC_4 Main Transformer Grid 0.390000 AC_42 35kV BUS #21 Box Transformer 0.018289 AC_46 35kV BUS #30 Box Transformer 0.023331 AC_50 35kV BUS #33 Box Transformer 0.027278 AC_6 #7 Box Transformer #2 Box Transformer 0.013594 AC_9 #2 Box Transformer #3 Box Transformer 0.003449 Name_10 #2 Box Transformer #1 Box Transformer 0.004750 Name_13 #10 Box Transformer #11 Box Transformer 0.001000 Name_14 #11 Box Transformer #12 Box Transformer 0.003028 Name_15 #9 Box Transformer #10 Box Transformer 0.004343 Name_21 #9 Box Transformer 35kV BUS 0.002299 Name_22 #15 Box Transformer #14 Box Transformer 0.003633 Name_23 #14 Box Transformer #13 Box Transformer 0.003550 Name_24 #16 Box Transformer #15 Box Transformer 0.004687 Name_25 #21 Box Transformer #22 Box Transformer 0.003221 Name_26 #23 Box Transformer #24 Box Transformer 0.003080 Name_27 #22 Box Transformer #23 Box Transformer 0.004385 Name_28 #16 Box Transformer 35kV BUS 0.032663 Name_29 #20 Box Transformer #25 Box Transformer 0.019992 Name_30 #20 Box Transformer #19 Box Transformer 0.004509 Name_31 #25 Box Transformer #26 Box Transformer 0.003080 Name_32 #16 Box Transformer 35kV BUS 0.029717 Name_33 #33 Box Transformer #32 Box Transformer 0.007830 Name_34 #18 Box Transformer #19 Box Transformer 0.002577 Name_36 #32 Box Transformer #31 Box Transformer 0.002777 Name_40 #30 Box Transformer #29 Box Transformer 0.003195 Name_41 #28 Box Transformer #27 Box Transformer 0.001670 Name_42 #30 Box Transformer #28 Box Transformer 0.006072 Name_9 #8 Box Transformer #6 Box Transformer 0.002021 Collector Line 35kV BUS #7 Box Transformer 0.005923

[0071] Table 2 Transformer table

[0072]

[0073] According to the above data, the following modified node admittance matrix B can be calculated: ​

[0074] Table 3 modified nodal admittance matrix B

[0075]

[0076]

[0077] The eigenvalues and eigenvectors of B are calculated and the eigenvectors are sorted by the size of the eigenvalues to get the new eigenvalues and the corresponding eigenvector sequence (see Table 4):

[0078] Table 4

[0079] 1 2 3 4 5 6 7 8 9 10 ... 63 1 0.13 -0.07 -0.05 -0.09 -0.06 -0.52 0.23 0.01 0.07 0.12 ... 0.00 2 0.13 -0.04 -0.02 -0.03 -0.02 -0.04 -0.38 -0.08 0.08 0.09 ... 0.00 3 0.13 -0.01 0.00 0.00 0.00 0.00 -0.01 0.00 0.00 0.00 ... 0.13 4 0.13 -0.04 -0.02 -0.03 -0.02 -0.05 -0.43 -0.10 0.14 0.16 ... 0.00 5 0.13 -0.01 0.00 0.00 0.00 0.00 -0.02 0.00 0.00 0.00 ... -0.06 6 0.13 -0.04 -0.02 -0.04 -0.02 -0.05 -0.52 -0.13 0.24 0.32 ... 0.00 7 0.13 -0.01 0.00 0.00 0.00 0.00 -0.02 0.00 0.00 0.00 ... 0.01 8 0.13 -0.08 -0.07 -0.18 -0.43 0.26 0.17 0.01 0.09 0.17 ... 0.00 9 0.13 -0.01 -0.01 -0.02 -0.03 0.02 0.01 0.00 0.00 0.00 ... 0.00 10 0.13 -0.08 -0.07 -0.17 -0.39 0.23 0.14 0.01 0.05 0.07 ... 0.00 ... ... ... ... ... ... ... ... ... ... ... ... 63 0.13 -0.01 0.00 0.00 0.00 0.00 -0.01 0.00 0.00 0.00 ... 0.00

[0080] Since the division coefficient m is 3, the 2nd to 4th eigenvectors are selected to form a new matrix (see Table 5):

[0081] Table 5

[0082]

[0083]

[0084]

[0085] According to the second step in step 5.2, the allocation matrix C is calculated as shown in Table 6:

[0086] Table 6

[0087]

[0088]

[0089] According to the third step in step 5.2, the division vector Q is calculated from C D , see Table 7:

[0090] Table 7

[0091]

[0092]

[0093] The elements in Q D are re-encoded according to the order from small to large, with non-repeated elements being 0, 1, 3, 7, 0 re-encoded as 1, 1 re-encoded as 2, 3 re-encoded as 3, and 7 re-encoded as 4, to get the final Q D as shown in Table 8:

[0094] Table 8

[0095]

[0096]

[0097]

[0098] According to the fifth step of 5.2, the node N i is divided into the block Q Di , see Table 9:

[0099] Table 9

[0100]

[0101]

[0102] According to the above grouping result, the grouping can be obtained as shown in Figure 2 . In the figure, the nodes with the same color are in one partition. It can be seen from the figure that the nodes connected to the same branch are allocated into the same partition, verifying the correctness of the method of the application.

[0103] The above is only a specific embodiment of the application, but the protection scope of the application is not limited to the step response test of the generator excitation system, and any step response test of an automatic control system is included in the technical field of the application. Any changes or replacements that can be easily thought of by those skilled in the art within the technical scope disclosed by the application should be covered in the protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1.A new energy unit clustering method based on modifying the admittance matrix of nodes, characterized in that, Comprising the following steps: (1) reading power grid topology data and element parameters to obtain a branch table of the power grid: ； wherein is the power grid branch table, , is the node on both sides of the branch, is the node and is the branch reactance between the nodes n is the number of network nodes (2) constructing a modified node susceptance matrix B as follows according to the branch table of the power grid: ； (3) calculating eigenvalues D and eigenvectors V of the modified node susceptance matrix B: ， ； wherein to modify eigenvalues of the nodal admittance matrix, to modify eigenvectors of the nodal admittance matrix; (4) sorting n eigenvalues in ascending order to obtain new eigenvalues and corresponding eigenvector sequences: ， ； (5) based on the new eigenvalues and corresponding eigenvector sequences obtained in step (4), dividing the nodes in the power grid by using a mean value division method based on Laplacian spectrum; Step (5) specifically comprises: Step 5.1: Arrange the 2nd to m+lst ordered eigenvectors into a matrix of size m x (m-l). ； Where m is a division coefficient; Step 5.2: Based on each node's corresponding... To divide the nodes based on the relationship between the value and the average of the elements in each column, the steps are as follows: 1) Read the column data , compute the average value for each column: ； wherein is the average value of the jth column vector; 2) define a new allocation matrix , the allocation matrix Each element in the matrix is calculated as follows: ； 3) the following is calculated The elements in each column are calculated as follows, resulting in a partition vector :​ ； wherein is the i-th element of 4) The division vector is divided into non-repeating elements, and the non-repeating elements are sequentially re-encoded in ascending order: The non-repeating elements in the division vector are sequentially re-encoded in ascending order: n is In the order of elements is not repeated, the nodes are partitioned into chunks .

Images