A quantitative evaluation method for coupling degree between distribution network and communication network

Abstract

The present invention relates to a quantitative evaluation method for coupling degree between a distribution network and a communication network. Firstly, a correlation matrix K is established according to the node dependency relationship between the distribution network and the communication network, and the node degree, node betweenness and node clustering coefficient of the distribution network topology and the communication network topology are solved respectively; secondly, the Pearson correlation coefficient is corrected by using the matrix K, and then the node degree similarity, node betweenness similarity and node clustering coefficient similarity of the coupling network between the distribution network and the communication network are solved, and the comprehensive weights of the three similarity parameters are solved by using the hierarchical analysis method and the entropy weight method; finally, the correlation of the coupling network is quantified by using the comprehensive weights of the three similarity parameters, and then the coupling correlation degree between the distribution network and the communication network is evaluated.

Description

Technical Field

[0001] The invention relates to a method for quantitatively evaluating the coupling degree between a distribution network and a communication network, and belongs to the field of economic evaluation of power system distribution network construction. Background Art

[0002] As the power network with the closest relationship with users, the reliable operation of the distribution network is the basis for ensuring the normal operation of users' daily lives. With the increasing attention paid by the country to the distribution network and the rapid development of smart grids, the distribution network is no longer just a single network. The corresponding communication network is also increasing year by year in terms of scale and complexity. The increasingly complex distribution network has brought huge data transmission business volume to the communication network. Various power business data are transmitted through the communication network so that the staff can use it to evaluate the status of the power grid and control the power grid.

[0003] When the communication network transmits power communication services, it establishes a coupling relationship between the distribution network and the communication network. This coupling relationship brings convenience to the power grid, but also brings corresponding hidden dangers: the coupling relationship between the distribution network and the communication network increases the vulnerability of the two networks. At the same time, the shortcomings of the coupling relationship are particularly obvious after a certain network node fails. Take the failure of a power grid node as an example. When a certain node of the power grid fails, the area within its power supply radius will be cut off, and the communication node coupled with it will lose power supply, thereby disconnecting the communication service. At this time, if the communication node is responsible for the transmission of relatively important power communication services, it may cause some power grid control signals to be unable to be transmitted, and then cause other nodes that originally did not fail to become faulty nodes due to cascading effects, which eventually causes the scope of the fault to spread and increases the scope of the power outage.

[0004] Therefore, in view of the coupling relationship between the distribution network and the communication network, a method is needed to evaluate and quantify the degree of coupling correlation between the two networks, and based on the quantified results, to evaluate the probability of cascading failure after a certain network node fails. Summary of the invention

[0005] In order to evaluate and quantify the coupling correlation degree between the distribution network and the communication network, the present invention provides an evaluation method for quantifying the coupling degree between the distribution network and the communication network.

[0006] The present invention is implemented by the following technical solutions: A method for quantitatively evaluating the coupling degree between a distribution network and a communication network, comprising the following steps:

[0007] Step 1: According to the dependency relationship between the distribution network and the communication network nodes, establish the association matrix K; the elements in the association matrix K are k ij , k ij Represents the distribution network node V iWith the communication network node V j The connection relationship between them, where the element k in the association matrix K ij The values ​​are:

[0008]

[0009] Step 2, solving the node degree, node betweenness and node clustering coefficient of the distribution network topology and the communication network topology;

[0010] Step 3: Combined with the characteristics of the distribution network-communication network coupling, the Pearson correlation coefficient is corrected, and the corrected Pearson correlation coefficient is used to solve the node degree similarity ρ under the distribution network-communication network coupling topology D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C ;

[0011] Step 4: Use the analytic hierarchy process to solve the degree similarity of the nodes in the coupling network between the distribution network and the communication network ρ D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The subjective weight W c ;

[0012] Step 5: Use the entropy weight method to solve the node degree similarity ρ between the distribution network and the communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The objective weight W s ;

[0013] Step 6: The subjective weight W obtained by AHP c , the objective weight W obtained by entropy weight method s , weight the two to get the comprehensive weight W, the calculation formula is as follows:

[0014] W=αW s +βW c

[0015] Where α is the objective weight coefficient, which is used to characterize the importance of the objective weight in the comprehensive weight; β is the subjective weight coefficient, which is used to characterize the importance of the subjective weight in the comprehensive weight. The value range of α and β is between 0 and 1, and satisfies the relationship of α+β=1.

[0016] Step 7: According to the method of solving the comprehensive weight of the index in step 6, the comprehensive weights of the three similarity indexes in step 3 are weighted and summed to obtain the quantitative result ρ of the coupling relationship between the distribution network and the communication network coupling network sum :

[0017]

[0018] In the formula, is the comprehensive weight of node degree similarity, ρ D is the node degree similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node betweenness similarity, ρ B is the node betweenness similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node clustering coefficient similarity, ρ C is the similarity of node clustering coefficient of the coupled network between distribution network and communication network.

[0019] Further preferably, in step 3, in view of the different node dimensions of the distribution network and the communication network, when calculating the correlation between the distribution network and the communication network, the Pearson correlation coefficient is corrected as follows:

[0020]

[0021] In the formula, k ij is the element of the incidence matrix K, X i is the distribution network node V i Topological parameters, i = 1 ~ n, n is the number of distribution network nodes, is the average value of topological parameters of distribution network nodes, Y j is the communication network node V j Topological parameters, i = 1 ~ m, m is the number of distribution network nodes, is the average value of topological parameters of communication network nodes;

[0022] The network topology parameters of the distribution network and the communication network solved in step 2: node degree, node betweenness and node clustering coefficient are substituted into the modified Pearson correlation coefficient to obtain the node degree similarity ρ of the coupled network of the distribution network and the communication network. D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C .

[0023] Further preferably, in step 4, first, construct a node degree similarity ρ for solving the coupling network of the distribution network and the communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The judgment matrix Xc of subjective weights, the elements of the judgment matrix Xc are α ij ,and All eigenvectors of the judgment matrix Xc are geometrically averaged and then normalized as follows:

[0024]

[0025] Where W e is the result of normalizing the geometric mean of the e-th vector in the judgment matrix Xc, and W e The value of is the subjective weight W of the e-th evaluation index c , is the geometric mean of the e-th vector in the judgment matrix Xc, is the geometric mean of the t-th vector in the judgment matrix Xc, and T is the number of vectors in the judgment matrix Xc.

[0026] Solve the maximum eigenvalue λ based on the judgment matrix max In order to measure the error in the subjective weight calculation of the solution, the consistency index CI is introduced s , which is calculated as follows:

[0027]

[0028] where λ s is the sth eigenvalue of the judgment matrix, CI s is the consistency index corresponding to the sth eigenvalue, N is the number of evaluation indicators; s Size, introduce random consistency index RI:

[0029]

[0030] In order to check whether the judgment matrix Xc is consistent, the test coefficient CR is introduced:

[0031]

[0032] Where CI max is the maximum eigenvalue λ max Corresponding consistency indicators;

[0033] When CR<0.1, the judgment matrix Xc satisfies the consistency test. If not, modify the judgment matrix Xc.

[0034] Further preferably, in step 5, an initial matrix is ​​established according to the parameter data of the objective weight to be solved and standardized to obtain a standardized matrix X, and the element in the fth row and gth column of the standardized matrix X is x fg , calculate the corresponding probability matrix H based on the standardized matrix X. The calculation formula of the probability matrix H is as follows. Let h fg is the element in the fth row and gth column of the probability matrix H:

[0035]

[0036] Where k is the number of columns in the normalized matrix X.

[0037] Calculate the information entropy E of each row indicator from the probability matrixv :

[0038]

[0039] Where h vg is the element in the vth row and gth column of the probability matrix H, E v is the information entropy of the index in the vth row, and q is the number of index columns;

[0040] Define the information utility value p v :

[0041] p v =1-E v

[0042] Normalizing the information utility value can obtain the objective weight W of the vth indicator: s (v):

[0043]

[0044] Further preferably, in step 2, the node degree deg(V i ): represents a distribution network node V in the distribution network topology i The sum of the number of links directly connected to other nodes in the same topology. When the distribution network topology is a directed topology, the flow direction of the business in the distribution network needs to be considered. In this case, the node out-degree deg + (V i ) and node in-degree deg - (V i ), node degree deg(V i ) and the node out-degree deg + (V i ) and node in-degree deg - (V i ) is as follows:

[0045]

[0046] When the distribution network topology is an undirected network, the network node degree of the distribution network is equal to the number of links directly connected to the node.

[0047] Further preferably, in step 2, the node degree deg(V j ): represents a communication network node V in the communication network topology j The sum of the number of links directly connected to other nodes in the same topology. When the communication network topology is a directed topology, the flow of services in the communication network needs to be considered. In this case, the node out-degree deg + (V j ) and node in-degree deg - (V j), node degree deg(V j ) and the node out-degree and deg + (V j )Node in-degree deg - (V j ) is as follows:

[0048]

[0049] When the communication network topology is an undirected network, the degree of a network node in the communication network is equal to the number of links directly connected to the node.

[0050] Further preferably, in step 2, the node betweenness B(V a ): represents a node V in the topology of the distribution network or communication network a The importance of the topology and the flow through different node pairs in the same topology (V b , V c ) is handled by node V a The utilization efficiency is calculated as follows:

[0051]

[0052] In the formula, δ bc For node V b , V c The number of shortest paths between b,c (a) is δ bc Passing through node V a The number of paths.

[0053] Further preferably, in step 2, the node clustering coefficient C(V d ): A coefficient that reflects the degree of aggregation between nodes in the distribution network or communication network topology, which can characterize the node V d The degree of connection with its adjacent nodes is calculated as follows:

[0054] C d =[2deg(V d )] / [n(n-1)]

[0055] Where n represents the total number of nodes in the distribution network or communication network.

[0056] Working principle of the invention

[0057] 1Correction of Pearson correlation coefficient:

[0058] The Pearson correlation coefficient can be used to solve the correlation between data, which is applicable to the case where the dimensions of the data are the same; for the coupling relationship between the distribution network and the communication network, in practice, the relationship between the nodes of the two networks is not just a simple one-to-one relationship, but often a one-to-many relationship, that is, a power grid node is coupled with multiple communication network nodes, or a communication network is connected to multiple power grid nodes. This one-to-many relationship causes the dimensions of the data to be inconsistent when determining the similarity of the topological parameters of the two networks, and it is more troublesome to solve it using the Pearson correlation coefficient. Therefore, considering the coupling relationship in the unique scenario of the distribution network-communication network, the Pearson correlation coefficient calculation formula is corrected. The correction method uses the association matrix established in step 1 to reflect this one-to-many node relationship. The corrected Pearson calculation formula is as follows:

[0059]

[0060] 2. Analytical Hierarchy Process:

[0061] The analytic hierarchy process (AHP) is a common decision-making method used by decision makers to transform qualitative analysis of their thinking into quantitative analysis. When conducting AHP research, the evaluation object is first stratified. The stratification method generally divides the indicators into three layers according to the affiliation between the indicators, namely the target layer, the primary indicator and the secondary indicator. The subjective weight of each indicator is calculated layer by layer. The steps for calculating the subjective weight of indicators by AHP are as follows:

[0062] (1) Establish an indicator hierarchy model: According to the object to be evaluated, the decision-making goal, the factors to be considered, and the decision-making object are divided into the goal layer, indicator layer, and factor layer according to their mutual relationship;

[0063] (2) Investigate expert opinions and establish a judgment matrix

[0064] The establishment of the judgment matrix is ​​mainly used to compare the indicators in complex problems with each other and to rate them according to their importance. ij It represents the result of comparing the importance of factor i and factor j. In the judgment matrix, the importance comparison result has the following properties:

[0065]

[0066] Then establish the judgment matrix Xc.

[0067] (3) Solving the eigenvector

[0068] According to the obtained judgment matrix Xc, the eigenvalues ​​and corresponding eigenvectors of the judgment matrix are calculated, and all eigenvectors are geometrically averaged and then normalized to obtain the objective weights of each indicator.

[0069] (4) Perform consistency check

[0070] CI s , which is calculated as follows:

[0071]

[0072] where λ s is the sth eigenvalue of the judgment matrix, CI s is the consistency index corresponding to the sth eigenvalue, N is the number of evaluation indicators; s Size, introduce random consistency index RI:

[0073]

[0074] The RI values ​​are as follows:

[0075]

[0076] The consistency ratio is recorded as CR:

[0077]

[0078] When CR<0.1, we can consider that the established judgment matrix has satisfactory consistency. If the calculated CR>0.1, it is necessary to change the elements in the judgment matrix to achieve satisfactory consistency.

[0079] 3 Entropy weight method: In information theory, entropy is a measure of uncertainty. When the amount of known information is larger, the corresponding entropy is smaller, and when the amount of information is smaller, the corresponding entropy is larger. Therefore, the entropy value can be used to judge the degree of dispersion of the indicator. The greater the dispersion of the indicator, the greater the impact of the indicator on the comprehensive evaluation.

[0080] The present invention first establishes an association matrix K according to the node dependency relationship between the distribution network and the communication network, and respectively solves the node degree, node betweenness and node clustering coefficient of the distribution network topology and the communication network topology; secondly, the matrix K is used to correct the Pearson correlation coefficient, and then the node degree similarity, node betweenness similarity and node clustering coefficient similarity of the coupling network of the distribution network and the communication network are solved, and the hierarchical analysis method and the entropy weight method are used to solve the comprehensive weights of the three similarity parameters; finally, the comprehensive weights of the three similarity parameters are used to quantify the correlation of the coupling network, and then the coupling correlation degree of the distribution network and the communication network is evaluated. The present invention can be used to evaluate and quantify the degree of correlation of the coupling between the two networks, and according to the quantified results, to evaluate the probability of cascading failure after a certain network node fails. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1It is a flow chart for quantitative evaluation of coupling degree between distribution network and communication network as described in the specification of the present invention;

[0082] Figure 2 It is the IEEE-13 node distribution network-communication network coupling relationship diagram;

[0083] Figure 3 This is the coupling relationship diagram of the variable topology IEEE-13 node distribution network and communication network. DETAILED DESCRIPTION

[0084] The present invention is further explained in detail below with reference to the accompanying drawings and embodiments.

[0085] Reference Figure 1 , a method for quantitatively evaluating the coupling degree between a distribution network and a communication network comprises the following steps:

[0086] Step 1: According to the interdependence between the distribution network and the communication network nodes, the association matrix K is established as follows:

[0087]

[0088] The element in the incidence matrix K is k ij , k ij Represents the distribution network node V i With the communication network node V j The connection relationship between them, where the element k in the association matrix K ij The values ​​are:

[0089]

[0090] Step 2, solving the node degree, node betweenness and node clustering coefficient of the distribution network topology and the communication network topology;

[0091] Node degree deg(V i ): represents a distribution network node V in the distribution network topology i The sum of the number of links directly connected to other nodes in the same topology. When the distribution network topology is a directed topology, the flow direction of the business in the distribution network needs to be considered. In this case, the node out-degree deg + (V i ) and node in-degree deg - (V i ), node degree deg(V i ) and the node out-degree deg + (V i ) and node in-degree deg - (V i ) is as follows:

[0092]

[0093] When the distribution network topology is an undirected network, the network node degree of the distribution network is equal to the number of links directly connected to the node;

[0094] Similarly, the node degree deg(V j ): represents a communication network node V in the communication network topology j The sum of the number of links directly connected to other nodes in the same topology. When the communication network topology is a directed topology, the flow of services in the communication network needs to be considered. In this case, the node out-degree deg + (V j ) and node in-degree deg - (V j ), node degree deg(V j ) and the node out-degree and deg + (V j )Node in-degree deg - (V j ) is as follows:

[0095]

[0096] When the communication network topology is an undirected network, the degree of a network node in the communication network is equal to the number of links directly connected to the node.

[0097] Node betweenness B(V a ): represents a node V in the topology of the distribution network or communication network a The importance of the topology and the flow through different node pairs in the same topology (V b , V c ) is handled by node V a The utilization efficiency is calculated as follows:

[0098]

[0099] In the formula, δ bc For node V b , V c The number of shortest paths between b,c (a) is δ bc Passing through node V a The number of paths;

[0100] Node clustering coefficient C(V d ): A coefficient that reflects the degree of aggregation between nodes in the distribution network or communication network topology, which can characterize the node V d The degree of connection with its adjacent nodes is calculated as follows:

[0101] C d =[2deg(V d)] / [n(n-1)]

[0102] In the formula, n represents the total number of nodes in the distribution network or communication network;

[0103] Step 3: Combined with the characteristics of the distribution network-communication network coupling, the Pearson correlation coefficient is corrected, and the corrected Pearson correlation coefficient is used to solve the node degree similarity ρ under the distribution network-communication network coupling topology D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C ;

[0104] In view of the different node dimensions of the distribution network and the communication network, the Pearson correlation coefficient is corrected when calculating the correlation between the distribution network and the communication network, as shown in the following formula:

[0105]

[0106] In the formula, k ij is the element of the incidence matrix K, X i is the distribution network node V i Topological parameters, i = 1 ~ n, n is the number of distribution network nodes, is the average value of topological parameters of distribution network nodes, Y j is the communication network node V j Topological parameters, i = 1 ~ m, m is the number of distribution network nodes, is the average value of topological parameters of communication network nodes.

[0107] The network topology parameters of the distribution network and the communication network solved in step 2: node degree, node betweenness and node clustering coefficient are substituted into the modified Pearson correlation coefficient to obtain the node degree similarity ρ of the coupled network of the distribution network and the communication network. D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C .

[0108] Step 4: Use the analytic hierarchy process to solve the degree similarity of the nodes in the coupling network between the distribution network and the communication network ρ D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The subjective weight W c ; First, construct the node degree similarity ρ for solving the coupling network of distribution network and communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The judgment matrix Xc of subjective weights, the elements of the judgment matrix Xc are α ij ,and as follows:

[0109]

[0110] All eigenvectors of the judgment matrix Xc are geometrically averaged and then normalized as follows:

[0111]

[0112] Where W e is the result of normalizing the geometric mean of the e-th vector in the judgment matrix Xc, and W e The value of is the subjective weight W of the e-th evaluation index c , is the geometric mean of the e-th vector in the judgment matrix Xc, is the geometric mean of the t-th vector in the judgment matrix Xc, and T is the number of vectors in the judgment matrix Xc.

[0113] Solve the maximum eigenvalue λ based on the judgment matrix max In order to measure the error in the subjective weight calculation of the solution, the consistency index CI is introduced s , which is calculated as follows:

[0114]

[0115] where λ s is the sth eigenvalue of the judgment matrix, CI s is the consistency index corresponding to the sth eigenvalue, and N is the number of evaluation indicators. s Size, introduce random consistency index RI:

[0116]

[0117] N is the number of evaluation indicators;

[0118] In order to check whether the judgment matrix Xc is consistent, the test coefficient CR is introduced:

[0119]

[0120] Where CI max is the maximum eigenvalue λ max The corresponding consistency index.

[0121] When CR<0.1, the judgment matrix Xc satisfies the consistency test. If not, modify the judgment matrix Xc.

[0122] Step 5: Use the entropy weight method to solve the node degree similarity ρ between the distribution network and the communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ CThe objective weight W s ;

[0123] Entropy weight method: According to the parameter data of the objective weight to be solved, the initial matrix is ​​established and standardized to obtain the standardized matrix X. The element in the fth row and gth column of the standardized matrix X is x fg , calculate the corresponding probability matrix H based on the standardized matrix X. The calculation formula of the probability matrix H is as follows. Let h fg is the element in the fth row and gth column of the probability matrix H:

[0124]

[0125] Where k is the number of columns in the normalized matrix X.

[0126] Calculate the information entropy E of each row indicator from the probability matrix v :

[0127]

[0128] Where h vg is the element in the vth row and gth column of the probability matrix H, E v is the information entropy of the index in the vth row, and q is the number of index columns;

[0129] Define the information utility value p v :

[0130] p v =1-E v

[0131] Normalizing the information utility value can obtain the objective weight W of the vth indicator: s (v):

[0132]

[0133] Step 6: The subjective weight W obtained by AHP c , the objective weight W obtained by entropy weight method s , weight the two to get the comprehensive weight W, the calculation formula is as follows:

[0134] W=αW s +βW c

[0135] Where α is the objective weight coefficient, which is used to characterize the importance of the objective weight in the comprehensive weight; β is the subjective weight coefficient, which is used to characterize the importance of the subjective weight in the comprehensive weight. The value range of α and β is between 0 and 1, and satisfies the relationship of α+β=1.

[0136] Step 7: According to the method of solving the comprehensive weight of the index in step 6, the comprehensive weights of the three similarity indexes in step 3 are weighted and summed to obtain the quantitative result ρ of the coupling relationship between the distribution network and the communication network coupling network sum :

[0137]

[0138] In the formula, is the comprehensive weight of node degree similarity, ρ D is the node degree similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node betweenness similarity, ρ B is the node betweenness similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node clustering coefficient similarity, ρ C is the similarity of node clustering coefficient of the coupled network between distribution network and communication network.

[0139] According to the quantification result ρ sum Evaluate the coupling degree between the distribution network and the communication network. Given the quantitative range of coupling network correlation: when ρ sum is greater than 0.7, it is determined that the distribution network and the communication network topology are strongly correlated. sum If it is less than 0.7, it is determined that the topology of the distribution network and the communication network is weakly correlated.

[0140] Example

[0141] Taking the IEEE-13 node distribution network-communication network coupling model as an example, the normal topology and Figure 2 Coupled network attachment under lower and variable topology Figure 3 The degree of correlation is verified by experiments:

[0142] Calculation of coupling network parameters under normal topology:

[0143] according to Figure 2 The coupling network topology is established, and the coupling matrix K is established:

[0144]

[0145] The corresponding values ​​of each sub-matrix of the coupling matrix are as follows:

[0146]

[0147]

[0148]

[0149] The node degree, node betweenness and node clustering coefficient of the distribution network and the communication network are solved. The calculation results are shown in Table 1 and Table 2:

[0150] Table 1 IEEE-13 distribution network topology parameters

[0151]

[0152] Table 2 IEEE-13 communication network topology parameters

[0153]

[0154] According to the data obtained in Table 1 and Table 2, we put them into the modified Pearson correlation coefficient to obtain the node degree similarity, node betweenness similarity and node clustering coefficient similarity. The results are shown in Table 3:

[0155] Table 3 IEEE-13 distribution network-communication network coupling topology parameters

[0156]

[0157] The comprehensive weights of each similarity obtained by analytic hierarchy process and entropy weight method are:

[0158] Using the analytic hierarchy process to obtain the weight W c :

[0159] Table 4 Weight results obtained by AHP under normal topology

[0160]

[0161] The weight W obtained using the entropy weight method s :

[0162] Table 5 Weight results obtained by entropy weight method under normal topology

[0163]

[0164] Use objective weights and subjective weights to solve the comprehensive weight W:

[0165] Table 6 Comprehensive weight results under normal topology

[0166]

[0167] By weighted summing the results in Table 3, we can obtain the similarity of the IEEE-13 node distribution network-communication network coupling network:

[0168]

[0169] Since 0.841>0.7, it can be judged that there is a strong correlation between the distribution network and the communication network. When the distribution network fails, considering that the distribution network topology changes, the similarity of the distribution network-communication network coupling network after the IEEE-13 node topology changes is calculated. Assuming that the dependency between the distribution network and the communication network does not change after the topology changes, the correlation matrix of the two networks used for calculation is still K.

[0170] Under the variable topology: solve the network parameters of the distribution network and the communication network. The calculation results are shown in Table 7 and Table 8:

[0171] Table 7 IEEE-13 distribution network topology parameters after topology change

[0172]

[0173] Table 8 IEEE-13 communication network topology parameters after topology change

[0174]

[0175] According to the data obtained in Table 7 and Table 8, the data are substituted into the modified Pearson correlation coefficient to obtain three similarity coefficients. The results are shown in Table 9:

[0176] Table 9 IEEE-13 distribution network-communication network coupling topology parameters after topology change

[0177]

[0178] Similarly, using the values ​​of node degree similarity, node betweenness similarity, and node clustering coefficient similarity calculated by the distribution network-communication network after the changed topology, combined with the comprehensive weights of the three similarities, the similarity after the changed topology is obtained:

[0179]

[0180] Because 0.873>0.7, it can be judged that after the topology is changed, the distribution network and the communication network still have a strong correlation.

[0181] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for quantitatively evaluating the coupling degree between a distribution network and a communication network, characterized in that: The steps include: Step 1: According to the dependency relationship between the distribution network and the communication network nodes, establish the association matrix K; the elements in the association matrix K are k ij , k ij Represents the distribution network node V i With the communication network node V j The connection relationship between them, where the element k in the association matrix K ij The values ​​are: Step 2: Solve the node degree, node betweenness and node clustering coefficient of the distribution network topology and communication network topology; Step 3: Combined with the characteristics of the distribution network-communication network coupling, the Pearson correlation coefficient is corrected, and the corrected Pearson correlation coefficient is used to solve the node degree similarity ρ under the distribution network-communication network coupling topology D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C ; Step 4: Use the analytic hierarchy process to solve the degree similarity ρ of the nodes in the coupling network between the distribution network and the communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The subjective weight W c ; Step 5: Use the entropy weight method to solve the node degree similarity ρ between the distribution network and the communication network D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The objective weight W s ; Step 6: The subjective weight W obtained by AHP c , the objective weight W obtained by entropy weight method s , weight the two to get the comprehensive weight W, the calculation formula is as follows: W=αW s +βW c Where α is the objective weight coefficient, which is used to characterize the importance of the objective weight in the comprehensive weight; β is the subjective weight coefficient, which is used to characterize the importance of subjective weight in the comprehensive weight. The value range of α and β is between 0 and 1, and satisfies the relationship of α+β=1; Step 7: According to the method of solving the comprehensive weight of the index in step 6, the comprehensive weights of the three similarity indexes in step 3 are weighted and summed to obtain the quantitative result ρ of the coupling relationship between the distribution network and the communication network coupling network sum : In the formula, is the comprehensive weight of node degree similarity, ρ D is the node degree similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node betweenness similarity, ρ B is the node betweenness similarity of the coupling network between the distribution network and the communication network, is the comprehensive weight of node clustering coefficient similarity, ρ C is the similarity of node clustering coefficient of the coupled network between distribution network and communication network.

2. A method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1, characterized in that: In step 3, the Pearson correlation coefficient ρ is corrected when calculating the correlation between the distribution network and the communication network due to the different node dimensions of the distribution network and the communication network, as shown in the following formula: In the formula, k ij is the element of the incidence matrix K, X i is the distribution network node V i Topological parameters, i = 1 ~ n, n is the number of distribution network nodes, is the average value of topological parameters of distribution network nodes, Y j is the communication network node V j Topological parameters, i = 1 ~ m, m is the number of distribution network nodes, is the average value of topological parameters of communication network nodes; The network topology parameters of the distribution network and the communication network solved in step 2: node degree, node betweenness and node clustering coefficient are substituted into the modified Pearson correlation coefficient to obtain the node degree similarity ρ of the coupled network of the distribution network and the communication network. D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C .

3. The method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1 is characterized in that: In step 4, first, the node degree similarity ρ for solving the coupling network between the distribution network and the communication network is constructed. D , node betweenness similarity ρ B Similarity with node clustering coefficient ρ C The judgment matrix Xc of subjective weights, the elements of the judgment matrix Xc are α ij ,and All eigenvectors of the judgment matrix Xc are geometrically averaged and then normalized as follows: Among them, W e is the result of normalizing the geometric mean of the e-th vector in the judgment matrix Xc, and W e The value of is the subjective weight of the e-th evaluation index. is the geometric mean of the e-th vector in the judgment matrix Xc, is the geometric mean of the t-th vector in the judgment matrix Xc, T is the number of vectors in the judgment matrix Xc; Solve the maximum eigenvalue λ based on the judgment matrix max In order to measure the error in the subjective weight calculation of the solution, the consistency index CI is introduced s , which is calculated as follows: where λ s is the sth eigenvalue of the judgment matrix, CI s is the consistency index corresponding to the sth eigenvalue, N is the number of evaluation indicators; s Size, introduce random consistency index RI: In order to check whether the judgment matrix Xc is consistent, the test coefficient CR is introduced: Where CI max is the maximum eigenvalue λ max Corresponding consistency indicators; When CR<0.1, the judgment matrix Xc satisfies the consistency test. If not, modify the judgment matrix Xc.

4. The method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1 is characterized in that: In step 5, the initial matrix is ​​established according to the parameter data of the objective weight to be solved and standardized to obtain the standardized matrix X. The element in the fth row and gth column of the standardized matrix X is x fg , calculate the corresponding probability matrix H based on the standardized matrix X. The calculation formula of the probability matrix H is as follows. Let h fg is the element in the fth row and gth column of the probability matrix H: Where k is the number of columns of the normalized matrix X; Calculate the information entropy E of each row indicator from the probability matrix v : Where h vg is the element in the vth row and gth column of the probability matrix H, E v is the information entropy of the index in the vth row, and q is the number of index columns; Define the information utility value p v : p v =1-E v By normalizing the information utility value, we can get the objective weight W of each vth indicator: s (v):

5. The method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1 is characterized in that: In step 2, the node degree deg(V i ): represents a distribution network node V in the distribution network topology i The sum of the number of links directly connected to other nodes in the same topology. When the distribution network topology is a directed topology, the flow direction of the business in the distribution network needs to be considered. In this case, the node out-degree deg + (V i ) and node in-degree deg - (V i ), node degree deg(V i ) and the node out-degree deg + (V i ) and node in-degree deg - (V i ) is as follows: When the distribution network topology is an undirected network, the network node degree of the distribution network is equal to the number of links directly connected to the node.

6. A method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1, characterized in that: In step 2, the node degree deg(V j ): represents a communication network node V in the communication network topology j The sum of the number of links directly connected to other nodes in the same topology. When the communication network topology is a directed topology, the flow of services in the communication network needs to be considered. In this case, the node out-degree deg + (V j ) and node in-degree deg - (V j ), node degree deg(V j ) and the node out-degree and deg + (V j )Node in-degree deg - (V j ) is as follows: When the communication network topology is an undirected network, the degree of a network node in the communication network is equal to the number of links directly connected to the node.

7. A method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1, characterized in that: In step 2, the node betweenness B a :Represents a node V in the topology of the distribution network or communication network a The importance of V in its topology and the traffic flowing through the node V in the same topology b With V c When the business is node V a The utilization efficiency is calculated as follows: In the formula, δ bc For node V b With V c The number of shortest paths between bc (a) is δ bc Passing through node V a The number of paths.

8. The method for quantitatively evaluating the coupling degree between a distribution network and a communication network according to claim 1, characterized in that: In step 2, the node clustering coefficient C d : A coefficient that reflects the degree of aggregation between nodes in the distribution network or communication network topology, which represents the node V d The degree of connection with its adjacent nodes is calculated as follows: C d =[2deg(V d )] / [n(n-1)] Where n represents the total number of nodes in the distribution network or communication network.

Images