Method for identifying weak line of power distribution network

Abstract

The invention discloses a weak line identification method for a power distribution network. The method comprises the following steps: step 1, establishing an importance index of each line in the power distribution network, a line transmission contribution betweenness index in the power distribution network, a voltage deviation index and a line centrality index; 2, respectively endowing all indexes established in the step 1 with combination weights; and step 3, using an optimized TOPSIS algorithm to evaluate the current operation quality degree of each index according to the weight of each index.

Description

Technical Field

[0001] The present invention relates to the field of distribution network identification methods, and specifically to a method for identifying weak lines in a distribution network. Background Art

[0002] With the large-scale access of distributed energy, the structure of the distribution network has changed from radial to multi-source interconnection. The surge in nodes and the bidirectional power flow have increased the difficulty of assessing line vulnerability. Existing methods mostly start from a single dimension of topology or operating state: the former is based on complex network theory and uses structural parameters such as node degree and betweenness centrality to identify weak links; the latter analyzes electrical risks through power flow calculation and cascading failure simulation. Although some studies have tried to integrate multiple indicators and use the analytic hierarchy process or entropy weight method to construct an evaluation system, there are still significant defects. First, the coupling relationship between topology and operating parameters has not been fully modeled. For example, the dynamic power flow changes caused by distributed energy are difficult to reflect in static structure indicators, and network connection characteristics are often ignored in operating analysis for their impact on fault propagation. Second, single indicators are one-sided: topology indicators are prone to missing the risks of high-load lines, while operating indicators are difficult to capture inherent defects such as structural islands. In addition, the traditional evaluation model uses Euclidean distance to calculate weights, without eliminating the correlation between indicators (such as the strong correlation between load rate and voltage drop), resulting in distorted weights. More critically, existing methods rely on static ideal solutions and cannot adapt to the spatio-temporal differences in vulnerability caused by changes in the penetration rate of distributed energy, resulting in evaluation results deviating from reality. These problems lead to insufficient accuracy of traditional methods in complex distribution networks and are difficult to support the high-reliability power supply demand. There is an urgent need to establish a comprehensive evaluation model with dynamic coupling, index decoupling, and scenario adaptability. Summary of the Invention

[0003] Aiming at the defects in the existing methods for identifying weak lines in distribution networks, such as insufficient coupling between topology and operating parameters, inaccurate handling of index correlation, and poor dynamic adaptability, the present invention proposes a method for evaluating the vulnerability of distribution network lines based on multi-source data fusion and dynamic weight optimization.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] A method for identifying weak lines in a distribution network, comprising the following steps:

[0006] Step 1: Establish the importance index of each line in the distribution network, the betweenness index of line transmission contribution in the distribution network, the voltage offset index, and the line centrality index;

[0007] Step 2: Assign combined weights to all the indexes established in Step 1;

[0008] Step 3: Use the TOPSIS algorithm based on Mahalanobis distance to evaluate the current operation quality of each index according to the weight of each index obtained in Step 2.

[0009] Furthermore, in step 1, an importance index of each line in the distribution network is established based on the PageRank algorithm, and the importance index of each line in the distribution network established based on the PageRank algorithm is iterated using the LeaderRank optimization algorithm to obtain the importance index of each line after iterative stabilization.

[0010] Furthermore, in step 1, the transmission contribution index M of the distribution network line l As shown in the following formula:

[0011]

[0012] Where: S ij-max is the maximum transmission power of line l; S ij is the transmission power of line l during normal operation; i and j are the nodes at both ends of line l respectively.

[0013] Furthermore, in step 1, the voltage offset indicator V * As shown in the following formula:

[0014]

[0015] Where: V represents the actual operating voltage of the circuit; V N Expressed as the rated voltage of the line; ΔV max Indicates the maximum allowed voltage offset.

[0016] Furthermore, in step 1, the line centrality index C l =As shown in the following formula:

[0017]

[0018] Where N * represents the number of lines in the system; i and j represent two different nodes; d ij Indicates the impedance value of the line connecting each node.

[0019] Furthermore, in step 2, all indicators established in step 1 are assigned combined weights based on the grey relational weighting method and the anti-entropy weight method.

[0020] The method of the present invention first constructs a comprehensive evaluation index system covering topological connectivity, power flow balance, voltage stability, and distributed energy penetration rate to achieve multi-dimensional coupling analysis of structural parameters and operating states; secondly, an improved TOPSIS evaluation model is adopted. By introducing Mahalanobis distance to replace the traditional Euclidean distance to calculate the index closeness, the interference of correlation between indicators is eliminated, and a virtual negative ideal solution is set based on dynamic scenario characteristics to optimize the ideal solution selection mechanism; further, an entropy weight - analytic hierarchy process is combined to generate a time-varying weight matrix to dynamically adapt to the impact of the change in distributed energy penetration rate on the vulnerability distribution; finally, by establishing an IEEE 39-node simulation model with a high proportion of distributed energy, the effectiveness and robustness of the evaluation model in complex scenarios are verified.

[0021] Compared with the prior art, the advantages of the present invention are as follows:

[0022] By integrating multi-source indicators such as topological connectivity, power flow balance, voltage stability, and distributed energy penetration rate, the present invention constructs a multi-dimensional coupling evaluation system, solving the limitation of single-dimensional analysis of traditional methods; using Mahalanobis distance to replace Euclidean distance to calculate the index closeness, eliminating the redundant interference of highly correlated indicators such as load rate and voltage sag, and significantly improving the objectivity of weight assignment; combining entropy weight - analytic hierarchy process to dynamically generate a time-varying weight matrix, and introducing a virtual negative ideal solution to optimize the ideal solution selection mechanism, effectively adapting to the spatio-temporal differences in vulnerability caused by the change in distributed energy penetration rate, and overcoming the scenario adaptability defects of traditional static models. The present invention solves the problems of misjudgment caused by one-sidedness of indicators, weight distortion, and static ideal solutions in traditional methods, significantly improving the identification accuracy of weak lines and scenario adaptability, and providing technical support for the planning and operation and maintenance of high-reliability distribution networks. Brief Description of the Drawings

[0023] Figure 1 It is the schematic diagram of the embodiment of the present invention. Detailed Embodiment

[0024] The present invention will be further described below with reference to the drawings and embodiments.

[0025] As Figure 1 shown, this embodiment discloses a method for identifying weak lines in a distribution network, including the following steps:

[0026] Step 1: Establish multiple indicators for distribution network lines, including the importance index of each line in the distribution network, the betweenness index of the transmission contribution of lines in the distribution network, the voltage offset index, and the line centrality index. The specific descriptions are as follows:

[0027] (1) In this embodiment, the importance index of each line in the distribution network is established based on the PageRank algorithm, and the importance index of each line in the distribution network established based on the PageRank algorithm is iterated using the LeaderRank optimization algorithm to obtain the importance index of each line after iterative stabilization.

[0028] The basic idea behind the PageRank algorithm is to define a random walk model, or first-order Markov chain, on a directed graph. This model describes the behavior of a random walker randomly visiting various paths along the directed graph. Under certain conditions, the probability of visiting each path converges to a stationary distribution in the extreme case. At this point, the stationary probability of each path is its PageRank value, representing the importance of the path. PageRank is defined recursively, and PageRank calculations can be performed using an iterative algorithm.

[0029] When the PageRank algorithm is applied to the identification of vulnerable lines in the distribution network of this embodiment, the concept of a web page is replaced by lines in the distribution network. The specific calculation formula of the PageRank algorithm in this embodiment is as follows:

[0030]

[0031] In formula (1): n is the total number of lines in the system; l i is the line connected to line l; I(l) is the set of lines connected to line l; C out (l i ) is line l i The chain outflow is equal to the number of lines connected to it; σ is the damping coefficient, which has a random value but can affect the convergence speed; R(l) and R(l i ) are the lines l and l calculated by the PageRank algorithm respectively. i The importance index of , referred to as importance here.

[0032] The convergence criterion of the PageRank algorithm in this embodiment is as follows:

[0033] max|R t+1 (l)-R t (l)|≤δ (2)

[0034] In formula (2): R t+1 (l) and R t (l) are line l at the t+1th and tth iterations respectively i The importance of ; δ is the convergence accuracy. According to the expression formula (1) of the PageRank algorithm above, it can be converted into a matrix representation to better reflect its iterative characteristics, as shown below:

[0035] R t+1 = R t ·G(3)

[0036]

[0037] In formulas (3), (4), and (5): R t+1 and R t are the importance column vectors of each line for the (t + 1)-th and t-th iterations; G is a non-periodic matrix with randomly taken and irreducible values; H is the connection relationship matrix, which has been normalized; e is an n-dimensional column vector with all elements in each row being 1; a is an n-dimensional marking vector. If line l i is not connected to any line, correspondingly let a i be taken as 1, otherwise 0; S is the matrix that fuses the original connection relationship and the special scenario weight adjustment.

[0038] Therefore, when using the PageRank algorithm to sort a network, there are still certain problems: when changing different connection graphs each time, it is necessary to optimize the parameters in the algorithm to maximize the importance R(l) of each target object calculated; secondly, each iteration of the PageRank algorithm needs to evenly distribute the R(l) of the target object to other evaluation objects in the network, and this equal distribution method is unreasonable for the value calculation in the network.

[0039] Therefore, to address the above problems, LU et al. proposed the LeaderRank algorithm, which calculates the ranking score of each link based on the degree of each connection link, the LeaderRank scores of neighbor links, and the connection relationship between links. The LeaderRank algorithm not only focuses on the local connection of nodes but also, by introducing "super nodes" in the network and performing iterative propagation, can comprehensively consider the structural information of the entire distribution network. This enables it to more comprehensively evaluate the status and role of nodes in the network, thereby more accurately judging the vulnerability of nodes. In an actual distribution network, the operating state may change over time, such as load fluctuations, access and withdrawal of distributed energy, and line fault repair. The LeaderRank algorithm can adapt to these changes by dynamically adjusting the score values of nodes and timely reflecting the dynamic changes in node vulnerability. Unlike some traditional vulnerability assessment methods that require specific prior knowledge or assumption conditions, the LeaderRank algorithm can be directly applied to distribution networks of different scales and structures without in-depth understanding of the specific characteristics of the network. This makes it more flexible and convenient in practical applications and can adapt to various complex distribution network environments. For newly added nodes or lines, the LeaderRank algorithm can easily incorporate them into the calculation without re-designing the assessment method. This provides convenience for the expansion and upgrade of the distribution network and can continuously and effectively evaluate node vulnerability.

[0040] Therefore, in this embodiment, the LeaderRank optimization algorithm is used to iterate the importance index of each line in the distribution network established based on the PageRank algorithm, and the importance index of each line after the iteration is stabilized is obtained. The specific algorithm process is as follows:

[0041] At the initial time, a unit of importance value is assigned to other targets in the network except the background target K g as follows:

[0042]

[0043] In formula (6): R i represents the initial importance value of other targets in the network except the background target K g ; R g is the initial importance value of the background target K g .

[0044] During the iteration process of the LeaderRank algorithm, the R(l) values of all lines are updated according to formula (3) as follows:

[0045]

[0046] In formula (7): k j-out is the out-degree of the target; a ji is the element in the adjacency matrix; R l (t + 1) is the importance index value of line l at the (t + 1)-th iteration; R l (t) is the importance index value of line l at the t-th iteration.

[0047] When the importance values of all targets are stabilized, the iteration is stopped, and the R(l) values of the background target are evenly distributed to each ordinary node, and the final importance in the network can be obtained as follows:

[0048]

[0049] In formula (8): R l * represents the importance after the iteration is stabilized, n represents the number of targets in the network; R l is the importance index value of line l.

[0050] (2) Since the PangRank algorithm only analyzes the topological relationship between an interconnected system. However, when evaluating a power network, not only its topological connection relationship should be considered, but also its operating state should be analyzed. Therefore, in this embodiment, the line transmission contribution betweenness index, voltage deviation index, and line centrality index are also established. Among them:

[0051] 2a) Line transmission contribution betweenness index

[0052] The traditional betweenness index indicates the number of times the shortest path passes between any two nodes. However, considering that there is power flow on any line in the power grid, the traditional index is no longer applicable. The power flow in the system is reasonably distributed according to circuit laws and does not propagate on the path with the shortest electrical distance, which can exactly complement the original betweenness index. Therefore, this embodiment proposes a line transmission contribution betweenness index to measure the ability of a line to withstand system power fluctuations. The line transmission contribution betweenness index in the distribution network is shown in the following formula:

[0053]

[0054] In formula (9): S ij-max is the maximum transmission power of line l; S ij is the transmission power of line l under normal operation; i and j are the nodes at both ends of line l respectively.

[0055] When the line power flow is close to or has exceeded the line capacity limit, the index value will increase significantly, so it can better reflect the importance difference of lines with different load rates.

[0056] 2b) Voltage deviation index

[0057] During the peak load period, the load absorbs more power from the power grid to achieve system stability. When the load power growth exceeds the maximum range of the transmission power, the voltage drops too much, resulting in a voltage collapse phenomenon. On the contrary, when the input power is greater than the output power, the voltage will rise. Therefore, the line voltage deviation can reflect the real-time operating state of the bus and evaluate the relationship between the bus load size and the power grid input power at that moment. The voltage deviation index established in this embodiment is shown in the following formula:

[0058]

[0059] In formula (10): V represents the actual operating voltage of the line; V N represents the rated voltage of the line; ΔV max represents the maximum allowable voltage deviation.

[0060] In most cases, voltage deviation is more common in lines that play an important role in power transmission, output lines of generators, and connecting lines of major loads.

[0061] 2c) Line centrality index

[0062] The line centrality index, like the closeness index in the previous chapter, represents the relative position of the line in the entire power network. The larger the index value, the more lines are connected between power networks through it, indicating that it has a greater interconnection function. As shown in the following formula:

[0063]

[0064] In formula (11), N * represents the number of lines in the system; i and j represent two different nodes; d ij Indicates the impedance value of the line connecting each node.

[0065] Step 2: Assign combined weights to all indicators established in step 1. In this embodiment, a weighting method based on grey correlation and anti-entropy weighting is used to assign combined weights to all indicators established in step 1. The specific description is as follows:

[0066] (A) Weighting method based on grey relational degree

[0067] Grey relational analysis is a system analysis method. This method reveals the correlation between internal factors of a system by calculating the correlation between each factor. It has low data requirements and does not require a large number of samples and typical distributions, which is very beneficial for many practical problems. The steps of the grey relational weighting method are as follows:

[0068] A1) Data standardization. Since the dimensions of each influencing factor are different, each dimension is standardized. The standardization method used in this embodiment is Max-Min, and the calculation formula is as follows:

[0069] X i =(X i -X imax ) / (X imax -X imin ) (12)

[0070] In formula (12): X i represents the value of the i-th data sample; X imax Indicates the maximum value in the sample data; X imin Indicates the minimum value in the sample data.

[0071] A2) Determine expert weights using the AHP method.

[0072] The nine-scale method is used to compare the indicators of each layer, and the judgment matrix A=a is obtained. ij (m×m), and after comparing them pairwise, we can get the weight ω of each indicator. ijIt is the importance degree interval of index i relative to index j when comparing the importance of indexes; m is the order of the judgment matrix.

[0073] A3) Calculation of grey correlation coefficient.

[0074] For the target object, each dimension is calculated to obtain a new sequence. Each point in this sequence represents the correlation between the subsequence and the corresponding dimension of the reference sequence. The correlation degree is shown in the following formula:

[0075]

[0076] In formula (13): x0(k) is the value of the reference sequence at time k, x i (k) is the value of the i-th characteristic sequence at time k, and ρ is the resolution coefficient, which is used to adjust the difference degree of the correlation coefficient.

[0077] A4) Calculation of grey correlation coefficient, the formula is as follows:

[0078] r i = ω i ·ζ i (k) (14)

[0079] In formula (14): ω i is the weight of each index; ζ i (k) is the grey correlation coefficient of each index.

[0080] Finally, taking the grey correlation coefficient as the objective benchmark, the indexes are weighted according to the principle of "the higher the correlation degree, the greater the weight", effectively getting rid of the limitations of traditional subjective weighting methods, making the weight distribution more in line with the internal law of the data, and improving the comprehensive credibility of the weight.

[0081] (B), Objective weight assignment method based on anti-entropy weight method

[0082] The anti-entropy weight method is an objective weighting method, and its concept comes from the concepts of coefficient of variation and entropy weight. This method believes that in the evaluation problem, if the difference of the indexes is larger, then the anti-entropy is larger, and the larger the anti-entropy, the higher the corresponding weight should be. The steps to calculate the objective weight of the evaluation indexes by using the anti-entropy weight method are as follows:

[0083] B1) For each evaluation object, obtain its original data under each index and organize it into an evaluation matrix X, as shown in the formula:

[0084]

[0085] In formula (15): X ij is the value of the j-th index of the i-th evaluation object

[0086] B2) Normalize the evaluation matrix X to obtain a normalized matrix. Here, the Max-Min normalization method is applied to obtain matrix ζ, as shown in the formula:

[0087]

[0088] In formula (16): X ij is the value of the j-th index of the i-th evaluation object, and X j is the value of the j-th index.

[0089] B3) According to the concept of the anti-entropy weight method, calculate the anti-entropy value of each index, as shown in the following formula:

[0090]

[0091] In formula (16): p ij is the proportion of the normalized value ζ ij of the i-th evaluation object in the total sum of the normalized values of all evaluation objects for this index, reflecting the relative contribution of a single evaluation object within this index; ζ ij is the value of the i-th evaluation object corresponding to the j-th index after Max-Min normalization.

[0092] B4) Determine the weight of each index, as shown in the following formula:

[0093]

[0094] In formula (18): h * is the anti-entropy value of a single index, and n is the total number of indexes participating in the weight calculation.

[0095] (C) Combined weight

[0096] After assigning weights to each index through the weight assignment method based on grey correlation degree in step (A) above and assigning weights to each index through the objective weight assignment method based on the anti-entropy weight method in step (B) above, then based on the weights obtained from these two methods, determine the combined weight of each index, as shown in the following formula:

[0097]

[0098] In formula (19), α1 and α2 are distribution factors; ω C1 , ω k1 , ω N1 are the weights of the line centrality index, line transmission contribution degree betweenness index, and voltage deviation index of the subjective weight respectively; ω C2 , ω K2 , ω N2 are the weights of the line centrality index, line transmission contribution degree betweenness weight, and voltage deviation index of the objective weight respectively.

[0099] Step 3: Use the TOPSIS algorithm based on Mahalanobis distance to evaluate the pros and cons of the current operation of each indicator according to the weight of each indicator obtained in Step 2.

[0100] The research object of this embodiment is the distribution network in a single area, and there is no situation of ranking and making choices by the relative distances between alternative solutions and the positive and negative ideal solutions. Therefore, it is necessary to optimize the selection of the optimal and worst values of the indicators of the traditional evaluation object, and generally use the TOPSIS method, that is, the method of distance between the best and worst solutions, for optimized selection.

[0101] When traditional TOPSIS uses Euclidean distance to calculate the distance between solutions, it does not consider the coupling relationship between data. In the distribution network, there is often a certain correlation between various indicators. For example, voltage stability and power factor may affect each other. The Euclidean distance calculation cannot effectively handle this data coupling problem, which will lead to a decrease in evaluation accuracy. Therefore, this embodiment improves and optimizes the traditional TOPSIS, uses Mahalanobis distance to replace Euclidean distance to characterize the pros and cons between various indicators of the distribution network, forms an optimized TOPSIS method, and can effectively reduce the interference of the correlation between indicators.

[0102] In the evaluation of weak lines in the distribution network, using Mahalanobis distance can more accurately reflect the actual performance of each indicator and avoid evaluation deviations caused by indicator correlation. Compared with Euclidean distance, Mahalanobis distance can more accurately measure the pros and cons between various indicators of the distribution network. It weights different indicators more reasonably by considering the covariance matrix of the indicators, making the evaluation results more in line with the actual operation of the distribution network. And a virtual negative ideal solution is introduced to optimize the calculation of the closeness.

[0103] The optimized TOPSIS method in this embodiment can evaluate a single distribution network indicator more accurately and perfectly. The process of the optimized TOPSIS method in this embodiment is as follows:

[0104] (1) Establish an initial evaluation matrix according to the evaluation indicators and define the initial matrix C 3×n , as shown in Equation (17):

[0105]

[0106] In Equation (20): n represents the number of evaluation indicators; A1 is the actual observed or calculated value of the indicator; A2 and A3 are the optimal value and the worst value of the indicator; c 11 -c 1n , c 21 -c 2n , c 31 -c 3n respectively represent the actual value, the optimal value, and the worst value of the indicator.

[0107] (2) Normalization of the evaluation index matrix

[0108] Normalize each index in the initial matrix according to Equation (17) to obtain the normalized matrix Where:

[0109]

[0110] In Equation (21): c ij represents the element of the evaluation index matrix.

[0111] (3) Weight the normalized matrix

[0112] Obtain the weighted normalized matrix Z = (z ij ) according to the combined weights calculated in Step 2 3×n , where:

[0113]

[0114] In Equation (22): ω j is the weight of the j-th evaluation index; represents the element after normalization of the evaluation index matrix.

[0115] (4) Determine the optimal value vector and the worst value vector

[0116] The second row z + = (z 21 , z 22 ..., z 2n ) and the third row z - = (z 31 , z 32 ..., z 3n ) in the weighted normalized matrix Z are the optimal value vector and the worst value vector respectively, and the virtual negative ideal vector z* is obtained as shown in the following formula:

[0117]

[0118] In Equation (23): represent the virtual negative ideal values under different evaluation indexes respectively.

[0119] (5) Calculate the Mahalanobis distance and ratio between the actual value, the optimal value and the virtual negative ideal value vectors of the indexes, as shown in the following formula:

[0120]

[0121] In Equations (24) and (25): ∑ -1 is the inverse matrix of the covariance matrix formed between each index; z 1jrepresents the actual value of the j-th index; z 2j represents the optimal value of the j-th evaluation index; z ij represents the element of the weighted normalized matrix; represents the virtual negative ideal value of the first evaluation index.

[0122] (6) Calculate the relative closeness degree as shown in the following formula:

[0123]

[0124] In formula (26): is the Mahalanobis distance between the actual value and the optimal value of the i-th evaluation object's index; d i * is the Mahalanobis distance between the actual value of the i-th evaluation object's index and the virtual negative ideal value.

[0125] S i represents the closeness degree between a certain index and its optimized index. If sorted from small to large, S i the smaller the value, the closer it is to the optimal value level, indicating that its current operation is better. Conversely, the farther and worse it is.

[0126] Thus, based on the closeness degree S i the current operation quality of each index can be obtained, and each index can be sorted in descending order according to the quality degree, and then the weakness of each line can be determined.

[0127] To further elaborate the influence degree of the selected weak points on the distribution network in this embodiment, this embodiment also uses network efficiency to characterize the operation ability of the power grid system. Network efficiency expresses the ease of network information transmission evaluated from a global network perspective. When the network structure changes, its value also changes accordingly. In the power system, network efficiency usually refers to the shortest electrical distance E between the generator and the load, and its specific formula is as follows:

[0128]

[0129] In formula (24): n represents the number of nodes in the power system; d ij represents the shortest electrical distance between the generator node i and the load node j; G and D represent all the generator and load nodes in the power system.

[0130] When a certain node in the power system fails, the value of the network efficiency index becomes smaller. It indicates that when the node affecting the system weakness fails, the transmission ability of the power system also decreases accordingly, that is, the network efficiency decreases. And when the failed node in the system has a higher impact on the system weakness, the network efficiency is also lower.

[0131] The network power supply capacity is an indicator characterizing the effective power supply capacity of the distribution network, which can reflect the power transmission capacity of the distribution network during operation. If the decline in the power supply capacity of the system caused by the failure of a node in the distribution network is greater, it indicates that the node has a greater influence on the power supply of the system. The formula for the network power supply capacity F is as follows:

[0132]

[0133] In Equation (24): P represents the load level of the system at steady state; P * represents the load level of the system after the node is attacked.

[0134] The preferred embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings. The embodiments described in the present invention are only descriptions of the preferred embodiments of the present invention, and do not limit the concept and scope of the present invention. Among the various specific technical features described in the above specific embodiments, they can be combined in any suitable manner without contradiction. As long as such a combination does not violate the idea of the present invention, it should also be regarded as the content disclosed in this disclosure. To avoid unnecessary repetition, the present invention will not separately describe various possible combination methods.

[0135] The present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention and without departing from the design idea of the present invention, various variations and improvements made by those skilled in the art to the technical solutions of the present invention should fall within the protection scope of the present invention. The technical content claimed by the present invention has been fully recorded in the claims.

Claims

1. A method for identifying weak lines in a distribution network, characterized in that, It includes the following steps: Step 1: Establish the importance index of each line in the distribution network, the betweenness index of line transmission contribution in the distribution network, the voltage deviation index, and the line centrality index; Step 2: Assign combined weights to all the indexes established in Step 1 respectively; Step 3: Use the TOPSIS algorithm based on Mahalanobis distance to evaluate the current operation quality of each index according to the weight of each index obtained in Step 2.

2. The method for identifying weak lines in a distribution network according to claim 1, characterized in that, In Step 1, establish the importance index of each line in the distribution network based on the PageRank algorithm, and use the LeaderRank optimization algorithm to iterate the importance index of each line in the distribution network established based on the PageRank algorithm to obtain the importance index of each line after iterative stability.

3. The method for identifying weak lines in a distribution network according to claim 1, wherein In step 1, the line transmission contribution betweenness index M in the distribution network l is shown as follows: Where: S ij-max is the maximum transmission power of line l; S ij is the transmission power of line l under normal operation; i and j are the nodes at both ends of line l respectively.

4. The method for identifying weak lines in a distribution network according to claim 1, wherein In Step 1, the voltage offset index V * is as shown in the following formula: Where: V represents the actual operating voltage of the line; V N represents the rated voltage of the line; ΔV max represents the maximum allowable voltage deviation.

5. A method for identifying weak lines in a distribution network according to claim 1, characterized in that, In step 1, the line centrality index C l is shown as follows: Where N * represents the number of lines in the system; i and j represent two different nodes; d ij represents the impedance value of the lines connected between each pair of nodes.

6. The method for identifying weak lines in a distribution network according to claim 1, characterized in that In Step 2, assign combined weights to all the indexes established in Step 1 respectively based on the weighting method of grey correlation degree and the anti-entropy weight method.

Images