CN-122020543-A - High-speed railway network key station identification method and system based on multi-feature fusion TOPSIS

CN122020543ACN 122020543 ACN122020543 ACN 122020543ACN-122020543-A

Abstract

The invention discloses a high-speed railway network key site identification method based on multi-feature fusion TOPSIS, which comprises the steps of S1, constructing a high-speed railway network model, carrying out multi-dimensional feature extraction based on the high-speed railway network model, S2, constructing a decision matrix, standardizing to obtain a standardized decision matrix, S3, constructing a weighted normalization matrix, determining index weights according to the distribution condition of each index data by utilizing an entropy weight method, carrying out weighting treatment on the standardized matrix based on the weights to obtain the weighted normalization matrix, S4, comparing an object to be evaluated with an optimal solution and a worst solution based on the TOPSIS method, and calculating relative closeness coefficients of the object to be evaluated, thereby realizing comprehensive descending order of the objects, wherein sites with larger relative closeness coefficients are key sites. Corresponding systems, apparatuses, electronic devices, and computer-readable storage media are also disclosed.

Inventors

DING SHUXIN
FENG SHUANGYIN
PU CUNLAI
LI RUICHEN
SUN YANHAO
FU QINGYUN

Assignees

中国铁道科学研究院集团有限公司
中国铁道科学研究院集团有限公司通信信号研究所
北京华铁信息技术有限公司
北京锐驰国铁智能运输系统工程技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260129

Claims (10)

1. A high-speed railway network key station identification method based on multi-feature fusion TOPSIS is characterized by comprising the following steps: s1, constructing a high-speed rail network model and extracting multidimensional features based on the high-speed rail network model; S2, constructing a decision matrix and standardizing to obtain a standardized decision matrix; S3, constructing a weighted normalization matrix, wherein the weighted normalization matrix is obtained by determining index weights according to the distribution condition of each index data by utilizing an entropy weight method and carrying out weighting treatment on the normalized matrix based on the weights; s4, comparing the object to be evaluated with the optimal solution and the worst solution based on the TOPSIS method, and calculating the relative closeness coefficient of the object to be evaluated, so that comprehensive descending order of the objects is realized, wherein the sites with larger relative closeness coefficient are key sites.
2. The method for identifying the key sites of the high-speed railway network based on the multi-feature fusion TOPSIS as claimed in claim 1, wherein the step S1 comprises the following steps: s11, modeling a high-speed rail network by adopting a complex network idea; s12, extracting multidimensional features based on a multidimensional characterization method combining the high-speed rail network model with global indexes, local indexes and propagation capacity indexes; The global index comprises a medium centrality, a near centrality and a PageRank value, wherein the medium centrality is used for measuring the role of a node serving as an intermediary or a bridge in a network and describing key control force of the node on a delay propagation path; The local index comprises a degree centrality and a local clustering coefficient, wherein the degree centrality is used for reflecting the connection quantity of the nodes and the neighbors thereof and reflecting the diffusion capacity of the nodes in local propagation; the sum propagation capability index comprises a propagation centrality and is used for measuring influence of the node in the propagation process.
3. The method for identifying the key sites of the high-speed railway network based on the multi-feature fusion TOPSIS according to claim 2, wherein the step S11 comprises the following steps: (1) Abstracting an actual high-speed rail transportation system into a network model based on topological structure characteristics, and adopting a graph Representing the network model, wherein A collection of sites is represented and, Representing a set of edges, and wherein the nodes Indicating the high-speed rail station, edge Representing a site With the site A direct railway connection relationship exists between the two; (2) Defining the weights of edges in the network model as deferred propagation probabilities, first, for the weights of edges In the sense of delaying a slave site Propagation to adjacent sites According to the running rule and delay propagation mechanism of the high-speed train, the delay propagation probability is determined by the running frequency and the running reliability of the train between stations.
4. The method for identifying the key stations of the high-speed railway network based on the multi-feature fusion TOPSIS as claimed in claim 3, wherein the betweenness centrality is defined as the ratio of the number of the shortest paths passing through a certain point to all the shortest paths in the network, the betweenness centrality The mathematical definition of (a) is shown in the following formula (1): (1); Wherein, the Representing nodes And The number of shortest paths between them indicates Node And Passing through nodes between The number of shortest paths of (a); proximity centrality is defined as the inverse of the average of the distances of a node to all other nodes, the greater the proximity centrality of a node, the closer the node is to the other nodes, in a centered position Is defined as shown in the following formula (2): (2); Wherein, the Representing nodes To all other nodes Is used for the distance of (a), Representing the total number of all nodes in the network; The PageRank value The calculation formula of (2) is shown in the following formula (3): (3); Wherein, the Representing nodes Representing the importance score of the node in the network; For a damping coefficient, for indicating the probability that the user continues to browse the next node, The probability that the user randomly jumps to any node; Representing the total number of all nodes in the network, and calculating the denominator of the basic weight; Representing all pointing nodes Of the node sets, i.e Is a set of ingress neighbor nodes; Representing into a neighboring node PageRank value of (C) representing a node The importance of itself; Representing nodes Degree of egress, i.e. node Number of points to other nodes; the centrality is defined as the normalized value of the node Is defined as shown in the following formula (4): (4); Wherein, the Is a node Is of degree centrality, representing nodes A degree-related importance score in the network; Is a node Refers to the total number of other nodes directly connected with the node; For the total number of all nodes in the network, Maximum value that may be owned by a single node in an undirected network; the local aggregation coefficient reflects the possibility that the neighboring nodes of the node are also mutually communicated, and the local aggregation coefficient The calculation formula of (2) is shown in the following formula (5): (5); Wherein, the Is a node Reflecting the node The aggregation tightness degree of the neighbor nodes; Is a node The number of edges actually present between all neighboring nodes; Is a node Degree value of (i.e. with node) The total number of directly connected neighbor nodes; The number of edges which can exist among the neighbor nodes of the node at most is represented, so as to unify the counting logic of the undirected graph; The transmission center PC is used for measuring influence of the node in the transmission process, is used for measuring infection and easy comprehensive capacity of the node, and is used for measuring the node Is of the propagation center of (3) Is defined as the following formula (6): (6); Wherein, the Representing the propagation centrality of a node, representing the comprehensive influence score of the node in the propagation process; the weight coefficient is in the range of 0 to 1, and is used for balancing the importance of propagation capability and susceptibility by adjusting Nodes with strongest propagation, highest susceptibility or greatest comprehensive influence can be found; Representing nodes The outward propagation capability index of (a) reflects the capability of a node to propagate information or viruses to other nodes; weight coefficient corresponding to susceptibility, and The summation is 1, so that the weight distribution of the two indexes is ensured to be reasonable; Representing nodes And the infected susceptibility index of the node reflects the difficulty level of the influence of other node transmission.
5. The method for identifying the key sites of the high-speed railway network based on the multi-feature fusion TOPSIS according to claim 4, wherein the step S2 comprises the following steps: s21, constructing a decision matrix, including the steps of Individual evaluation objects The evaluation targets are high-speed rail stations, and the original data matrix The expression is shown as a formula (7): (7); Wherein, the Represent the first The object of evaluation is at the first Observations on the individual indicators; S22, for the decision matrix And (3) carrying out standardization treatment on the evaluation indexes, wherein the standardization method is shown as a formula (8): (8); Wherein, the And Respectively represent the first The maximum value and the minimum value of the individual indexes, and marking the matrix after the standardized treatment as: 。
6. The method for identifying the key sites of the high-speed railway network based on the multi-feature fusion TOPSIS according to claim 5, wherein the step S3 comprises the following steps: s31, normalizing the data in the matrix subjected to the normalization processing according to columns to obtain the first Under the index of Specific gravity of individual object As shown in formula (9): (9); Wherein, the Representing the data in the matrix after normalization, The number of the evaluation objects is the column number; s32, calculating the first Entropy of information entropy of each index, wherein the information entropy of the index is used for measuring uncertainty of the index, the first Entropy value of information entropy of individual index The calculation formula of (2) is shown as formula (10): (10); S33, based on the introduced difference coefficient To reflect the magnitude of the ability of the index to distinguish the evaluation object, as shown in formula (11): (11); Coefficient of difference The larger the index, the stronger the distinguishing ability of the index is, and the difference coefficient is based Determining weights of various indexes The following formula (12) shows: (12);; The weight of each index has the following characteristics: ; S34, utilizing the weight of each index Weighting each data in the matrix after normalization to obtain a weighted normalization matrix The following formula (13): (13)。
7. The method for identifying the key sites of the high-speed railway network based on the multi-feature fusion TOPSIS according to claim 6, wherein the step S4 comprises the following steps: S41, determining a positive ideal solution and a negative ideal solution, wherein the positive ideal solution is the maximum value of each index, the negative ideal solution is the minimum value of each index according to a formula (14), and the positive ideal solution is determined according to a formula (15): (14); (15); s42, calculating the first based on the Euclidean distance The distance between the individual object and the positive ideal solution and the negative ideal solution is determined according to formula (16) Determining the distance between the object and the ideal solution according to equation (17) Distance of individual objects from the negative ideal solution; (16); (17); S43, based on the first Distance of individual object from positive ideal and the first The distance between the object and the negative ideal solution determines the proximity of the object to the positive ideal solution and the negative ideal solution, thereby defining the relative proximity coefficient As shown in formula (18): (18); Wherein when (when) The closer to 1, the description object The closer to the ideal solution, the better the overall performance, when The closer to 0, the description object The closer to the negative ideal solution, the worse its overall performance; S44, according to the relative closeness coefficient Sequencing all objects in a descending order to obtain a final comprehensive evaluation result, wherein in the identification problem of key stations of a high-speed rail network, the relative closeness coefficient is calculated according to the relative closeness coefficient The larger site is the critical site.
8. A multi-feature fusion TOPSIS-based high-speed railway network key station identification system for implementing the method of any one of claims 1-7, comprising: The model construction and feature extraction module is used for constructing a high-speed rail network model and extracting multidimensional features based on the high-speed rail network model; the matrix construction and standardization module is used for constructing a decision matrix and standardizing to obtain a standardized decision matrix; The weighted normalization matrix construction module is used for constructing a weighted normalization matrix and comprises the steps of determining index weights according to the distribution condition of each index data by utilizing an entropy weight method, and carrying out weighted treatment on the normalized matrix based on the weights to obtain the weighted normalization matrix; The key site identification module is used for comparing the object to be evaluated with the optimal solution and the worst solution based on the TOPSIS method and calculating the relative closeness coefficient of the object to be evaluated, so that comprehensive descending order sorting of the objects is realized, wherein the sites with larger relative closeness coefficient are key sites.
9. An electronic device comprising a processor and a memory, the memory storing a plurality of instructions, the processor configured to read the instructions and perform the method of any of claims 1-7.
10. A computer readable storage medium storing a plurality of instructions readable by a processor and for performing the method of any one of claims 1-7.

Description

High-speed railway network key station identification method and system based on multi-feature fusion TOPSIS Technical Field The invention relates to the technical field of transportation and complex network analysis, in particular to a method and a system for identifying key stations of a high-speed railway network based on multi-feature fusion TOPSIS, which comprehensively utilize complex network theory, a statistical learning method and a multi-index decision model, the importance of the stations in the high-speed railway network is quantitatively evaluated and ordered, so that the efficient identification of the key stations is realized, and the method belongs to the cross research category of railway transportation management, traffic planning and network safety protection. Background Along with the rapid development of high-speed railways in China, a high-speed railway network becomes a high-speed railway system with the maximum world regulation and highest operation efficiency. The huge high-speed rail network not only bears the transregional rapid passenger flow transportation task, but also plays an important role in promoting regional economic integration and improving social trip efficiency. However, as the scale of networks continues to expand, operational safety and stability issues become increasingly prominent. Once some critical sites are subjected to operation interruption, equipment failure or external interference, large-scale propagation may be delayed, and even the transportation capacity of the whole network is greatly reduced, which has serious influence on traffic order and socioeconomic operation. Therefore, how to identify key sites in a huge high-speed rail network becomes an important research topic in academia and engineering applications. The conventional key node identification method mostly depends on a single network topology index, such as centrality, betweenness centrality, compactness centrality, feature vector centrality, pageRank and the like. Although such methods can reflect some of the characteristics of the station in the network topology, the following disadvantages exist: 1. Index singleness-a single index cannot fully reflect the multidimensional function of a site in a complex network. For example, centrality emphasizes the number of local connections, while mediacy emphasizes path intermediation, with different emphasis, and it is difficult to cover the comprehensive impact of sites. 2. Neglecting the operation characteristics, the traditional method is mostly based on a static topological structure, but the operation characteristics (such as train operation performance) of the high-speed rail cannot be fully combined, so that the result is deviated from the actual operation condition. 3. The existing method usually adopts simple weighting or experience weighting when processing multiple indexes, lacks an objective and reasonable index weight determining method, and has strong subjectivity and insufficient stability. In recent years, multi-attribute decision methods have been introduced into complex network node evaluation problems. The TOPSIS (Technique for Order Preference by Similarity to Ideal Solution, approaching the ideal solution sequencing method) has the advantages of simple and convenient calculation and visual result because the TOPSIS can sequence the solutions according to the relative proximity to the ideal solution and the negative ideal solution on the basis of comprehensively considering multiple indexes. However, in the problem of identifying key sites of a high-speed railway network, a technical scheme capable of effectively fusing a plurality of network topology indexes with operation attribute features and determining index weights by using an objective method so as to efficiently, objectively and comprehensively identify the key sites is not yet available. In summary, the existing key site identification method still has defects in terms of comprehensiveness, objectivity and practicability, and an improved method based on multi-feature fusion and TOPSIS is needed to be provided so as to improve the accuracy and the robustness of key site identification. Disclosure of Invention The invention aims to overcome the defects of the prior art and provides a method for identifying key stations in a high-speed railway network based on multi-feature fusion TOPSIS. The method fully considers the complexity and multidimensional characteristics of the high-speed railway network, synthesizes global indexes, local indexes and transmission capacity indexes, combines an objective weighting method (entropy weighting method) with a TOPSIS decision model, and realizes comprehensive evaluation and sequencing of the importance of the sites, thereby scientifically identifying the key sites with the greatest influence on the stability and the transportation efficiency of the network. The first aspect of the invention provides a method for identifying key sites o