Comparison method for multi-source topological data of digital railway road network

Abstract

The invention discloses a digital railway network multi-source topological data comparison method, which comprises the following steps: S1, constructing a railway operation topological network based on train operation data to form a time sequence topological reference for reflecting the actual use state of a railway network; s2, constructing a railway physical topology network based on the infrastructure related data, wherein the railway physical topology network is used for describing railway lines, stations and engineering connection relations thereof; and S3, performing topology comparison and consistency analysis between the railway operation topology network and the railway physical topology network to form a topology check result. According to the method, the specific linear space-time continuity constraint of the railway network is introduced, the railway operation topology is constructed as the space-time reference, and the railway operation topology is gradually compared and reasoned with the physical topology data with heterogeneous sources, so that the automatic consistency analysis and checking among the multi-source topology data are realized, and the problem that the consistency of the multi-source topology data can only be checked manually item by item is solved. And large-scale and continuously updated engineering requirements are difficult to meet.

Description

Technical Field

[0001] This invention relates to the field of railway informatization and transportation information processing technology, specifically to a method for comparing multi-source topology data of a digital railway network. Background Technology

[0002] With the continuous advancement of digital and intelligent railway construction, railway network topology data has become a fundamental data resource supporting line planning, transportation organization, simulation analysis, capacity assessment, and digital twin modeling. However, compared with road traffic, power networks, or communication networks, railway networks have highly specialized and strongly constrained structural characteristics. Their topological form is simultaneously constrained by multiple factors, including engineering construction structure, transportation organization logic, and train operation rules, resulting in significant uniqueness in the expression methods, granular levels, and evolution mechanisms of railway network data. Currently, methods for constructing and comparing multi-source topology data are primarily mature in the fields of highways, aviation, and waterways. These methods typically rely on high-precision maps, remote sensing imagery, or continuous GPS trajectories, achieving topological element matching through spatial proximity, field consistency, or geometric similarity. These methods generally implicitly assume the following preconditions: data sources are from the same or nearly the same source; the semantic definitions of nodes and edges are highly consistent; and a reliable high-precision spatial reference exists. However, these preconditions are often difficult to simultaneously satisfy in a multi-source railway data environment, making it difficult for existing technologies to be directly applied to the comparison and verification between railway operational topology and physical topology.

[0003] Within the railway industry, the maintenance and updating of existing network topology data still heavily relies on manual entry and review. On the one hand, this method is labor-intensive, time-consuming, and lacks real-time performance; on the other hand, any delays in the entry, review, or aggregation stages can easily lead to distortion or asynchrony of topology information across the country, thus affecting the reliability of capacity analysis, transportation organization optimization, and decision support. Furthermore, different business systems, catering to different application scenarios, exhibit significant differences in their recording methods and focus areas for the network, resulting in substantial variations in scale, structure, and representation of multi-source data. The lack of a unified topology alignment mechanism necessitates manual item-by-item verification, making it difficult to meet the demands of large-scale, continuous updates.

[0004] In the actual railway information system, railway network topology data is mainly distributed from the following types of systems: (1) Planning and design and engineering construction system, which forms a physical structure topology with lines, stations and sections as the core. Data updates have obvious phased and lagging characteristics, and at the macro scale, the left line of the section or station is often the main recording object; (2) Operation data formed by transportation organization and dispatch system, with trains as the core, records the arrival and departure times, running order and passing relationships of trains between different stations, which reflects the actual use status of the railway network and reflects the operation logic at the meso scale; (3) Existing railway geographic information or ledger system, which mainly maintains station location and line information in a static way. Some data lack complete line geometric descriptions, while professional systems such as signaling and dispatching focus on equipment or track circuit information at the micro scale.

[0005] For a long time, the aforementioned data has been maintained separately by different railway bureaus and different business lines, lacking a unified and automated topology comparison and verification mechanism. This has made it difficult to form a unified topology network that truly reflects the railway's operational status on a national scale in a timely manner. Therefore, there is an urgent need for a technical solution that can still achieve effective comparison and verification of multi-source railway topology data under conditions of incomplete spatial information and semantic inconsistencies. Summary of the Invention

[0006] This invention addresses the problems in existing technologies by disclosing a method for comparing multi-source topology data of digital railway networks. By introducing the linear spatiotemporal continuity constraint unique to railway networks, this invention constructs a railway operation topology as a spatiotemporal benchmark and compares and infers it step by step with physical topology data from heterogeneous sources. This enables automatic consistency analysis and verification between multi-source topology data, solving the problem that relying solely on manual item-by-item verification makes it difficult to meet the needs of large-scale, continuous updates in engineering projects.

[0007] This invention is achieved through the following technical solution:

[0008] This invention first provides a method for comparing multi-source topology data of a digital railway network, comprising the following steps:

[0009] S1: Construct a railway operation topology network based on train operation data to form a time-series topology benchmark that reflects the actual usage status of the railway network;

[0010] S2: Construct a railway physical topology network based on infrastructure-related data to describe railway lines, stations, and their engineering connections;

[0011] S3: Perform topology comparison and consistency analysis between the railway operation topology network and the railway physical topology network to generate topology verification results.

[0012] As a further step, the method for constructing the railway operation topology network includes:

[0013] S11: Obtain railway train operation data. The train operation data is divided by railway bureau and stored in log form line by line, including: train code, station / line junction name, station / line junction code, arrival / departure information;

[0014] S12: Preprocess railway train operation data;

[0015] S13: Construct a set of station nodes;

[0016] S14: Construct train operation routes;

[0017] S15: Construct directed connections in the operation topology based on the sequential relationships in the train's running path.

[0018] S16: Aggregate multiple logical sections with the same starting and ending stations and the same direction of operation to form a stable operating section node.

[0019] As a further step, S12 includes:

[0020] S121: Read trajectory data, remove records with invalid or missing key information, and convert station data into interval representation;

[0021] S122: Group the data by train code, and within each train group, sort it in ascending order by station number to ensure the formation of the correct set of operating trajectories.

[0022] As a further step, constructing the station node set includes: traversing all trajectory records to extract unique station codes and station names, creating a graph node for each unique station, and adding the graph node to the set.

[0023] As a further step, S15 includes:

[0024] S151: Construct the logical sub-interval topology;

[0025] S152: Construct the logical interval topology.

[0026] As a further step, the method for constructing the logical sub-interval topology includes:

[0027] S1511: Traverse the train path for each train ordered site sequence Iterate to generate train operating sections. Establish connection and logical sub-intervals and put gather;

[0028] S1511: According to the train's direction of travel Establish in the diagram The directed edges, where, Indicates the starting station. This indicates the termination site.

[0029] As a further step, the method for constructing the logical interval topology includes:

[0030] S1521: Traverse the set of logical subintervals generated by S1511 According to the origin and destination stations Group and aggregate the sub-intervals to form a single interval;

[0031] S1522: Construction Directed edges, and simultaneously establish logical intervals. With all the sites it includes The mapping relationship is used to form a complete road network topology.

[0032] As a further step, the method for constructing the railway physical topology network includes:

[0033] S21: Load data from planning and design systems, infrastructure ledger systems, or publicly available geographic information;

[0034] S22: Analyze the node and connection elements in the data of S21, extract their basic attribute information, unify the attribute field structure of nodes and connections in different data sources, and remove elements that are obviously irrelevant or lack key attributes.

[0035] S23: Constructing a railway microphysical topology network based on node elements and their connection relationships;

[0036] S24: Perform hierarchical aggregation on the railway micro-physical topology network to construct a railway meso-physical topology network suitable for operational topology comparison.

[0037] As a further step, S24 includes:

[0038] S241: Within the same station area, multiple micro-nodes with the same engineering attributes or belonging to the same functional unit are aggregated and merged into a single physical station node.

[0039] S242: Merge consecutive nodes in the same direction and within the same engineering interval to form physical topological units that express the interval structure;

[0040] S243: During the aggregation process, the mapping relationship between nodes before and after aggregation is preserved to support subsequent backtracking from the meso-topology to the micro-structure.

[0041] As a further step, S3 includes:

[0042] S31: Identify high-confidence anchor nodes across the entire network, with the anchor nodes serving as the initial constraints in the comparison process;

[0043] S32: Starting from the anchor node, perform topology diffusion search in the running topology network according to the train running sequence, and simultaneously search for candidate nodes in the preset neighborhood in the physical topology network to form directional path matching.

[0044] S33: During the topology diffusion process, a multi-level confidence level determination is performed on the correspondence between running topology nodes and physical topology nodes;

[0045] S34: After completing one round of topology diffusion, new anchor points are automatically selected from the confirmed high-confidence matching nodes. S31~S33 are repeated to gradually cover a larger range of network intervals with the comparison results. Through multiple rounds of iteration, the correspondence between the running topology and the physical topology gradually converges.

[0046] S305: For nodes that still cannot be automatically determined, force-guided layout or manual verification methods are used for auxiliary analysis to form topology verification results.

[0047] The features and beneficial effects of this invention are as follows:

[0048] (1) This invention introduces the linear spatiotemporal continuity constraint unique to railway networks, constructs railway operation topology as spatiotemporal benchmark, and compares and infers it step by step with physical topology data from heterogeneous sources, thereby realizing automatic consistency analysis and verification between multi-source topology data, solving the problem that it can only rely on manual item-by-item verification, which is difficult to meet the engineering needs of large-scale and continuous updates.

[0049] (2) This invention systematically adapts to the industry characteristics of railway operation data being “time-dense and space-sparse”; it does not rely on one-to-one node correspondence or high-precision spatial trajectory, significantly reducing the data prerequisite requirements; and it achieves stable comparison of railway topology at the national scale through anchor point driving and gradual diffusion mechanism.

[0050] (3) This invention shifts from a focus on spatial geometry to a focus on temporal order and topological relationships, fully utilizing the advantages of high temporal granularity and wide coverage of railway operation data; it moves from one-time global matching to progressive comparison and convergence, allowing for local uncertainty and gradual correction; and it shifts from a strict one-to-one correspondence assumption to a cognitive framework based on confidence level grading, dynamically characterizing node correspondences. Instead of pursuing one-time, global, precise matching under imperfect data conditions, it treats railway operation data as long-term, repetitive, and temporally continuous observations of the physical railway network, gradually approximating the true topological structure under uncertain conditions through anchor-driven topological diffusion and multi-round inference mechanisms. It supports the formation of daily, updatable railway operation topology views, providing real-time support for management decisions. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of the method for comparing multi-source topology data of digital railway networks according to an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of the initial operation topology of passenger transport nationwide.

[0054] Figure 3 This is a schematic diagram of the initial operation topology of freight transport nationwide.

[0055] Figure 4 (a) Schematic diagram of topology identification for introducing incomplete geographic information constraints; Figure 4 (b) Schematic diagram for initializing locations with incomplete geographic information;

[0056] Figure 5 This is the topology diagram after applying this method as described in the embodiments of the present invention. Detailed Implementation

[0057] To facilitate understanding of the present invention, a more comprehensive description of the present invention will be given below, and embodiments of the present invention will be provided, but this does not limit the scope of the present invention.

[0058] Figure 2 and Figure 3The initial operating topologies for passenger and freight transport nationwide are respectively. As we know, the data for both are quite chaotic and cannot directly identify specific line information. Due to the following significant characteristics of railway topology data: (1) Inherent lack of spatial information. Railway operation data takes station-level arrival and departure records as the basic unit and does not provide continuous, high-frequency spatial trajectory information, making it difficult to support trajectory-based spatial matching methods; (2) There is no natural one-to-one correspondence between the operating topology and the physical topology. The two have systematic differences in naming system, coding rules and topology granularity, making it impossible to assume a simple correspondence; (3) In the national-scale railway network, duplicate node names, aliases, missing and incorrect nodes are common, making the reliability of methods that rely solely on string or field comparison insufficient; (4) The large network scale and frequent cross-bureau operation place higher demands on the computational complexity, stability and engineering feasibility of the method. Therefore, this invention provides a method for comparing multi-source topology data of digital railway networks. By introducing the linear spatiotemporal continuity constraint unique to railway networks, a railway operation topology is constructed as a spatiotemporal benchmark. This topology is then compared and reasoned with heterogeneous physical topology data from different sources step by step. This achieves automatic consistency analysis and verification between multi-source topology data, solving the problem that relying solely on manual item-by-item verification makes it difficult to meet the needs of large-scale, continuous updates in engineering projects.

[0059] like Figure 1 As shown, a method for comparing multi-source topology data of a digital railway network includes...

[0060] S1: Construct a railway operation topology network based on train operation data to form a time-series topology benchmark that reflects the actual usage status of the railway network;

[0061] S2: Construct a railway physical topology network based on infrastructure-related data to describe railway lines, stations, and their engineering connections;

[0062] S3: Perform topology comparison and consistency analysis between the railway operation topology network and the railway physical topology network to generate topology verification results.

[0063] This invention shifts the focus from spatial geometry to temporal order and topological relationships, fully leveraging the advantages of high temporal granularity and wide coverage of railway operation data. It moves from one-time global matching to progressive comparison and convergence, allowing for local uncertainty and gradual correction. Furthermore, it moves from a strict one-to-one correspondence assumption to a cognitive framework based on confidence levels, dynamically characterizing node correspondences. Instead of pursuing a one-time, globally precise match under imperfect data conditions, it treats railway operation data as a long-term, repetitive, and temporally continuous observation of the physical railway network. Through anchor-driven topological diffusion and multi-round inference mechanisms, it gradually approximates the true topological structure under uncertain conditions. This supports the formation of daily, updatable railway operation topology views, providing real-time support for management decisions.

[0064] In some embodiments, constructing a railway operation topology network includes:

[0065] S11: Obtain railway train operation data;

[0066] The data includes at least the train identification, station name or code, arrival time, and departure time.

[0067] Train operation data is divided by railway bureau and stored in log format, row by row. Each row contains the following fields: 1) Train code, a unique identifier belonging to the railway bureau; 2) Station / line junction name: the name of the physical station; 3) Station / line junction code: the globally unique ID of the physical station (used as a vertex in the topology graph); 4) Arrival / departure information: including arrival train number, arrival time, departure train number, and departure time. See Table 1 for details.

[0068] Table 1

[0069]

[0070] Note: Trains (carriages) with the same number have the same number within the jurisdiction of the same railway bureau. The number may change when crossing railway bureaus, but the train track remains continuous in time and space.

[0071] S12: Preprocess railway train operation data;

[0072] S121: Read the trajectory data, remove records with invalid or missing key information (such as station codes and times), and convert the station data into interval representation.

[0073] S122: Sorting and Grouping: Group the data by train code, and within each train group, sort in ascending order by station number to ensure the formation of a correct set of train tracks. .

[0074] S13: Construct the set of station nodes:

[0075] Iterate through all trajectory records, extract unique (station code, station name) pairs, create a graph node for each unique station, and add it to the set. .

[0076] S14: Construct the train running path; the running path does not contain continuous spatial coordinate information, but implicitly expresses the train's travel direction and adjacent relationships in the network through time sequence.

[0077] The operation records are grouped according to train identification, and within each train group, they are sorted according to time sequence or station number to form a train operation path sequence. Each train operation path consists of a series of ordered station nodes, which are used to describe the actual operation sequence of the train in the railway network.

[0078] By aggregating a large number of train operation paths, a set of operation paths with high redundancy covering the entire railway network can be formed, providing a stable data foundation for subsequent topology abstraction.

[0079] S15: Construct directed connections in the operation topology based on the sequential relationships in the train's running path.

[0080] The station records in the running path are abstracted into station nodes in the running topology, which are used to represent logical nodes with clear stopping or passing significance during the train operation. For the continuous running process of the train between adjacent stations, interval nodes or logical interval nodes are introduced to describe the interval structure in the running path, thereby avoiding the simple compression of complex interval relationships into a single edge.

[0081] The logical interval is an abstract set concept used to express the macroscopic fact that station A and station B are connected, without focusing on the specific physical tracks. It can contain multiple logical sub-intervals. For example, the Beijing South to Tianjin South section of the Beijing-Shanghai High-Speed ​​Railway (logical interval) includes the Beijing South to Tianjin South up section (logical sub-interval 1) and the Beijing South to Tianjin South down section (logical sub-interval 2).

[0082] The logical intervals include single-track intervals and double-track intervals. A single-track interval is a physical space where only one main line (physical track) is laid between two adjacent stations. Within this interval, trains traveling in both directions share the same physical track resource. In terms of train operation, only trains traveling in one direction are typically allowed to occupy the space at any given time (or they must pass through a passing station to meet), exhibiting strict physical conflict constraints. Logically, the two logical sub-intervals corresponding to these intervals are mapped onto the same physical entity. A double-track interval is a physical space where two independent main lines (physical tracks) are laid between two adjacent stations. Typically, one track is designated to primarily handle upward transport, while the other handles downward transport. Physically, the two tracks are spatially isolated and do not interfere with each other, supporting independent operation of trains traveling in both directions on their respective physical tracks simultaneously. This provides a higher line throughput capacity than a single-track interval. Logically, each logical interval corresponds to a single-track interval and contains two logical sub-intervals, with each logical interval mapped onto an independent physical track (entity).

[0083] A logical subinterval is the smallest directed connection unit in the train operation topology, specifically referring to a unidirectional virtual path between two adjacent nodes (such as from station A to station B). Subintervals possess directionality and atomicity, and must be unidirectional, forming the basis for constructing macroscopic lines; their attributes may include basic operational attributes such as mileage.

[0084] S151: Constructing the logical sub-interval topology:

[0085] S1511: Interval Generation: Traverse the train path for each train ordered site sequence Iterate to generate train operating sections. Establish connection and logical sub-intervals and put gather.

[0086] S1511: Topology Construction: Based on Train Direction of Travel Establish in the diagram The directed edge.

[0087] S152: Construct logical interval topology

[0088] S1521: Interval Merging and Generation: Traversing the logical sub-interval set generated in step three According to the origin and destination stations Group and aggregate the sub-intervals to form a single interval.

[0089] S1522: Establish Topological Relationships: Establish the connection relationships between logical intervals and stations in the graph, i.e., construct... Directed edges, and simultaneously establish With all of its contents The mapping relationship is used to form a complete road network topology.

[0090] Operational topology is an abstract representation of a system built upon nodes and connections. It is primarily used to describe the connectivity and logical structure between various elements (stations, track sections) within a railway network. Operational topology focuses only on topological connection logic and does not involve actual geometric distances, geographical coordinates, physical directions, or spatial scales. Through this abstraction, operational topology can express not only "station-to-station" relationships but also the hierarchical structure of "station-section-to-station" that has engineering and organizational significance in railway operations.

[0091] For each adjacent station node in a train route, a directed edge structure is constructed, pointing from the preceding station node to the corresponding interval node, and then from the interval node to the following station node, to explicitly represent the directionality of train operation. The direction of the directed edges strictly follows the chronological order of train operation, thereby introducing inherent directional and sequential constraints from railway operation at the topology level. By aggregating recurring connections in different train routes, the actual operational topology can be gradually strengthened, while mitigating the impact of occasional anomalies or single erroneous data.

[0092] S16: Aggregate multiple logical intervals with the same starting and ending stations and consistent operating directions to form stable operating interval nodes; count the frequency of interval occurrence during the merging process to measure the stability and reliability of the operating interval.

[0093] In some embodiments, a method for constructing a railway physical topology network based on infrastructure-related data includes:

[0094] S21: Load data from planning and design systems, infrastructure ledger systems, or publicly available geographic information. The loaded data must at least contain descriptions of railway-related nodes and their connections, but is not required to contain complete, continuous line geometry.

[0095] S22: Preprocess the data in S21;

[0096] The node and connection elements are parsed to extract their basic attribute information; the attribute field structure of nodes and connections in different data sources is unified; and elements that are obviously irrelevant or lack key attributes are removed to reduce the uncertainty of subsequent topology construction.

[0097] S23: Construct a railway microphysical topology network based on node elements and their connection relationships.

[0098] Engineering entities such as station nodes and line nodes in physical data are abstracted into physical topology nodes, and the engineering connections between them are abstracted into topology connection relationships. These connection relationships can originate from line route descriptions, adjacent node references, or engineering structural associations.

[0099] S24: Perform hierarchical aggregation on the railway micro-physical topology network to construct a railway meso-physical topology network suitable for operational topology comparison. Specifically, this includes:

[0100] S241: Within the same station area, multiple micro-nodes with the same engineering attributes or belonging to the same functional unit are aggregated and merged into a single physical station node.

[0101] S242: Merge consecutive nodes in the same direction and within the same engineering interval to form physical topological units that express the interval structure;

[0102] S243: During the aggregation process, the mapping relationship between nodes before and after aggregation is preserved to support subsequent backtracking from the meso-topology to the micro-structure.

[0103] Through the above processing, a railway physical topology network based on engineering structure and expressible at multiple scales is formed, providing a candidate space for comparison between operational topology and physical topology.

[0104] In some embodiments, a method for performing topology comparison and consistency analysis between the railway operating topology network and the railway physical topology network includes:

[0105] S31: Global Anchor Locking: Identify high-confidence anchor nodes across the entire network. These anchor nodes serve as initial constraints in the comparison process to reduce the propagation of uncertainty.

[0106] The anchor node is a node that has a high degree of uniqueness or reliable identification characteristics in both the running topology and the physical topology. The methods for determining it include, but are not limited to: having a name that is completely identical after removing common suffixes, having an authoritative system confirmation mark, or being a key node that has obvious uniqueness in the topology structure.

[0107] S32: Directed Path Matching: Starting from the anchor node, a topology diffusion search is performed in the running topology network according to the train operation sequence, and candidate nodes are simultaneously searched within a preset neighborhood in the physical topology network to form a directed path match. During the diffusion process, the unique linear structure characteristics and operational continuity constraints of the railway network are utilized, that is, adjacent nodes have strong stability in terms of topological order, upstream and downstream relationships, and number of branches.

[0108] S33: Multi-level confidence determination: During the topology diffusion process, a multi-level confidence determination is performed on the correspondence between the running topology nodes and the physical topology nodes.

[0109] When nodes are highly consistent in name, attributes, and topological adjacency, a high-confidence matching relationship is established; when node attributes are missing or ambiguous, topological reasoning is performed based on the continuity relationship of its upstream and downstream confirmed nodes to infer its most likely corresponding position; for nodes with severely insufficient information or anomalies, a corresponding relationship is not established for the time being and they are recorded as pending processing.

[0110] S34: Iterative Comparison. After completing one round of topology diffusion, new anchor points are automatically selected from the confirmed high-confidence matching nodes, and S31~S33 are repeated to gradually cover a larger range of network intervals with the comparison results. Through multiple rounds of iteration, the correspondence between the running topology and the physical topology gradually converges.

[0111] S35: Auxiliary analysis and manual verification of unmatched nodes:

[0112] For nodes that cannot be automatically determined, force-guided layout or manual verification methods are used for auxiliary analysis to form the final usable topology verification results.

[0113] This invention systematically adapts to the industry characteristics of railway operation data, which are "time-dense and spatially sparse." It does not rely on one-to-one node correspondence or high-precision spatial trajectories, significantly reducing the data prerequisite requirements. Through anchor-driven and progressive diffusion mechanisms, it achieves stable comparison of railway topology at the national scale. This method enables the automatic and effective comparison and verification of multi-source railway topology data even under conditions of incomplete spatial information and semantic inconsistencies, solving the problem that relying solely on manual item-by-item verification is insufficient to meet the needs of large-scale and continuous updates in engineering projects.

[0114] Example

[0115] like Figure 4 and Figure 5 As shown, this embodiment uses the national-scale railway network as an example to specifically illustrate the comparison method of multi-source topology data of the digital railway network proposed in this invention. However, this embodiment does not constitute a limitation on the scope of protection of this invention. It includes the following steps:

[0116] A1. Data Loading and Parameter Setting:

[0117] Railway operational topology data and railway physical topology data are loaded as multi-source topology data for comparison. Among them:

[0118] The running topology network contains more than 9,000 topology nodes and more than 10,000 topology edges;

[0119] The physical topology network contains more than 12,000 topology points with geographic coordinate attributes.

[0120] To ensure the engineering feasibility of topology expansion and candidate search, the spatial neighborhood search range of physical topology nodes is set to 300 km to limit the candidate matching space of running topology nodes in the physical topology.

[0121] A2. Determination of global high-confidence anchor nodes:

[0122] Within the entire road network, at least one pair of nodes with high determinism in both the running topology and the physical topology is selected as the initial anchor node pair through manual confirmation or automatic screening.

[0123] In this embodiment, a node with a consistent and unique name, "Line 104," is randomly selected as the initial anchor point to establish a one-to-one correspondence between the running topology and the physical topology. This anchor point node serves as the initial constraint for subsequent topology diffusion and inference.

[0124] A3. Anchor-based neighborhood topology expansion and initial round matching:

[0125] Starting from the anchor node, a breadth-first search is performed in the running topology along the train running sequence, and candidate nodes are searched in the physical topology within a preset neighborhood.

[0126] During the topology expansion process, a method combining local name matching and topology adjacency consistency is adopted, and the candidate node search process is accelerated by using a spatial index structure (such as a quadtree).

[0127] When a running topology node finds a candidate node in the physical topology that meets the matching rules, a corresponding relationship is established, and the node is included in the subsequent expanded anchor point set; nodes that fail to match are not matched and are temporarily left to wait for subsequent iterations.

[0128] After completing the initial topology expansion, the algorithm performed 179 expansion iterations, yielding the following preliminary comparison results (the data is sensitive and some parts are redacted):

[0129] Specific matching nodes: 50;

[0130] Fuzzy matching nodes: 11;

[0131] Unmatched nodes: 29.

[0132] A4. Iterative cleaning and anchor point regeneration:

[0133] After the initial matching is completed, non-explicitly matched nodes are restored to an unmatched state, and only confirmed explicitly matched nodes are retained as stable anchor points.

[0134] Subsequently, new anchor node pairs are automatically selected from the confirmed explicit matching nodes, and steps A2 to A3 are repeated.

[0135] During multiple iterations, anchor nodes included, but were not limited to, “Line 107 Junction”, “Dingjiazhuang”, and “Sankou”. As the anchor nodes expanded, the number of explicitly matched nodes continued to grow, while the number of fuzzy and unmatched nodes gradually decreased.

[0136] When no new explicit matching nodes are generated in multiple consecutive iterations, the matching results are considered to have basically converged.

[0137] In this embodiment, after multiple iterations, the number of explicitly matching nodes gradually increased from the initial 50 to over 82, achieving a precise matching rate of over 90%. Finally, within a 300km search radius, no new explicitly matching nodes were added, and matching ceased.

[0138] A5. Auxiliary Analysis of Unmatched Nodes

[0139] For the small number of nodes that still cannot be automatically matched, visualization methods such as force-guided layout are used for auxiliary analysis, and the final topology verification results are confirmed by manual review.

[0140] like Figure 2 As shown in this embodiment, it can be verified that the method proposed in this invention can achieve stable comparison and rapid convergence of multi-source railway topology data under the conditions of inconsistent naming, incomplete spatial information, and the existence of duplicate names and missing nodes in the national-scale railway network. It has good engineering feasibility and practical value.

[0141] Comparative Example

[0142] Using the traditional exact matching method based on node names, a comparison was made on national-scale railway network data, and the overall accurate matching rate was approximately 87.0%. This method relies on the consistency and uniqueness of node names, and it is difficult to effectively handle common problems in the railway field such as duplicate names, aliases, abbreviations, and historical names. Matching errors are mainly concentrated in hub sections, cross-bureau boundary sections, and areas with significant differences in station levels.

[0143] In comparison, the method of this invention achieves a precise matching rate of 90.0% under the same data conditions, with an overall improvement of approximately 3.0 percentage points. Although the improvement is limited in terms of global statistical indicators, it does not stem from a simple rule aggregation, but rather from a fundamental change in the matching mechanism.

[0144] Specifically, traditional name matching methods, once mismatched in a local area, are prone to generating a chain of errors in subsequent topology construction or fusion processes, leading to misconnected segments, broken paths, or distorted network structures. In contrast, this invention introduces topology continuity constraints based on railway operation logic and an anchor-driven diffusion reasoning mechanism, which can automatically correct local matching results based on upstream and downstream topology relationships in the event of name inconsistencies or missing information, thereby significantly reducing structural errors in key nodes and key segments.

[0145] Therefore, the technical effect of this invention is not only reflected in the improvement of the overall matching ratio, but also in its ability to ensure the integrity and consistency of the railway network topology. This capability has a decisive impact on subsequent capacity assessment, transportation organization analysis, and digital railway applications, and is something that traditional name-based matching methods cannot achieve. It effectively solves the problem that relying solely on manual item-by-item verification is insufficient to meet the needs of large-scale, continuously updated engineering projects.

[0146] This embodiment also provides a computer device applicable to a method for comparing multi-source topology data of a digital railway network, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the method for comparing multi-source topology data of a digital railway network as proposed in the above embodiment.

[0147] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a method for comparing multi-source topology data of a digital railway network as proposed in the above embodiments.

[0148] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0149] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0151] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0152] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0153] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for comparing multi-source topology data of a digital railway network, characterized in that: Includes the following steps: S1: Construct a railway operation topology network based on train operation data to form a time-series topology benchmark that reflects the actual usage status of the railway network; S2: Construct a railway physical topology network based on infrastructure data to describe railway lines, stations and their engineering connections; S3: Perform topology comparison and consistency analysis between the railway operation topology network and the railway physical topology network to generate topology verification results.

2. The method for comparing multi-source topology data of a digital railway network according to claim 1, characterized in that: The method for constructing the railway operation topology network includes: S11: Obtain railway train operation data. The train operation data is divided by railway bureau and stored in log form line by line, including: train code, station / line junction name, station / line junction code, arrival / departure information; S12: Preprocess railway train operation data; S13: Construct a set of station nodes; S14: Construct train operation routes; S15: Construct directed connections in the operation topology based on the sequential relationships in the train's running path; S16: Aggregate multiple logical sections with the same starting and ending stations and the same direction of operation to form a stable operating section node.

3. The method for comparing multi-source topology data of a digital railway network according to claim 2, characterized in that: S12 includes: S121: Read trajectory data, remove records with invalid or missing key information, and convert station data into interval representation; S122: Group the data by train code, and within each train group, sort it in ascending order by station number to ensure the formation of the correct set of operating trajectories.

4. The method for comparing multi-source topology data of a digital railway network according to claim 2, characterized in that: Constructing the station node set includes: traversing all trajectory records to extract unique station codes and station names, creating a graph node for each unique station, and adding the graph node to the set.

5. The method for comparing multi-source topology data of a digital railway network according to claim 2, characterized in that: S15 includes: S151: Construct the logical sub-interval topology; S152: Construct the logical interval topology.

6. The method for comparing multi-source topology data of a digital railway network according to claim 5, characterized in that: The method for constructing the logical sub-interval topology includes: S1511: Traverse the train path for each train ordered site sequence Iterate to generate train operating sections. Establish connection and logical sub-intervals and put gather; S1511: According to the train's direction of travel Establish in the diagram The directed edges, where, Indicates the starting station. This indicates the termination site.

7. The method for comparing multi-source topology data of a digital railway network according to claim 5, characterized in that: The method for constructing the logical interval topology includes: S1521: Traversing the set of logical subintervals generated by S1511 According to the origin and destination stations Group and aggregate the sub-intervals to form a single interval; S1522: Construction Directed edges, and simultaneously establish logical intervals. With all the sites it includes The mapping relationship is used to form a complete road network topology.

8. The method for comparing multi-source topology data of a digital railway network according to claim 1, characterized in that: The method for constructing the railway physical topology network includes: S21: Load data from planning and design systems, infrastructure ledger systems, or publicly available geographic information; S22: Analyze the node and connection elements in the data of S21, extract their basic attribute information, unify the attribute field structure of nodes and connections in different data sources, and remove elements that are obviously irrelevant or lack key attributes. S23: Constructing a railway microphysical topology network based on node elements and their connection relationships; S24: Perform hierarchical aggregation on the railway micro-physical topology network to construct a railway meso-physical topology network suitable for operational topology comparison.

9. The method for comparing multi-source topology data of a digital railway network according to claim 8, characterized in that: S24 includes: S241: Within the same station area, multiple micro-nodes with the same engineering attributes or belonging to the same functional unit are aggregated and merged into a single physical station node. S242: Merge consecutive nodes in the same direction and within the same engineering interval to form physical topological units that express the interval structure; S243: During the aggregation process, the mapping relationship between nodes before and after aggregation is preserved to support subsequent backtracking from the meso-topology to the micro-structure.

10. The method for comparing multi-source topology data of a digital railway network according to claim 1, characterized in that: S3 includes: S31: Identify high-confidence anchor nodes across the entire network, with the anchor nodes serving as the initial constraints in the comparison process; S32: Starting from the anchor node, perform topology diffusion search in the running topology network according to the train running sequence, and simultaneously search for candidate nodes in the preset neighborhood in the physical topology network to form directional path matching. S33: During the topology diffusion process, a multi-level confidence level determination is performed on the correspondence between running topology nodes and physical topology nodes; S34: After completing one round of topology diffusion, new anchor points are automatically selected from the confirmed high-confidence matching nodes. S31~S33 are repeated to gradually cover a larger range of network intervals with the comparison results. Through multiple rounds of iteration, the correspondence between the running topology and the physical topology gradually converges. S305: For nodes that still cannot be automatically determined, force-guided layout or manual verification methods are used for auxiliary analysis to form topology verification results.

Images