Integrated Management Method of Digital Twin Model for the Entire Lifecycle of Railway Infrastructure

By constructing an engineering entity structure division system and coding mapping rules, processing multi-source data and performing semantic mapping and quality assessment, the problem of data inconsistency at different stages of railway infrastructure has been solved, achieving efficient management of digital twin models and improving data utilization and intelligence level in the operation and maintenance phase.

CN123070458APending Publication Date: 2026-11-10LANZHOU JIAOTONG UNIV +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611103717.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-23
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The inconsistent data coding system, model structure, and semantic definition of railway infrastructure at different stages result in low utilization of completed and delivered data during the operation and maintenance phase. The lack of unified coding mapping rules and automated data quality assessment methods makes it impossible to achieve dynamic updates of digital twin models and efficient service calls, thus hindering the improvement of intelligent operation and maintenance levels.

Method used

Construct an engineering entity structure partitioning system, formulate coding mapping rules, collect and process multi-source data, use a data translation and semantic mapping engine to translate and align data, adopt a data quality assessment system for evaluation, generate quality score reports, build a digital twin model, provide operation and maintenance views and decision support services, and realize closed-loop data flow.

Benefits of technology

It achieves efficient alignment and reuse of data at different stages, improves the utilization efficiency of completed and delivered data in the operation and maintenance stage, reduces the consumption of manual translation and verification, ensures data consistency and integrity, supports operation and maintenance task generation and decision-making, and realizes intelligent operation and maintenance management of railway infrastructure.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123070458A_ABST
    Figure CN123070458A_ABST
Patent Text Reader

Abstract

This invention discloses an integrated management method for the entire lifecycle of railway infrastructure digital twin models, comprising: constructing an engineering entity structure partitioning system, dividing railway engineering into several entities according to levels; mapping design stage identifiers, construction stage identifiers, and operation and maintenance asset identifiers; performing format recognition and metadata standardization processing on the accessed multi-source data; translating and aligning the multi-source data using a data translation and semantic mapping engine; evaluating the translated and aligned data; constructing a digital twin model; and writing the digital twin model, quality score report, and change records back to the design and construction data sources through standardized interfaces, forming a closed-loop data flow from design, construction, joint commissioning and testing, to operation and maintenance. This invention enables seamless data connection and closed-loop flow across the design, construction, joint commissioning and testing, and operation and maintenance stages, improves the utilization rate of completed and delivered data, and provides technical support for the intelligent operation and maintenance management of railway infrastructure.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of railway facility informatization, specifically to a method for integrated management of the entire lifecycle of a digital twin model of railway infrastructure. Background Technology

[0002] Currently, railway infrastructure generates a large amount of heterogeneous data during the design, construction, joint commissioning and testing, and operation and maintenance stages, including BIM models, point cloud measurement data, construction records, IoT data collected by sensors, and maintenance records.

[0003] However, due to inconsistencies in data coding systems, model structures, and semantic definitions at different stages, the utilization rate of completed and delivered data during the operation and maintenance phase is low, resulting in problems such as low translation rates and difficulties in accessing the data. Furthermore, existing technologies lack an engineering entity partitioning system for operation and maintenance management, unified coding mapping rules, and automated data quality assessment and consistency verification methods. This makes it impossible to support dynamic updates and efficient service calls for digital twin models throughout their entire lifecycle, severely hindering the improvement of intelligent operation and maintenance levels for railway infrastructure.

[0004] Therefore, to solve the above problems, a digital twin model of railway infrastructure is needed for integrated management throughout its entire lifecycle, which can improve the utilization efficiency of completed and delivered data in the operation and maintenance phase and realize intelligent operation and maintenance management of railway infrastructure. Summary of the Invention

[0005] In view of this, the purpose of this invention is to overcome the deficiencies in the prior art and provide an integrated management method for the entire life cycle of digital twin models of railway infrastructure, which can improve the utilization efficiency of completed and delivered data in the operation and maintenance phase and realize intelligent operation and maintenance management of railway infrastructure.

[0006] The integrated management method for the entire lifecycle of railway infrastructure digital twin model of the present invention includes the following steps:

[0007] S1. Construct an engineering entity structure division system, divide the railway project into several entities according to levels, and define a standard attribute set for each entity;

[0008] S2. Develop and maintain coding mapping rules to map design phase identifiers, construction phase identifiers, and operation and maintenance asset identifiers;

[0009] S3. Collect and access multi-source data, and perform format recognition and metadata standardization processing on the accessed multi-source data;

[0010] S4. Translate and align the multi-source data using a data translation and semantic mapping engine;

[0011] S5. Evaluate the translated and aligned data based on a predefined data quality assessment system and generate a quality score report;

[0012] S6. Based on the qualified data, construct a digital twin model and provide operation and maintenance view, alarm correlation, operation and maintenance task generation and decision support services;

[0013] S7. Write back the digital twin model, quality score report, and change record to the design and construction data source through a standardized interface to form a closed-loop data flow of design, construction, commissioning and testing, and operation and maintenance.

[0014] Furthermore, some of the entities include line segments, sections, stations, substructures, components, and equipment units.

[0015] Furthermore, the encoding mapping rules include namespace conventions, hierarchical separators, and check bit rules.

[0016] Furthermore, the multi-source data includes BIM files, point cloud measurement data, construction logs, commissioning and testing records, IoT sensor time-series data, inspection records, and related documents and images.

[0017] Furthermore, the multi-source data is subjected to format recognition and metadata standardization processing, specifically including:

[0018] Perform format parsing on files or data streams from different sources to extract file header information, data field types, and unit identifiers;

[0019] Based on the preset metadata specifications, the field names, encoding methods and unit systems are uniformly converted; and a check mark is generated for fields that cannot be automatically recognized to ensure the consistency of data in subsequent translation and mapping.

[0020] Furthermore, the data translation and semantic mapping engine includes a semantic mapping module; the semantic mapping module generates candidate mappings based on word vector similarity calculation or based on a trained classifier, and triggers a manual review process for candidate mappings with confidence scores below a preset threshold.

[0021] Furthermore, the data quality assessment system uses a weighted scoring method to comprehensively evaluate completeness, consistency, and geometric accuracy, and marks entities with scores below a preset threshold as objects to be reviewed.

[0022] Furthermore, based on the qualified data, a digital twin model is constructed, specifically including:

[0023] Load the data that has passed the quality assessment into a predefined digital twin model template;

[0024] Based on the engineering entity division system, the data is instantiated into corresponding line, station, tunnel and equipment unit objects; and the geometric information and attribute information of the objects are associated and bound with the time series data to generate an operation model that supports operation and maintenance visualization and real-time monitoring.

[0025] Furthermore, the operation and maintenance task generation includes analyzing sensor data based on a predictive maintenance model to automatically generate a priority list of maintenance plans.

[0026] Furthermore, the standardized interfaces include REST-based or message queue-based interfaces.

[0027] The beneficial effects of this invention are as follows: The integrated management method for the entire lifecycle of a digital twin model of railway infrastructure disclosed in this invention achieves efficient alignment and reuse of data at different stages by constructing a unified engineering entity partitioning system and coding mapping rules. This improves the utilization efficiency of completed and delivered data during the operation and maintenance phase, reduces the consumption of manual translation and verification, and ensures data consistency and integrity. Simultaneously, the generated digital twin operation model can provide data support for operation and maintenance task generation and decision-making, realizing intelligent operation and maintenance management of railway infrastructure. Attached Figure Description

[0028] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0029] Figure 1 This is a schematic diagram illustrating the principle and process of the integrated management method of the present invention. Detailed Implementation

[0030] The present invention will be further described below with reference to the accompanying drawings, as shown in the figures:

[0031] This embodiment discloses a method for integrated management of the entire lifecycle of a digital twin model for railway infrastructure, including the following steps:

[0032] S1. Construct an engineering entity structure division system, divide the railway project into several entities according to levels, and define a standard attribute set for each entity;

[0033] S2. Develop and maintain coding mapping rules to map design phase identifiers, construction phase identifiers, and operation and maintenance asset identifiers;

[0034] S3. Collect and access multi-source data, and perform format recognition and metadata standardization processing on the accessed multi-source data;

[0035] S4. Translate and align the multi-source data using a data translation and semantic mapping engine;

[0036] S5. Evaluate the translated and aligned data based on a predefined data quality assessment system and generate a quality score report;

[0037] S6. Based on the qualified data, construct a digital twin model and provide operation and maintenance view, alarm correlation, operation and maintenance task generation and decision support services;

[0038] S7. Write back the digital twin model, quality score report, and change record to the design and construction data source through a standardized interface to form a closed-loop data flow of design, construction, commissioning and testing, and operation and maintenance.

[0039] This invention improves the utilization rate of completed delivery data and the level of intelligent operation and maintenance management by constructing an engineering entity division system, formulating unified coding mapping rules, realizing automatic translation and semantic mapping of multi-source data, and conducting data quality assessment and consistency verification. Then, it combines with the operation and maintenance data platform to achieve seamless connection and cyclic flow of data in the design, construction, commissioning and testing and operation and maintenance stages.

[0040] In this embodiment, step S1 involves first hierarchically dividing the engineering entities in railway engineering. Specifically, railway engineering is divided into levels such as line segments, sections, stations, substructures, components, and equipment units according to spatial distribution and functional characteristics. Line segments, as the highest-level entity, describe the overall engineering project spanning multiple regions; sections correspond to specific line segments; stations, as independent functional units, include buildings, equipment, and ancillary facilities; substructures correspond to the subdivided parts of stations and lines, such as bridge piers and tunnel lining sections; components are the basic building blocks of substructures, such as beams, sleepers, and supports; and equipment units include signals, power equipment, and monitoring sensors. For each type of entity, a standard attribute set is defined, including geometric parameters, physical attributes, operational status parameters, and maintenance parameters, to ensure data consistency and traceability across different levels.

[0041] By defining standard attribute sets for entities at different levels, such as line segments, sections, stations, substructures, components, and equipment units, the standardization and unified management of engineering data are achieved, improving the traceability and reusability of the data.

[0042] In this embodiment, step S2 requires the establishment of coding mapping rules to achieve integrated association of data from different stages. These rules are used to establish a one-to-one correspondence between design stage identifiers, construction stage identifiers, and operation and maintenance asset identifiers. The coding rules include namespace conventions, hierarchical separators, and checksum rules. For example, component codes generated in the design stage can be mapped to construction log entries formed during the construction stage through namespaces, and further aligned with asset codes in the operation and maintenance stage. Checksum rules prevent erroneous mappings caused by manual input or inconsistent formats, thereby ensuring the uniqueness and integrity of engineering data throughout its entire lifecycle.

[0043] By mapping design identifiers, construction identifiers, and operation and maintenance asset identifiers, the problem of data silos caused by inconsistent coding systems at different stages in traditional methods is solved, and data mutual recognition and interconnection throughout the entire life cycle are realized.

[0044] In this embodiment, step S3 involves generating multi-source heterogeneous data during engineering construction and operation, including BIM model files, laser point cloud measurement data, construction logs, commissioning and testing records, IoT sensor time-series data, inspection records, and multimedia documents and images. To ensure that this data can be used uniformly in the digital twin environment, it is necessary to first identify its format. Specifically, format parsing is performed on files or data streams from different sources to extract file header information, data field types, and unit identifiers; then, according to preset metadata specifications, field names, encoding methods, and unit systems are uniformly converted. For example, millimeters are converted to meters, and free text fields are mapped to standardized attributes. For fields that cannot be automatically parsed, a check mark is generated for manual confirmation and supplementation, ensuring data consistency during subsequent translation and mapping processes.

[0045] By parsing, uniformly converting, and marking multi-source heterogeneous data such as BIM files, point clouds, logs, and sensor data, the efficiency of integrating and utilizing multi-source heterogeneous data is improved, and the risk of inconsistency in subsequent data translation and mapping is reduced.

[0046] In this embodiment, in step S4, the normalized data needs to be unified at the semantic level. To this end, a data translation and semantic mapping engine is used to translate and align multi-source data. This engine includes a semantic mapping module that generates candidate mappings using word vector similarity calculation methods or based on the prediction results of a pre-trained classifier. For example, for the fields "pier height" and "pier height," they are identified as the same attribute through semantic similarity calculation, and the mapping is automatically completed. When the confidence level of a candidate mapping is lower than a preset threshold, a manual review process is triggered, thereby ensuring semantic consistency and correctness while maintaining the efficiency of automated processing.

[0047] By combining word vector similarity and classifier models with a semantic mapping module, candidate mappings are automatically generated and manual review is triggered, which significantly improves the accuracy and automation of data alignment and reduces manual workload.

[0048] In this embodiment, step S5 requires a quality assessment of the translated and semantically aligned data. This method predefines three core indicators: completeness, consistency, and geometric accuracy, and employs a weighted scoring method for comprehensive evaluation. For example, if a station's BIM model lacks information on some equipment units, the completeness score decreases; if the status records of a component in the construction log and sensor data are inconsistent, the consistency score decreases; and if the deviation between point cloud measurement data and the BIM geometric model exceeds a threshold, the geometric accuracy score decreases. After comprehensive evaluation, a quality score report is generated, and entities with scores below a preset threshold are marked as requiring review, providing a basis for subsequent manual intervention and data correction.

[0049] By employing a weighted scoring method to comprehensively evaluate completeness, consistency, and geometric accuracy, low-quality data can be identified and flagged in a timely manner, ensuring the accuracy and reliability of the digital twin model.

[0050] In this embodiment, in step S6, data that passes the quality assessment is loaded into a predefined digital twin model template. Based on the engineering entity structure division system, the data is instantiated into line, station, tunnel, and equipment unit objects, and geometric information, attribute information, and time-series data are bound together. The resulting digital twin model not only has visualization capabilities but also supports real-time monitoring. For example, in a tunnel lining structure unit, real-time sensor data can be directly mapped to the digital twin object, enabling dynamic monitoring of the lining's health status. Furthermore, sensor data can be analyzed based on a predictive maintenance model to automatically generate a maintenance task plan list and set priorities according to fault risk levels, allowing managers to efficiently schedule maintenance tasks.

[0051] By loading qualified data into a digital twin model template, instantiating engineering objects and binding geometric, attribute, and time-series information, an operation and maintenance view, alarm correlation, and predictive maintenance can be realized, supporting intelligent operation and maintenance and risk warning of railway infrastructure.

[0052] In this embodiment, in step S7, the digital twin model, quality score report, and change records are finally written back to the design and construction data sources through a standardized interface, thereby achieving a data closed loop between design, construction, commissioning and testing, and operation and maintenance. The standardized interface can be implemented based on a REST API or message queue mechanism, supporting cross-system data interaction. For example, the design unit can receive equipment hazard information discovered during the operation and maintenance phase in real time and make improvements in subsequent design optimizations, thus achieving continuous accumulation and optimized utilization of data throughout its entire lifecycle.

[0053] By writing back models, quality reports, and change records to design and construction data sources through REST or message queue interfaces, a closed-loop flow of data throughout its entire lifecycle is achieved, promoting continuous optimization and iteration of infrastructure information in the design, construction, and operation and maintenance stages.

[0054] This invention achieves standardized and unified management of engineering data. By constructing an engineering entity structure division system and standard attribute set, it ensures that the data of each level of railway engineering entity can be uniformly expressed and associated in the design, construction and operation and maintenance stages, thereby improving the traceability and reusability of the data.

[0055] It establishes a data link across the entire lifecycle of design, construction, and operation and maintenance. By formulating coding mapping rules, it maps design identifiers, construction identifiers, and operation and maintenance asset identifiers, enabling cross-stage data mutual recognition and interconnection, and avoiding the data silo problem caused by inconsistent coding systems in traditional methods.

[0056] It improves the efficiency of integrating and utilizing multi-source heterogeneous data. By performing format recognition and metadata standardization on the accessed multi-source data, and combining data translation and semantic mapping engines, it achieves efficient alignment and integration of multi-source heterogeneous data, significantly improving the utilization rate of data during the construction phase and the operation and maintenance phase.

[0057] It ensures data quality and reliability. Through a predefined data quality assessment system, it performs weighted scoring and comprehensive evaluation of integrity, consistency and geometric accuracy, which can promptly identify and mark low-quality data, thus ensuring the accuracy and reliability of the digital twin model.

[0058] It supports intelligent operation and maintenance and predictive maintenance, building or updating digital twin models based on qualified data to realize operation and maintenance views, alarm correlation, task generation, and decision support services. By analyzing sensor data through predictive maintenance models, potential risks can be identified in advance, and maintenance plan lists can be automatically generated, thereby reducing operation and maintenance costs and improving the safety and reliability of railway infrastructure.

[0059] A closed-loop data flow mechanism covering the entire lifecycle has been established. Through standardized interfaces, digital twin models, quality reports, and change records are written back to design and construction data sources, enabling closed-loop data flow throughout the entire lifecycle and promoting continuous optimization and iteration of railway infrastructure in the design, construction, and operation and maintenance stages.

[0060] Finally, 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 integrated management of the entire lifecycle of a digital twin model of railway infrastructure, characterized in that: Includes the following steps: S1. Construct an engineering entity structure division system, divide the railway project into several entities according to levels, and define a standard attribute set for each entity; S2. Develop and maintain coding mapping rules to map design phase identifiers, construction phase identifiers, and operation and maintenance asset identifiers; S3. Collect and access multi-source data, and perform format recognition and metadata standardization processing on the accessed multi-source data; S4. Translate and align the multi-source data using a data translation and semantic mapping engine; S5. Evaluate the translated and aligned data based on a predefined data quality assessment system and generate a quality score report; S6. Based on the qualified data, construct a digital twin model and provide operation and maintenance view, alarm correlation, operation and maintenance task generation and decision support services; S7. Write back the digital twin model, quality score report, and change record to the design and construction data source through a standardized interface to form a closed-loop data flow of design, construction, commissioning and testing, and operation and maintenance.

2. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The entities mentioned include line segments, sections, stations, substructures, components, and equipment units.

3. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The encoding mapping rules include namespace conventions, hierarchical separators, and check bit rules.

4. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The multi-source data includes BIM files, point cloud measurement data, construction logs, commissioning and testing records, IoT sensor time-series data, inspection records, and related documents and images.

5. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The system performs format recognition and metadata normalization on the accessed multi-source data, specifically including: Perform format parsing on files or data streams from different sources, and extract file header information, data field types, and unit identifiers; Based on the preset metadata specifications, the field names, encoding methods and unit systems are uniformly converted; and a check mark is generated for fields that cannot be automatically recognized to ensure the consistency of data in subsequent translation and mapping.

6. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The data translation and semantic mapping engine includes a semantic mapping module; the semantic mapping module generates candidate mappings based on word vector similarity calculation or based on a trained classifier, and triggers a manual review process for candidate mappings with confidence scores below a preset threshold.

7. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The data quality assessment system uses a weighted scoring method to comprehensively evaluate completeness, consistency, and geometric accuracy, and marks entities with scores below a preset threshold as objects to be reviewed.

8. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: Based on the evaluated and qualified data, a digital twin model is constructed, specifically including: Load the data that has passed the quality assessment into a predefined digital twin model template; Based on the engineering entity division system, the data is instantiated into corresponding line, station, tunnel and equipment unit objects; and the geometric information and attribute information of the objects are associated and bound with the time series data to generate an operation model that supports operation and maintenance visualization and real-time monitoring.

9. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The operation and maintenance task generation includes analyzing sensor data based on a predictive maintenance model and automatically generating a priority list of maintenance plans.

10. The integrated management method for the entire lifecycle of railway infrastructure digital twin model according to claim 1, characterized in that: The standardized interfaces include those based on REST or message queues.