Pipeline health assessment method, device and equipment and storage medium

Abstract

The invention provides a pipeline health assessment method and device, equipment and a storage medium, and relates to the technical field of information. The method comprises the following steps: acquiring real-time pipeline detection data, wherein the real-time pipeline detection data comprises real-time meteorological data, real-time geological data and real-time pipeline stress data; according to the real-time pipeline detection data, multiple real-time characteristics are determined, and the multiple real-time characteristics comprise the temperature change characteristic, the frozen soil expansion characteristic, the pipeline stress characteristic, the accumulated snow load characteristic and the freeze-thaw cycle characteristic. And inputting the plurality of real-time features into a pre-constructed association model to determine a quantitative association relationship among the plurality of real-time features. Wherein the correlation model is obtained by training based on historical pipeline detection data and is used for representing a coupling action relationship among the plurality of real-time features. According to the quantitative incidence relation and the real-time characteristics, a pipeline health assessment result is determined, and the pipeline health assessment result comprises a pipeline metal embrittlement index and a frozen soil displacement risk value.

Description

Technical Field

[0001] This application relates to the field of information technology, and in particular to a pipeline health assessment method, apparatus, equipment, and storage medium. Background Technology

[0002] Oil and gas pipelines, as critical infrastructure for energy transportation, are often laid in regions with complex climates such as high altitudes and permafrost. In extremely cold environments, the pipeline's metallic materials are prone to low-temperature embrittlement, and displacement caused by permafrost activity can lead to additional stress on the pipeline, seriously threatening its structural safety.

[0003] Current assessments of oil and gas pipelines often focus on the identification and early warning of single geological disaster events, such as using remote sensing or detection data to determine risks of landslides and collapses.

[0004] These methods primarily focus on the disaster event itself, making it difficult to characterize the relationship between meteorological changes, permafrost evolution, and pipeline structure, thus failing to accurately assess the overall health status of the pipeline. Summary of the Invention

[0005] This application provides a pipeline health assessment method, apparatus, equipment, and storage medium for accurately assessing the overall health status of a pipeline.

[0006] Firstly, this application provides a pipeline health assessment method. This method is applied to electronic devices. The subject executing the method can be an electronic device, a component or device applied to the electronic device (e.g., a processor, chip, or chip system), or a logic module or software capable of realizing all or part of the functions of the electronic device, including: Acquire real-time pipeline inspection data, which includes real-time meteorological data, real-time geological data, and real-time pipeline stress data; Based on real-time pipeline monitoring data, several real-time features were determined, including temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. Multiple real-time features are input into a pre-built association model to determine the quantitative association between the multiple real-time features; wherein, the association model is trained based on historical pipeline detection data and is used to characterize the coupling relationship between multiple real-time features; Based on quantitative correlations and real-time characteristics, the pipeline health assessment results are determined. The pipeline health assessment results include the pipeline metal embrittlement index and the frozen soil displacement risk value.

[0007] In the first aspect, by acquiring multi-dimensional monitoring data including meteorological, geological, and stress data, comprehensive environmental and structural information affecting pipeline safety can be captured. Key features such as temperature changes, permafrost expansion, pipeline stress, snow load, and freeze-thaw cycles can then be extracted from this data, providing specific quantitative basis for analyzing pipeline condition. By inputting multi-dimensional real-time features into a correlation model trained on historical data, the complex coupling relationships between multiple factors such as temperature, permafrost, stress, and snow can be dynamically and quantitatively analyzed, overcoming the limitations of traditional assessment methods that rely on isolated analysis of single factors. The model's coupling and extrapolation capabilities based on historical patterns enable the assessment results to more comprehensively and accurately reflect the current overall health status of the pipeline, improving the reliability of pipeline health status perception and the effectiveness of decision support in extremely complex environments, thus providing a more comprehensive reflection of the pipeline's true health status under complex conditions.

[0008] In conjunction with the first aspect, in one possible implementation, the association model is a knowledge graph model, which includes: Multiple feature nodes, each feature node corresponds one-to-one with a type of real-time feature; The associated edges connecting feature nodes, each associated edge is associated with a quantitative association rule between the two feature nodes connected.

[0009] In this implementation, a knowledge graph model is constructed. By setting multiple feature nodes that correspond one-to-one with real-time features such as temperature changes and permafrost expansion, each factor affecting pipeline health has an independent representation unit in the model, enhancing the model's ability to characterize complex factors. Associative edges are introduced between feature nodes to explicitly represent the quantitative association rules between two feature nodes, making the interactions between factors in the model explicit and structured. Through this node-edge graph structure, the knowledge graph model can clearly present the causal chain and transmission path between environmental factors and pipeline response, providing an intuitive and traceable reasoning basis for subsequent analysis of the coupling effects between features.

[0010] In conjunction with the first aspect, in one possible implementation, multiple feature nodes include: temperature change node, permafrost expansion node, pipeline stress node, snow load node, and freeze-thaw cycle node.

[0011] In this implementation, the specific types of feature nodes are clearly defined in the knowledge graph model, including temperature change nodes, permafrost expansion nodes, pipeline stress nodes, snow load nodes, and freeze-thaw cycle nodes. These nodes correspond to key physical factors affecting pipeline safety. Temperature change nodes characterize the dynamic evolution of ambient temperature, permafrost expansion nodes focus on the state of permafrost volume changes, pipeline stress nodes directly reflect the stress state of the pipeline structure, snow load nodes describe the weight and distribution of surface snow, and freeze-thaw cycle nodes depict the process of repeated freezing and thawing of permafrost. By setting these targeted nodes in the model, the complex geological, meteorological, and structural response elements in extremely cold environments can be modularly expressed, making the knowledge graph's characterization of the pipeline's environment more accurate.

[0012] In conjunction with the first aspect, in one possible implementation, the associated edges include: Connect the temperature change node and the permafrost expansion node with the first associated edge; The second associated edge connecting the permafrost expansion node and the snow load node; The third associated edge connecting the snow load node and the freeze-thaw cycle node; The fourth associated edge connects the freeze-thaw cycle node and the pipeline stress node.

[0013] In this implementation, the types of edges connecting each feature node are further defined in the knowledge graph model. The first edge connects the temperature change node and the permafrost expansion node, used to quantify the driving effect of temperature fluctuations on permafrost volume expansion; the second edge connects the permafrost expansion node and the snow load node, used to express the impact of surface morphology changes after permafrost expansion on snow distribution and load; the third edge connects the snow load node and the freeze-thaw cycle node, revealing the mechanism by which snow regulates the rate and intensity of the freeze-thaw process; and the fourth edge connects the freeze-thaw cycle node and the pipeline stress node, used to quantitatively describe the stress effect of soil displacement on the pipeline during repeated freezing and thawing of permafrost. Through these edge connections, the knowledge graph model fully constructs the complete transmission path from temperature change to permafrost response, then to snow regulation, and finally to the influence on pipeline stress.

[0014] In conjunction with the first aspect, in one possible implementation, the pipeline health assessment results are determined based on quantitative correlations and real-time characteristics, including: Based on the characteristics of temperature change, frozen soil expansion, snow load, freeze-thaw cycle, and pipe and pipeline stress, and combined with the association rules represented by the first, second, third, and fourth associated edges, the frozen soil expansion rate and soil displacement are determined. The embrittlement index of the pipeline metal is determined based on the temperature change characteristics; The risk value of frozen soil displacement is determined based on the amount of soil displacement and the rate of frozen soil expansion.

[0015] In this implementation, when determining the pipeline health assessment results, multiple real-time features such as temperature changes, permafrost expansion, snow load, freeze-thaw cycles, and pipeline stress are first utilized. This is combined with the association rules carried by the first, second, third, and fourth associated edges to perform reasoning, enabling accurate calculation of the permafrost expansion rate and soil displacement. The permafrost expansion rate directly reflects the intensity of permafrost activity, while the soil displacement directly reflects the actual displacement impact of permafrost changes on the pipeline. Furthermore, based on the temperature change characteristics, a pipeline metal embrittlement index is calculated. This index quantifies the risk of brittle failure of the pipeline under the combined effects of low-temperature environment and material properties, allowing for an independent and accurate assessment of material degradation risk. Simultaneously, based on the obtained soil displacement, and combined with the permafrost expansion rate, a permafrost displacement risk value is calculated. This value comprehensively reflects the threat level posed to pipeline safety by the magnitude of permafrost displacement, pipeline burial conditions, and surrounding soil properties.

[0016] In conjunction with the first aspect, one possible implementation method for acquiring real-time pipeline monitoring data includes: Real-time meteorological data is generated based on weather information and real-time temperature monitoring data of the pipeline; Real-time geological data is generated based on surface temperature information, geological exploration disaster report information, and geological permafrost detection information. Real-time pipeline stress data is generated based on real-time pipeline stress detection information.

[0017] In this implementation, when acquiring pipeline inspection data, real-time meteorological data is generated by fusing weather information with real-time temperature monitoring information along the pipeline route. This ensures that the meteorological data possesses both regional macroscopic scope and precise local temperature measurements, improving the data's environmental adaptability. In generating real-time geological data, surface temperature information, geological exploration disaster reports, and frozen soil monitoring information are integrated. This combines data from three different sources—remote sensing observations, manual reports, and professional inspections—allowing the geological data to corroborate and complement each other, effectively enhancing the completeness and reliability of the information. For real-time pipeline stress data, it is directly generated based on the pipeline's own real-time stress monitoring information, ensuring the originality and accuracy of the structural response data. Through this categorized and fused acquisition method, this approach provides a high-quality, multi-dimensional data foundation for subsequent analysis.

[0018] In conjunction with the first aspect, in one possible implementation, the method further includes: The pipeline health index is calculated based on the pipeline metal embrittlement index and the frozen soil displacement risk value. The pipeline health index is compared with a preset threshold to generate a first identifier, which is used to characterize the risk warning level based on the pipeline health index assessment.

[0019] In this implementation, after obtaining the pipeline metal embrittlement index and the permafrost displacement risk value, this method further integrates the two to calculate a comprehensive pipeline health index. The pipeline metal embrittlement index quantifies the degree of degradation of the pipeline material itself under low-temperature conditions, while the permafrost displacement risk value reflects the displacement threat level posed to the pipeline by external permafrost activity. By integrating these two core indicators reflecting risks from different dimensions, the pipeline health index can comprehensively reflect the superimposed effect of material performance degradation and geological environment effects, avoiding the one-sidedness of single-indicator assessment. The calculated pipeline health index is compared with a preset threshold. Through this quantitative comparative analysis, the deviation of the current pipeline health status from the preset safety boundary can be objectively judged, providing a unified measurement standard for risk level classification. The final generated first identifier is used to characterize the risk warning level based on the pipeline health index assessment, transforming the complex numerical assessment results into intuitive risk level information. This allows pipeline operation and maintenance personnel to quickly grasp the overall risk status of the pipeline, providing a clear and operable basis for subsequent monitoring, maintenance, or emergency response decisions.

[0020] Secondly, this application provides a method for constructing an association model for pipeline health assessment, including: Acquire historical pipeline inspection data, which includes historical meteorological data, historical geological data, and historical pipeline stress data; Based on historical pipeline inspection data, multiple historical features are extracted. These features include at least temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. Building a knowledge graph model includes: Create multiple feature nodes, each corresponding to a type of historical feature; Based on multiple historical features, the coupling relationship between features is analyzed, and the associated edges connecting feature nodes are generated, where each associated edge represents the quantitative association rule between the two feature nodes it connects. Based on multiple feature nodes and associated edges, a pre-built association model is formed, which is used to characterize the coupling relationship between multiple historical features.

[0021] In the second aspect, acquiring historical pipeline inspection data, including meteorological, geological, and stress data, provides comprehensive and realistic foundational information for modeling. Several key historical features, such as temperature changes, permafrost expansion, pipeline stress, snow load, and freeze-thaw cycles, are extracted from this historical data, ensuring that subsequent models can focus on the core elements affecting pipeline health. When constructing the knowledge graph model, feature nodes are created that correspond one-to-one with each historical feature, giving each environmental or structural element an independent representation unit within the model. Through in-depth analysis of the coupling relationships between historical features, association edges connecting feature nodes are generated. Each association edge explicitly quantifies the quantitative association rules between the two connected feature nodes, transforming complex multi-factor interactions into explicit structured knowledge. The resulting association model fully characterizes the coupling relationships between multiple historical features, providing an interpretable and mechanistically supported reasoning basis for pipeline health assessment. This enables assessment methods based on this model to more accurately reflect the influence patterns between the environment and pipeline health.

[0022] Thirdly, this application provides a pipeline health assessment device, comprising: The data acquisition module is used to acquire real-time pipeline inspection data, which includes real-time meteorological data, real-time geological data, and real-time pipeline stress data. The feature determination module is used to determine multiple real-time features based on real-time pipeline inspection data. These real-time features include temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. The relationship determination module is used to input multiple real-time features into a pre-built association model to determine the quantitative association relationship between the multiple real-time features; wherein, the association model is trained based on historical pipeline detection data and is used to characterize the coupling relationship between multiple real-time features; The results determination module is used to determine the pipeline health assessment results based on quantitative correlations and real-time characteristics. The pipeline health assessment results include the pipeline metal embrittlement index and the frozen soil displacement risk value.

[0023] In conjunction with the third aspect, in one possible implementation, the association model is a knowledge graph model, which includes: Multiple feature nodes, each feature node corresponds one-to-one with a type of real-time feature; The associated edges connecting feature nodes, each associated edge is associated with a quantitative association rule between the two feature nodes connected.

[0024] In conjunction with the third aspect, in one possible implementation, multiple feature nodes include: temperature change node, permafrost expansion node, pipeline stress node, snow load node, and freeze-thaw cycle node.

[0025] In conjunction with the third aspect, in one possible implementation, the relationship determination module is also used to connect the first associated edge between the temperature change node and the permafrost expansion node; The second associated edge connecting the permafrost expansion node and the snow load node; The third associated edge connecting the snow load node and the freeze-thaw cycle node; The fourth associated edge connects the freeze-thaw cycle node and the pipeline stress node.

[0026] In conjunction with the third aspect, in one possible implementation, the result determination module is also used to determine the frozen soil expansion rate and soil displacement based on temperature change characteristics, frozen soil expansion characteristics, snow load characteristics, freeze-thaw cycle characteristics and pipeline stress characteristics, and in conjunction with the association rules characterized by the first associated edge, the second associated edge, the third associated edge and the fourth associated edge. The embrittlement index of the pipeline metal is determined based on the temperature change characteristics; The risk value of frozen soil displacement is determined based on the amount of soil displacement and the rate of frozen soil expansion.

[0027] In conjunction with the third aspect, in one possible implementation, the data acquisition module is also used to generate real-time meteorological data based on weather information and real-time temperature detection information of the pipeline; Real-time geological data is generated based on surface temperature information, geological exploration disaster report information, and geological permafrost detection information. Real-time pipeline stress data is generated based on real-time pipeline stress detection information.

[0028] In conjunction with the third aspect, in one possible implementation, the result determination module is also used to calculate the pipeline health index based on the pipeline metal embrittlement index and the frozen soil displacement risk value. The pipeline health index is compared with a preset threshold to generate a first identifier, which is used to characterize the risk warning level based on the pipeline health index assessment.

[0029] Fourthly, this application provides an apparatus for constructing an association model for pipeline health assessment, comprising: The data acquisition module is used to acquire historical pipeline inspection data, which includes historical meteorological data, historical geological data, and historical pipeline stress data. The feature extraction module is used to extract multiple historical features based on historical pipeline inspection data. These features include at least temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. The model building module is used to build knowledge graph models, including: Create multiple feature nodes, each corresponding to a type of historical feature; Based on multiple historical features, the coupling relationship between features is analyzed, and the associated edges connecting feature nodes are generated, where each associated edge represents the quantitative association rule between the two feature nodes it connects. Based on multiple feature nodes and associated edges, a pre-built association model is formed, which is used to characterize the coupling relationship between multiple historical features.

[0030] Fifthly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, causing the electronic device to implement the methods of the first or second aspect described above.

[0031] In a sixth aspect, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the methods described in the first or second aspect.

[0032] In a seventh aspect, this application provides a computer program product comprising a computer program; when the computer program is run in an electronic device, it causes the electronic device to implement the methods described in the first or second aspect above.

[0033] The beneficial effects of the third to seventh aspects mentioned above are described in the corresponding description of the first aspect and will not be repeated here. Attached Figure Description

[0034] Figure 1 This is a schematic diagram illustrating the application environment of a pipeline health assessment method provided in an embodiment of this application. Figure 2 A schematic diagram of a pipeline health assessment system architecture provided in this application embodiment; Figure 3 A schematic flowchart of a pipeline health assessment method provided in an embodiment of this application; Figure 4 A flowchart illustrating a method for constructing an association model for pipeline health assessment, provided in an embodiment of this application; Figure 5 This is a schematic diagram of the composition of a pipeline health assessment device provided in an embodiment of this application; Figure 6 A schematic diagram of the composition of the association model construction device for pipeline health assessment provided in the embodiments of this application; Figure 7 This is a schematic diagram of the composition of an electronic device provided in an embodiment of this application. Detailed Implementation

[0035] The following is a detailed description, with reference to the accompanying drawings, of a pipeline health assessment method, apparatus, equipment, and storage medium provided in this application.

[0036] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0037] The terms "first" and "second," etc., used in the specification and drawings of this application are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.

[0038] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the steps or units listed, but may optionally include other steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0039] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0040] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0041] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0042] Oil and gas pipelines, as key infrastructure for national energy transportation, are typically characterized by long routes, wide geographical spans, and long operating cycles. Many pipelines are laid in high-altitude, cold regions, seasonally frozen soil areas, or areas with complex climates. During long-term operation, pipelines must withstand the influence of multiple factors, including changes in ambient temperature, geological conditions, and external loads. Especially in extremely cold weather conditions, a sudden drop in ambient temperature can cause low-temperature embrittlement of the pipeline's metallic materials, significantly reducing its impact resistance and fracture toughness. Simultaneously, the permafrost layer undergoes expansion, contraction, and uneven displacement during alternating freezing and thawing processes. Combined with the effects of snow accumulation, this can easily lead to stress concentration, structural deformation, and even instability and failure, posing a severe challenge to pipeline operational safety.

[0043] Currently, for geological hazard early warning of oil and gas pipelines, there are methods that identify hazards by acquiring historical hazard points and geological assessment hazard points along the pipeline route, combined with multi-source remote sensing data such as optical remote sensing imagery and radar data. This can improve the completeness and accuracy of landslide or collapse hazard identification. These methods mainly focus on the identification and early warning of geological hazard events themselves, with the core being the determination of whether a certain pipeline section has the risk of landslides or collapses.

[0044] However, the aforementioned methods lack a systematic characterization of the long-term, multi-dimensional, and dynamic relationships between meteorological changes, permafrost evolution, external loads, and pipeline structural responses, making it difficult to reflect the true health status of pipelines under extreme environmental conditions in a timely and accurate manner. Specifically, this manifests in the following ways: single detection or experience-based judgment methods are insufficient to address the characteristics of sudden risks, complex evolution processes, and numerous influencing factors in extremely cold environments; the completeness and stability of the assessment cannot be guaranteed when local sensors fail, detection blind spots occur, or data is missing; and relying solely on fixed empirical thresholds for judgment can easily lead to misjudgments or omissions.

[0045] To address the aforementioned technical problems, this application provides a pipeline health assessment method, apparatus, equipment, and storage medium. The core idea is to comprehensively capture environmental and structural information affecting pipeline safety by acquiring multi-dimensional detection data, including meteorological, geological, and stress data. Key features such as temperature changes, permafrost expansion, pipeline stress, snow load, and freeze-thaw cycles are then extracted from this data, providing specific quantitative basis for analyzing pipeline condition. By inputting these multi-dimensional real-time features into a correlation model trained on historical data, the complex coupling relationships between multiple factors such as temperature, permafrost, stress, and snow can be dynamically and quantitatively analyzed, overcoming the limitations of traditional assessment methods that rely on isolated analysis of single factors. The model's coupling and extrapolation capabilities based on historical patterns enable the assessment results to more comprehensively and accurately reflect the current overall health status of the pipeline, improving the reliability of pipeline health status perception and the effectiveness of decision support in extremely complex environments, thus providing a more comprehensive reflection of the pipeline's true health status under complex conditions.

[0046] The embodiments provided in this application will now be described in detail with reference to the accompanying drawings.

[0047] The pipeline health assessment method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown. For example... Figure 1 As shown, the application environment includes: The first device, and the second device.

[0048] For example, the first device can be a terminal device or a server, and the second device can be a server.

[0049] Terminal device 100 and server 101.

[0050] The terminal device 100 includes an application 102 that supports pipeline health assessment. The client of the application 102, which supports pipeline health assessment, is used in the terminal device 100 to visualize the assessment results and warning information during the execution of the pipeline health assessment method according to this embodiment.

[0051] This client provides a user interface, which can take the form of a World Wide Web (Web) page accessed through a browser or a native application that needs to be downloaded and installed. The terminal is specifically a user equipment (UE), which includes, but is not limited to, smartphones, tablets, laptops, desktop computers, and Internet of Things (IoT) terminals. The terminal accesses the access network via a wireless air interface, possessing the capability to carry voice services, data transmission services, and multimedia services. It can also achieve direct communication between different terminals based on device-to-device (D2D) direct connection technology.

[0052] This client is used to receive data query requests from users and display pipeline health assessment results and early warning information to users.

[0053] In another example, the pipeline health assessment method provided in this application can be applied to server 101. Server 101 runs an application 102 that supports pipeline health assessment functions. This application is responsible for processing requests sent by clients and executing a pipeline health assessment method that includes acquiring real-time pipeline detection data, determining multiple real-time features, inputting the multiple real-time features into a pre-built correlation model to determine quantitative correlation relationships, and determining the pipeline health assessment result based on the quantitative correlation relationships and real-time features.

[0054] In one alternative embodiment, the terminal device 100 and the server 101 can be interconnected via a wired or wireless network.

[0055] Server 101 includes a first memory and a first processor. The first memory stores a pipeline health assessment program; the pipeline health assessment program is invoked and executed by the first processor to implement the pipeline health assessment method provided in this application. The first memory may include, but is not limited to, the following: random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), and electrically erasable programmable read-only memory (EEPROM). The first processor may consist of one or more integrated circuit chips. Optionally, the first processor may be a general-purpose processor, such as a central processing unit (CPU) or a network processor (NP). Optionally, the first processor can implement the pipeline health assessment method provided in this application by running programs or code.

[0056] Database system 103 is deployed on a dedicated server to store data related to pipeline health assessment. This database system includes a graph database and a relational database. The graph database stores pre-built association models, which are knowledge graph models containing multiple feature nodes and edges connecting them. The relational database stores historical pipeline inspection data, real-time pipeline inspection data, pipeline health assessment results, and system operation logs. Historical pipeline inspection data includes historical meteorological data, historical geological data, and historical pipeline stress data; real-time pipeline inspection data includes real-time meteorological data, real-time geological data, and real-time pipeline stress data; pipeline health assessment results include the pipeline metal embrittlement index and permafrost displacement risk value. Database system 103 can respond to events such as updates to real-time inspection data, adjustments to model parameters, or the generation of assessment results by synchronously updating or invalidating relevant data in the cache.

[0057] This application embodiment also provides a pipeline health assessment system, which can be installed in... Figure 1 In the application environment shown, such as Figure 2 As shown, the pipeline health assessment system 200 may include: The pipeline health assessment system front-end 201 is used to: receive user operation requests and display pipeline health assessment results, early warning information, and real-time monitoring data through a visual interface. The front-end supports users in querying historical assessment records, viewing pipeline health trend charts, and configuring system parameters.

[0058] The pipeline health assessment system backend 202 is used to execute the core logic of the pipeline health assessment method, including data acquisition, feature extraction, model inference, and health assessment. Specifically, the pipeline health assessment system backend 202 includes the following engines: The Data Acquisition Engine 2021 is used to acquire real-time pipeline inspection data, including real-time meteorological data, real-time geological data, and real-time pipeline stress data. This engine interfaces with various sensor data interfaces, meteorological service platforms, and geological monitoring systems, supporting real-time acquisition and preprocessing of multi-source data.

[0059] The Feature Extraction Engine 2022 identifies multiple real-time features based on real-time pipeline inspection data. These features include temperature change characteristics, permafrost expansion characteristics, pipeline stress characteristics, snow load characteristics, and freeze-thaw cycle characteristics. The engine incorporates feature engineering algorithms to extract key indicators characterizing pipeline health from raw data.

[0060] The Model Association Engine 2023 is used to store pre-built association models and input multiple real-time features into these models to determine the quantitative association relationships between them. The association models are trained based on historical pipeline detection data and are used to characterize the coupling relationships between multiple real-time features. Optionally, the association model is a knowledge graph model containing multiple feature nodes and association edges connecting these nodes. Each association edge represents the quantitative association rule between the two feature nodes it connects.

[0061] The Health Assessment Engine 2024 is used to determine pipeline health assessment results based on quantitative correlations and real-time characteristics. The pipeline health assessment results include the pipeline metal embrittlement index and the frozen soil displacement risk value. Based on temperature change characteristics, frozen soil expansion characteristics, snow load characteristics, freeze-thaw cycle characteristics, and pipeline stress characteristics, and combined with the correlation rules represented by the first, second, third, and fourth correlation edges, the frozen soil expansion rate and soil displacement are determined; the pipeline metal embrittlement index is determined based on temperature change characteristics; and the frozen soil displacement risk value is determined based on the soil displacement and the frozen soil expansion rate.

[0062] The Information Early Warning Engine 2025 is used to generate corresponding early warnings or safety indicators based on pipeline health assessment results. It calculates the pipeline health index based on the pipeline metal embrittlement index and frozen soil displacement risk value; compares the pipeline health index with a preset threshold to generate a first indicator, which characterizes the risk warning level derived from the pipeline health index assessment.

[0063] The model training engine 2026 is used to train or update association models based on historical pipeline detection data. This engine periodically reads historical data from database system 203, performs feature analysis and coupling relationship mining, generates new association rules, or optimizes existing model parameters, ensuring the accuracy and timeliness of the model.

[0064] Database system 203 is used to store data related to pipeline health assessment. Database system 203 includes a graph database and a relational database. The graph database stores pre-built association models, containing multiple feature nodes and associated edges connecting these feature nodes. The relational database stores historical pipeline inspection data, real-time pipeline inspection data, pipeline health assessment results, early warning records, and system operation logs. Historical pipeline inspection data includes historical meteorological data, historical geological data, and historical pipeline stress data; real-time pipeline inspection data includes real-time meteorological data, real-time geological data, and real-time pipeline stress data. Database system 203 can respond to events such as updates to real-time inspection data, adjustments to model parameters, or the generation of assessment results by synchronously updating or invalidating relevant data in the cache.

[0065] It should be noted that the system architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of system architecture, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0066] See Figure 3 This is a schematic diagram of a pipeline health assessment method provided in an embodiment of this application. Figure 3 As shown, the pipeline health assessment method provided in this application can be implemented through the aforementioned server, specifically including the following steps S300~S303.

[0067] S300, the server acquires real-time pipeline detection data.

[0068] Real-time pipeline monitoring data includes real-time meteorological data, real-time geological data, and real-time pipeline stress data.

[0069] The server acquires real-time pipeline inspection data. Real-time meteorological data reflects the meteorological conditions of the pipeline's surrounding environment, including ambient temperature, temperature changes, precipitation, and snow accumulation. Real-time geological data describes the geological conditions along the pipeline route, covering permafrost conditions, soil displacement, and geological structural changes. Real-time pipeline stress data directly characterizes the mechanical state of the pipeline, including axial stress and circumferential stress. These three types of data describe the pipeline's operating environment from environmental, geological, and structural dimensions, respectively. The server connects to various inspection devices through communication interfaces, continuously receiving and integrating this data to provide data support for subsequent feature extraction and correlation analysis.

[0070] S301. The server determines multiple real-time features based on real-time pipeline detection data.

[0071] Multiple real-time features include temperature change features, permafrost swelling features, pipeline stress features, snow load features, and freeze-thaw cycle features. The server extracts features from real-time pipeline monitoring data to generate these features. Temperature change features describe the patterns of temperature change, including parameters such as the rate of temperature change, the amplitude of temperature fluctuations, and the duration of low temperatures. Permafrost swelling features reflect the characteristics of permafrost volume changes, including parameters such as the permafrost swelling coefficient and the directionality of frost heave. These features are calculated by analyzing the correlation between permafrost depth, soil temperature, and displacement in real-time geological data, combined with soil type information. Pipeline stress features characterize the stress state of the pipeline, including axial stress, circumferential stress, shear stress, and their changing trends. These features are obtained by filtering and feature extraction from real-time pipeline stress data. Snow load features describe the impact of snow accumulation on the pipeline, including parameters such as snow thickness, snow density, and snow distribution uniformity. These features are extracted from surface information in real-time geological data and satellite remote sensing analysis results. Freeze-thaw cycle characteristics reflect the dynamic response of permafrost to temperature changes, including parameters such as freeze-thaw cycle length, freeze-thaw interface movement rate, and water migration during the freeze-thaw process. These characteristics are obtained by comprehensively analyzing temperature change characteristics and permafrost expansion characteristics, identifying the number of times the temperature crosses the freezing point and the corresponding permafrost state change cycles. These five characteristics characterize the critical states of the pipeline system from three dimensions: environmental, geological, and structural. There is a causal relationship between them: temperature changes drive permafrost expansion, snow load affects the freeze-thaw cycle process, and the freeze-thaw cycle further leads to permafrost expansion and pipeline stress changes.

[0072] S302. The server inputs multiple real-time features into a pre-built association model to determine the quantitative association between the multiple real-time features.

[0073] The server uses multiple extracted real-time features as input data and imports them into a pre-built correlation model. This correlation model, trained on historical pipeline inspection data, characterizes the coupling relationships between multiple real-time features. It is a computational model trained on a large amount of historical pipeline inspection data, encompassing temperature variation characteristics, permafrost expansion characteristics, pipeline stress characteristics, snow load characteristics, and freeze-thaw cycle characteristics under various environmental, geological, and operational conditions, along with actual observation results. During training, the model characterizes the intrinsic connections between these features, uncovering complex relationships such as how temperature changes affect permafrost expansion, how permafrost expansion interacts with snow load, and how freeze-thaw cycles drive pipeline stress changes. The server runs the correlation model to calculate the current real-time features, outputting quantitative correlation relationships. These relationships are presented in the form of correlation strength coefficients, influence weights, and change rate mapping tables, accurately describing the numerical dependencies between features under current environmental conditions. Through these quantitative correlation relationships, the server can extrapolate changes in a single feature to the entire feature chain even when real-time data is incomplete. For example, by using temperature change data, we can infer how much the frozen soil expansion will increase and how much the frozen soil expansion will cause the pipeline stress to rise, providing a scientific basis for the calculation of subsequent pipeline health assessment results, and upgrading the assessment process from judging isolated indicators to multi-factor coupled analysis.

[0074] S303. The server determines the pipeline health assessment results based on quantitative correlation and real-time characteristics.

[0075] The pipeline health assessment results include the pipeline metal embrittlement index and the permafrost displacement risk value. The pipeline metal embrittlement index is a quantitative indicator evaluating the degree of embrittlement risk of pipeline materials in low-temperature environments, reflecting the impact of ambient temperature on the mechanical properties of the pipeline material. The permafrost displacement risk value is an indicator evaluating the degree of displacement risk caused by permafrost expansion, reflecting the threat of permafrost activity to the structural stability of the pipeline. The server calculates the deviation of the pipeline metal material from its brittle transition temperature at the current ambient temperature based on quantitative correlations and maps this deviation to the pipeline metal embrittlement index. Simultaneously, the server calculates the amount of soil displacement caused by permafrost expansion and its impact on the pipeline based on quantitative correlations and maps this impact to the permafrost displacement risk value. By generating quantitative assessment indicators from both material properties and geological processes, the pipeline health assessment results comprehensively depict the health status of the pipeline in extremely cold environments.

[0076] In this embodiment, by acquiring multi-dimensional detection data including meteorological, geological, and stress data, comprehensive environmental and structural information affecting pipeline safety can be captured. Key features such as temperature changes, permafrost expansion, pipeline stress, snow load, and freeze-thaw cycles are then extracted from this data, providing specific quantitative basis for analyzing pipeline condition. By inputting multi-dimensional real-time features into a correlation model trained on historical data, the complex coupling relationships between multiple factors such as temperature, permafrost, stress, and snow can be dynamically and quantitatively analyzed, overcoming the limitations of traditional assessment methods that rely on isolated analysis of single factors. The model's coupling and extrapolation capabilities based on historical patterns enable the assessment results to more comprehensively and accurately reflect the current overall health status of the pipeline, improving the reliability of pipeline health status perception and the effectiveness of decision support in extremely complex environments, and allowing the assessment results to more comprehensively reflect the true health status of the pipeline in complex environments.

[0077] In one embodiment, the association model is a knowledge graph model, which includes: Multiple feature nodes, each corresponding one-to-one with a type of real-time feature. Edges connect the feature nodes, each edge representing a quantitative association rule between the two connected feature nodes. The knowledge graph model uses a graph structure to formally represent multiple real-time features and their relationships. The knowledge graph model contains two basic elements: feature nodes and edges. Feature nodes are the basic units in the knowledge graph; each feature node corresponds to a specific type of real-time feature, and internally stores information such as the feature's current value, historical trends, and attribute descriptions. Edges are directed or undirected lines connecting two feature nodes. Each edge represents a definite quantitative association rule between the two connected feature nodes. The association rule is defined in the form of a mathematical function, parameter matrix, or conditional probability, describing the numerical relationship between one feature node and its impact on another feature node when the latter changes.

[0078] In one possible implementation, multiple feature nodes include: temperature change node, permafrost expansion node, pipeline stress node, snow load node, and freeze-thaw cycle node.

[0079] Temperature change nodes correspond to temperature change characteristics, storing real-time values, rates of change, and historical extreme values ​​of ambient temperature, reflecting the impact of thermal conditions on the pipeline system. Frozen soil expansion nodes correspond to frozen soil expansion characteristics, storing data such as the amount, rate, and direction of expansion, characterizing the volumetric changes of the geological body. Pipeline stress nodes correspond to pipeline stress characteristics, storing data such as axial stress, circumferential stress, and rate of change of stress, characterizing the stress response of the pipeline structure. Snow load nodes correspond to snow load characteristics, storing data such as snow thickness, snow density, and load distribution, describing the effects of surface loads on frozen soil and pipelines. Freeze-thaw cycle nodes correspond to freeze-thaw cycle characteristics, storing data such as freeze-thaw cycles, freeze-thaw interface locations, and meltwater volume, reflecting the dynamic evolution of frozen soil.

[0080] In one possible implementation, the associated edges include: The first associated edge connects the temperature change node and the permafrost swelling node. This first associated edge represents the quantitative correlation rule between temperature change and permafrost swelling, describing how factors such as temperature drop and duration of low temperature affect the permafrost swelling coefficient and swelling rate.

[0081] A second association edge connects the permafrost swelling node and the snow load node. This second association edge connects the permafrost swelling node and the snow load node, characterizing the quantitative association rule between permafrost swelling and snow load, and describing the load effect of snow thickness and density on permafrost swelling.

[0082] The third association edge connects the snow load node and the freeze-thaw cycle node. This third association edge represents the quantitative association rule between snow load and freeze-thaw cycle, and describes the influence mechanism of snow melting on the movement of the freeze-thaw interface and meltwater infiltration.

[0083] The fourth correlation edge connects the freeze-thaw cycle node and the pipeline stress node. This fourth correlation edge represents the quantitative correlation rule between the freeze-thaw cycle and pipeline stress, describing how factors such as changes in the freeze-thaw interface location and freeze-thaw rate lead to increases and fluctuations in pipeline stress.

[0084] For example, taking a section of an oil and gas pipeline as an example, features were first extracted from real-time geological data. It was found that the ambient temperature of a certain section of the pipeline plummeted from -15℃ to -32℃ within 24 hours, with a temperature drop rate of 0.7℃ / hour. Frozen soil monitoring data showed that the frozen soil depth in this area increased by 22cm within 48 hours, with a frost heave rate of 4.6mm / hour. Pipeline stress monitoring recorded an increase in axial stress of 18.5MPa. Satellite remote sensing analysis showed that the snow thickness in this area reached 25cm, with a density of 0.35g / cm³. Freeze-thaw cycle analysis indicated that the freeze-thaw interface was moving towards the pipeline location at a rate of 3-4cm per day. Based on these features, corresponding feature nodes were constructed. The temperature change node stores the specific value and rate of change of the sudden drop in ambient temperature; the frozen soil expansion node stores the increase in frozen soil depth and the expansion rate; the pipeline stress node stores the specific value of the increase in axial stress; the snow load node stores the snow thickness and density; and the freeze-thaw cycle node stores the movement rate and location information of the freeze-thaw interface. In the association rule generation stage, a strong correlation (correlation coefficient 0.87) was found between temperature decrease and permafrost expansion in this region. A first association edge connecting the temperature change node and the permafrost expansion node was generated. The quantitative association rule contained in this first association edge indicates that in such high-altitude and cold regions, permafrost expansion increases by approximately 0.65% for every 1°C decrease in temperature. Analysis revealed that while 25cm of snow provides insulation and slows the permafrost expansion rate by approximately 18%, its weight increases the ground load. A second association edge connecting the permafrost expansion node and the snow load node was generated. This second association edge contains quantitative... The association rules quantified the balance between snow thickness and permafrost expansion. It was also found that the snow was melting rapidly, and meltwater seeping into the soil would exacerbate the subsequent freezing process, generating a third association edge connecting the snow load node and the freeze-thaw cycle node. The quantitative association rules contained in this third association edge predicted that the freeze-thaw cycle intensity would increase by 30%. Analysis showed that the freeze-thaw interface was only 8 cm away from the pipeline, and it was expected to affect the pipeline within 48 hours, generating a fourth association edge connecting the freeze-thaw cycle node and the pipeline stress node. The quantitative association rules contained in this fourth association edge predicted that the pipeline stress would further increase by 25-30 MPa.

[0085] In one possible implementation, step S303 includes: S3031. The server determines the frozen soil expansion rate and soil displacement based on the characteristics of temperature change, frozen soil expansion, snow load, freeze-thaw cycle, and pipeline stress, and in combination with the association rules represented by the first, second, third, and fourth associated edges.

[0086] The server performs multi-path reasoning calculations along the association edges in the knowledge graph. The temperature change characteristics data include the temperature time series data for the most recent 24 hours, including temperature values ​​recorded every 15 minutes, the rate of temperature change, and the predicted temperature trend. This data not only includes current measured values ​​but also integrates the results of short-term temperature prediction models. The server analyzes the driving effect of temperature changes on permafrost expansion through the first association edge, which considers the influence of multiple factors such as soil type, moisture content, and freezing rate. The current temperature change data is input into the rule model to calculate the expected change in permafrost expansion: when the temperature drops below freezing, the water in the soil freezes, causing volume expansion. Based on the magnitude and rate of temperature drop, combined with local soil characteristic parameters, the amount and direction of permafrost expansion are accurately calculated. For example, for clay soils, the permafrost expansion is approximately 0.5%-0.7% for every 1°C drop below -2°C; while for sandy soils, the expansion is only 0.3%-0.4% under the same temperature change.

[0087] The server calculates the impact of snow load on permafrost expansion based on changes in permafrost expansion via the second associated edge. When the snow thickness is less than 5cm, its insulation effect is weak, mainly manifesting as a load effect, with permafrost expansion increasing by about 3%-5%. When the snow thickness is between 5-15cm, the insulation effect begins to appear, and the overall permafrost expansion decreases slightly. When the snow thickness exceeds 15cm, the insulation effect dominates, and the permafrost expansion rate can be reduced by 15%-25%, but at the same time, the infiltration of water after snowmelt will cause more severe permafrost expansion.

[0088] Based on the impact of snow load on permafrost swelling, the interaction mechanism between snow load and freeze-thaw cycles is analyzed using a third correlation edge. Snow thickness and melting rate directly affect the depth and speed of freeze-thaw cycles, while the freeze-thaw cycle process in turn affects the snow melting pattern. When snow melts during a period of rapid warming, a large amount of meltwater seeps into the unfrozen soil layer. When the temperature drops again, this water forms larger ice crystals, leading to a stronger frost heave effect, forming a positive feedback loop of meltwater-refreezing-stronger swelling. By analyzing the current temperature change trend, snow condition, and soil moisture content, the timing and amount of snow melt are predicted, thereby assessing the intensity and scope of subsequent freeze-thaw cycles.

[0089] Based on the snow load node and freeze-thaw cycle node, the pipeline stress variation is generated using the fourth association rule. When the freeze-thaw interface is close to the pipeline but not yet in contact, the pipeline stress change is small; when the freeze-thaw interface contacts the pipeline and relative displacement occurs, the pipeline will bear the maximum stress; after the freeze-thaw process is complete and the soil stabilizes, the stress will gradually be released. Based on the intensity prediction of the freeze-thaw cycle and the direction of frozen soil expansion, the stress components of the pipeline in different directions, including axial tensile stress, circumferential stress, and shear stress, and their variation processes, are calculated.

[0090] Based on the changes in pipeline stress, and combining the data from the frozen soil expansion nodes and the pipeline stress nodes, the frozen soil expansion rate and soil displacement are calculated using a soil mechanics model.

[0091] By analyzing the spatial distribution characteristics of expansion rate and soil displacement, high-risk areas, such as stress concentration points like pipe bends and tee connections, can be identified. The spatial distribution characteristics of expansion rate and soil displacement are presented in a four-dimensional spatiotemporal format, including a predicted curve of frozen soil expansion rate for the next 24-72 hours, a soil displacement vector field distribution map, high-risk area markers, and corresponding confidence assessments.

[0092] For example, taking an oil and gas pipeline as an example, temperature change data for this pipeline section is obtained from the knowledge graph. It shows that the temperature rose from -18℃ to -5℃ in the past 24 hours, and is expected to continue rising to 2℃ in the next 12 hours, with a daily variation of 20℃, which is a typical example of drastic temperature fluctuation. Through the first correlation edge analysis, the permafrost expansion in this area is determined. Because the temperature is recovering from extremely low values, the permafrost is beginning to thaw, but the surface soil has thawed while the deeper layers remain frozen, forming a special state of "thawed on top and frozen below." It is predicted that the permafrost expansion will first increase and then decrease in the next 24 hours, with the maximum expansion occurring when the temperature reaches around -2℃. Through the second correlation edge calculation, it is found that the snow thickness in this area is 18cm, with a density of 0.32g / cm³. The insulating effect of the snow reduces the permafrost expansion rate by about 20%, but the snow is melting rapidly, and the meltwater seeping into the soil will accumulate greater expansion potential for a subsequent possible temperature drop. Overall, the impact of snow on permafrost expansion is assessed as moderate to high. The third correlation edge analysis revealed the interaction mechanism between snow load and freeze-thaw cycle: Snow is expected to melt completely within the next 36 hours, with a large amount of meltwater seeping into the partially thawed soil layer. If temperatures subsequently drop, this will lead to a stronger secondary freezing and expansion effect, forming a high-risk "meltwater-refreezing" cycle. The fourth correlation edge analysis generated a prediction of pipeline stress changes: When the freeze-thaw interface reaches the pipeline location (expected when the temperature drops to -1℃), the pipeline will bear the maximum axial stress, expected to increase by 22-28 MPa, exceeding 75% of the safety threshold for this pipeline segment, posing a high risk. Combining frozen soil expansion and pipeline stress data, the maximum frozen soil expansion rate is calculated to reach 5.2 mm / hour, with an expected vertical soil displacement of 18 mm and a horizontal displacement of 12 mm. The displacement distribution is uneven, with the pipeline forming a displacement concentration zone. The final analysis of the dynamic changes in the frozen soil showed that the pipeline section would undergo a complex freeze-thaw cycle in the next 48 hours, with a certain high-risk point. It was predicted that soil displacement would cause the pipeline stress to reach a critical value. It was recommended to take drainage measures and strengthen the monitoring of the area before the temperature drops.

[0093] In one possible implementation, the server determines the stress detection frequency adjustment value based on the permafrost expansion rate.

[0094] The rate of frozen soil expansion is converted into the rate of change of pipeline stress (in megapascals per hour). For example, for an X70 steel pipe with a diameter of 508 mm and a wall thickness of 12.7 mm, in a clay soil environment, the rate of frozen soil expansion is 3 mm / hour, and the rate of change of pipeline stress is approximately 4.2 megapascals per hour; while in sandy soil, the same rate of expansion may result in a rate of change of stress of only about 3.0 megapascals per hour. Based on the rate of change of pipeline stress, the adjustment value for the stress detection frequency is determined. The detection frequency needs to match the rate of stress change to ensure that key change points are captured without wasting resources. A baseline detection frequency is set (usually once per hour), and an adjustment factor is calculated based on the rate of change of stress. When the rate of change of stress is below 2 megapascals per hour, the detection frequency can be maintained at the baseline level or appropriately reduced; when the rate of change of stress is between 2 and 5 megapascals per hour, the detection frequency is increased to once every 30 minutes; when the rate of change of stress exceeds 5 megapascals per hour, the detection frequency is further increased to once every 15 minutes or higher.

[0095] In one possible implementation, the stress detection sensitivity adjustment value is determined based on the amount of soil displacement.

[0096] Based on the soil displacement, the stress detection sensitivity adjustment value is determined. The larger the soil displacement, the more pronounced the stress concentration phenomenon that the pipeline may experience, requiring higher detection sensitivity to detect minute but critical stress changes. A baseline sensitivity level is set, and a sensitivity adjustment coefficient is calculated based on the soil displacement. When the soil displacement is less than 5 mm, the baseline sensitivity is maintained; when the displacement is between 5 and 10 mm, the sensitivity is increased by 20%; when the displacement exceeds 10 mm, the sensitivity is increased by 40% or higher. Sensitivity adjustment mainly affects the signal processing parameters of strain gauges or fiber optic sensors, such as amplifier gain, filter settings, and data sampling accuracy. Based on the above calculation results, a pipeline stress detection parameter adjustment command is generated according to the stress detection frequency adjustment value and the stress detection sensitivity adjustment value. This command includes not only the specific adjustment value but also information such as the adjustment priority, execution time, and applicable scope. The pipeline stress detection frequency and sensitivity are adjusted according to the pipeline stress detection parameter adjustment command. Pipeline inspection supports remote parameter adjustment, sending new parameters to inspection equipment along the pipeline via a communication network to achieve real-time optimization of the inspection strategy. The adjusted pipeline stress detection frequency and sensitivity are incorporated into the extreme cold adaptability assessment parameter system. It is used not only for current detection but also as historical data for subsequent analysis and model optimization. The adjusted detection performance will be continuously evaluated, and further optimization will be carried out as necessary, forming a closed-loop control.

[0097] For example, taking an extreme low-temperature event encountered by an oil and gas pipeline as an example, the frozen soil expansion rate of this pipeline section was 4.8 mm / h, the vertical soil displacement was 12.5 mm, and the horizontal displacement was 8.2 mm, significantly higher than normal levels. Based on the frozen soil expansion rate of 4.8 mm / h, the pipeline stress change rate was calculated. Considering that this pipeline section is an X80 steel pipeline with a diameter of 711 mm, a wall thickness of 14.3 mm, a burial depth of 1.8 meters, and the soil type is silty clay, the soil-pipeline interaction model was applied, and the pipeline stress change rate was calculated to be 6.3 MPa / h, which is close to the high-risk threshold (7 MPa / h). Based on the stress change rate of 6.3 MPa / h, the stress detection frequency adjustment value was determined. Since the stress change rate exceeds the threshold of 5 MPa / h, the detection frequency was adjusted from the baseline once per hour to once every 10 minutes, i.e., the frequency was increased by 500%. Predicting that the stress change rate might further increase to 7.5 MPa / hour in the next 6 hours, an early warning mechanism was implemented, which would further increase the frequency to once every 5 minutes if the prediction came true. Based on a soil vertical displacement of 12.5 mm (exceeding the 10 mm threshold), a stress detection sensitivity adjustment value was determined to be increased by 45% to ensure the capture of small but critical stress changes. A pipeline stress detection parameter adjustment command was generated, setting the detection frequency to once every 10 minutes, increasing the overall detection sensitivity by 45%. This command was marked as "high priority" and required to be completed within 15 minutes. Upon receiving the command, the detection equipment immediately adjusted its operating parameters, shortening the data acquisition interval from 60 minutes to 10 minutes, increasing the signal amplifier gain by 45%, and adjusting the data filtering algorithm accordingly to accommodate the higher sampling frequency. The adjusted detection frequency (once every 10 minutes) and detection sensitivity (increased by 45%-55%) were recorded for subsequent pipeline health assessments.

[0098] In one possible implementation, the server determines the permafrost displacement warning threshold based on the permafrost expansion rate.

[0099] The calculation of the frozen soil displacement early warning threshold is based on a comprehensive analysis of soil mechanics models and historical monitoring data, taking into account various factors such as soil type, water content, freezing depth, and external load. A time-series prediction method is used to compare the current frozen soil swelling rate with historical swelling patterns to predict the frozen soil displacement within the next 24-72 hours. Different prediction models are applied for different soil types. For clay soils, due to their high plasticity, displacement prediction mainly considers the cumulative effect of the swelling rate; for sandy soils, the changing trend of the swelling rate is more important. An adjustment coefficient for the early warning threshold is determined based on the predicted frozen soil displacement values. This adjustment coefficient, a value between 0.7 and 1.3, is used to dynamically adjust the frozen soil displacement early warning threshold. A baseline early warning threshold is set, and the adjustment coefficient is calculated based on the predicted frozen soil displacement values. When the predicted displacement is less than 50% of the baseline threshold, the adjustment coefficient is set to 1.2 to appropriately increase the warning threshold and reduce false alarms. When the predicted displacement is between 50% and 80% of the baseline threshold, the adjustment coefficient is set to 1.0 to maintain the baseline threshold. When the predicted displacement exceeds 80% but is less than 100% of the baseline threshold, the adjustment coefficient is set to 0.85 for early warning. When the predicted displacement exceeds the baseline threshold, the adjustment coefficient is further reduced to 0.7 for significantly earlier warning. Based on the warning threshold adjustment coefficient, the dynamic adjustment value of the frozen soil displacement warning threshold is calculated. The dynamic adjustment value is the product of the baseline warning threshold and the adjustment coefficient, representing the actual warning threshold that should be used under the current environmental conditions. For example, if the baseline warning threshold is 20 mm and the adjustment coefficient is 0.85, then the dynamic adjustment value is 17 mm. An adjustment range limit is set, with a single adjustment not exceeding 30% of the baseline threshold to avoid drastic fluctuations in the threshold. The frozen soil displacement warning threshold is updated based on the dynamic adjustment value. The new threshold does not take full effect immediately but gradually transitions to the target value over 1-2 hours through linear interpolation to avoid false alarms caused by sudden threshold changes. The updated permafrost displacement early warning threshold will not only be used for current early warning judgments, but also as historical data for subsequent knowledge graph optimization and trend analysis.

[0100] S3032. The server determines the embrittlement index of the pipe metal based on the temperature change characteristics.

[0101] When calculating the pipeline metal embrittlement index in real time, pipeline stress detection data is collected based on the adjusted pipeline stress detection frequency and sensitivity. Based on the adjusted detection frequency (e.g., once every 10 minutes) and adjusted detection sensitivity (e.g., increased by 45%), the detection equipment along the pipeline, such as strain gauges and fiber optic sensors, is controlled to collect pipeline stress data according to the new parameters. These adjusted parameters ensure that key stress changes can be captured under extremely cold conditions, avoiding missed important information due to insufficient detection frequency or sensitivity. This high-precision detection data is received and processed in real time to form a continuous pipeline stress time series. Based on the ambient temperature value and the real-time pipeline stress detection data, the difference between the pipeline metal brittle transition temperature and the ambient temperature is calculated, and the pipeline metal embrittlement degree index is calculated based on this difference.

[0102] The brittle-brittle transition temperature (BCT) of pipeline metals is a key parameter in materials science, representing the critical temperature at which pipeline steel transitions from a ductile to a brittle state. This value is determined by the pipeline material and can be obtained through material testing. The BCT of the current pipe section (e.g., -15℃) is retrieved from a pipeline material database and compared with the current ambient temperature value provided by meteorological monitoring (e.g., -25℃), calculating the difference (-10℃ in this example). When the ambient temperature is lower than the BCT, the difference is negative; a larger absolute value indicates a higher degree of embrittlement. An empirical formula is used to convert the temperature difference into a embrittlement index between 0 and 1. For example, for every 0.1℃ decrease in the difference, the embrittlement index increases by 0.05, but the upper limit is set to 1.0. Based on the pipeline embrittlement index and pipeline material parameters, an initial pipeline embrittlement index is calculated.

[0103] Pipeline material parameters include steel grade (e.g., X70, X80), chemical composition, heat treatment state, and manufacturing process. These parameters affect the material's actual performance at low temperatures. A material influence factor model is established, converting different material parameters into correction coefficients between 0.8 and 1.2. For example, for X80 steel that has undergone special low-temperature treatment, the correction coefficient might be 0.9; while for ordinary X70 steel, the correction coefficient might be 1.1. The initial pipeline metal embrittlement index is obtained by multiplying the embrittlement degree index by the material correction coefficient. This index reflects the actual embrittlement state of the pipeline metal under current environmental conditions, and its value range is typically 0-1.2. The initial pipeline metal embrittlement index is then normalized to generate a standardized pipeline metal embrittlement index. Normalization is performed to ensure the comparability of evaluation results for different pipe sections and materials; a minimum-maximum normalization method is used to map the initial index to a standard range of 0-1. For example, a safe threshold for the embrittlement index is set at 0.7, and a dangerous threshold at 1.0. A linear transformation maps the initial index (0-1.0) to a standardized index (0-1), where 0.7 corresponds to a standardized value of 0.7, and 1.0 corresponds to a standardized value of 1.0. For extreme cases exceeding 1.0, the standardized index is fixed at 1.0. The standardized pipeline metal embrittlement index can be directly used for health status assessment and early warning level classification; the closer the value is to 1, the higher the risk of pipeline metal embrittlement.

[0104] S3033. The server determines the permafrost displacement risk value based on the soil displacement and permafrost expansion rate.

[0105] Based on the initial frozen soil displacement risk value, normalization processing is performed to generate standardized frozen soil displacement risk values.

[0106] When calculating the risk value of frozen soil displacement in real time, frozen soil displacement detection data is collected based on the adjusted pipeline stress detection frequency and sensitivity. The adjusted parameters are used to control frozen soil displacement detection equipment (such as ground displacement sensors, InSAR detection systems, and pipeline strain detection devices) to collect high-precision frozen soil displacement data. Based on the frozen soil displacement detection data, the actual frozen soil displacement is calculated. This calculation process includes steps such as data filtering, outlier removal, and multi-source data fusion, ultimately obtaining the vertical and horizontal displacement values, in millimeters, reflecting the actual movement of the frozen soil.

[0107] Actual frozen soil displacement is a direct indicator for assessing the impact of frozen soil on pipelines, providing fundamental data for subsequent risk assessment. Based on the actual frozen soil displacement and the updated frozen soil displacement warning threshold in the displacement warning sub-parameters, the proportion of displacement exceeding the standard is calculated.

[0108] The displacement early warning sub-parameter is a dynamically adjusted warning threshold based on previous analysis of frozen soil dynamic changes (e.g., adjusted from the baseline of 20 mm to 15 mm). The actual frozen soil displacement is compared with the current effective warning threshold to calculate the displacement exceedance ratio, using the formula: Displacement exceedance ratio = (Actual frozen soil displacement - Warning threshold) / Warning threshold × 100%. When the actual displacement is less than the warning threshold, the displacement exceedance ratio is negative, indicating a safety margin; when the actual displacement exceeds the warning threshold, the displacement exceedance ratio is positive, with a higher value indicating a higher risk. The displacement exceedance ratio directly reflects the degree of deviation of the current frozen soil displacement from the safety threshold and is a key intermediate parameter for risk assessment.

[0109] Based on the proportion of displacement exceeding the standard, combined with pipeline burial depth and soil type parameters, the frozen soil displacement risk coefficient is calculated. Pipeline burial depth represents the vertical distance (in meters) from the pipeline centerline to the ground, and soil type describes the physical properties of the soil surrounding the pipeline (e.g., sand, clay, or silt). A risk coefficient calculation model is established. For shallower pipelines, the same proportion of displacement exceeding the standard leads to a higher risk; therefore, a burial depth correction factor is applied (the correction factor increases by 0.1 for every 0.5-meter decrease in burial depth). For different soil types, the frozen soil expansion characteristics differ, so a soil type correction factor is applied (1.0 for sand, 1.2 for silt, and 1.5 for clay). The frozen soil displacement risk coefficient = proportion of displacement exceeding the standard × burial depth correction factor × soil type correction factor. This risk coefficient comprehensively considers the degree of displacement exceeding the standard and the influence of environmental factors, more accurately reflecting the actual risk level. Based on the frozen soil displacement risk coefficient, the initial frozen soil displacement risk value is calculated. The initial frozen soil displacement risk value is obtained by nonlinearly transforming the risk coefficient. An S-shaped function maps the risk coefficient to a range of 0-1.5. This function increases slowly when the risk coefficient is low and accelerates as it approaches the danger threshold, better reflecting the actual evolution of risk. For example, when the risk coefficient is 0.2, the initial risk value is approximately 0.15; when the risk coefficient is 0.5, the initial risk value is approximately 0.4; and when the risk coefficient is 1.0, the initial risk value is approximately 0.85. This nonlinear transformation highlights high-risk situations, making the assessment results more sensitive and practical. Based on the initial frozen soil displacement risk value, normalization is performed to generate standardized frozen soil displacement risk values. The normalization process uses a linear transformation method to map the initial risk value (range 0-1.5) to a standard 0-1 range, where 0.7 is set as the warning threshold and 1.0 as the danger threshold. For example, when the initial risk value is ≤0.7, the standardized risk value = the initial risk value; when 0.7 < initial risk value ≤1.0, the standardized risk value = 0.7 + (initial risk value - 0.7) × (1.0 - 0.7) / (1.0 - 0.7) = initial risk value; when the initial risk value >1.0, the standardized risk value = 1.0. The standardized frozen soil displacement risk value can be directly used for health status assessment, and together with the pipeline metal embrittlement index, it constitutes a complete pipeline health assessment result. The closer the value is to 1, the higher the frozen soil displacement risk.

[0110] In this embodiment, a knowledge graph model is constructed. By setting multiple feature nodes that correspond one-to-one with real-time features such as temperature changes and permafrost expansion, each factor affecting pipeline health has an independent expression unit in the model, enhancing the model's ability to characterize complex factors. Associative edges are introduced between feature nodes to explicitly represent the quantitative association rules between two feature nodes, making the interaction relationships between factors in the model explicit and structured. Through this node-edge graph structure, the knowledge graph model can clearly present the causal chain and transmission path between environmental factors and pipeline response, providing an intuitive and traceable reasoning basis for subsequent analysis of the coupling effects between features.

[0111] The knowledge graph model clarifies the specific types of feature nodes, including temperature change nodes, permafrost expansion nodes, pipeline stress nodes, snow load nodes, and freeze-thaw cycle nodes. These nodes correspond to key physical factors affecting pipeline safety. Temperature change nodes characterize the dynamic evolution of ambient temperature, permafrost expansion nodes focus on the state of permafrost volume changes, pipeline stress nodes directly reflect the stress state of the pipeline structure, snow load nodes describe the weight and distribution of surface snow, and freeze-thaw cycle nodes depict the process of repeated freezing and thawing of permafrost. By setting these targeted nodes in the model, the complex geological, meteorological, and structural response elements in extremely cold environments can be modularly expressed, making the knowledge graph's depiction of the pipeline's environment more accurate.

[0112] The knowledge graph model further defines the types of edges connecting each feature node. The first edge connects the temperature change node and the permafrost expansion node, quantifying the driving effect of temperature fluctuations on permafrost volume expansion. The second edge connects the permafrost expansion node and the snow load node, expressing the impact of surface morphology changes after permafrost expansion on snow distribution and load. The third edge connects the snow load node and the freeze-thaw cycle node, revealing the mechanism by which snow regulates the rate and intensity of the freeze-thaw process. The fourth edge connects the freeze-thaw cycle node and the pipeline stress node, quantitatively describing the stress effect of soil displacement on the pipeline during repeated freezing and thawing of permafrost. Through these edge connections, the knowledge graph model fully constructs the complete transmission path from temperature change to permafrost response, then to snow regulation, and finally to the influence on pipeline stress.

[0113] When determining the pipeline health assessment results, multiple real-time features, such as temperature changes, permafrost expansion, snow load, freeze-thaw cycles, and pipeline stress, are first utilized. These features are then combined with the association rules carried by the first, second, third, and fourth associated edges to perform reasoning, enabling the accurate calculation of the permafrost expansion rate and soil displacement. The permafrost expansion rate directly reflects the intensity of permafrost activity, while the soil displacement directly reflects the actual displacement impact of permafrost changes on the pipeline. Furthermore, based on the temperature change characteristics, the pipeline metal embrittlement index is calculated. This index quantifies the risk of brittle failure of the pipeline under the combined effects of low-temperature environment and material properties, allowing for an independent and accurate assessment of material degradation risk. Simultaneously, based on the obtained soil displacement and combined with the permafrost expansion rate, a permafrost displacement risk value is calculated. This value comprehensively reflects the threat level posed to pipeline safety by the magnitude of permafrost displacement, pipeline burial conditions, and surrounding soil properties.

[0114] In one embodiment, step S300 includes: S3001: The server generates real-time meteorological data based on weather information and real-time temperature detection information of the pipeline.

[0115] Weather information includes meteorological elements such as the type of warning, warning level, affected area, and expected duration issued by the meteorological department, reflecting the overall weather conditions in the areas through which the pipeline passes.

[0116] Meanwhile, the server collects real-time temperature detection information through a distributed temperature sensor network along the pipeline. The real-time temperature detection information is collected by high-precision temperature sensors at a fixed frequency, accurately reflecting the dynamic changes in the ambient temperature around the pipeline.

[0117] This process fuses temperature forecasts from weather information with real-time temperature monitoring data. First, both types of data undergo format standardization and quality checks to remove outliers and invalid data. Then, spatiotemporal alignment technology is used to map regional weather information to specific coordinates along the pipeline, ensuring temporal and spatial matching between the two types of data. Finally, a weighted fusion algorithm is applied, dynamically allocating weights based on the reliability and real-time performance of the data sources to generate ambient temperature. Ambient temperature, temperature change rate, and extreme temperature events constitute real-time meteorological data.

[0118] The current ambient temperature is compared with a preset temperature threshold to generate a temperature comparison result. The preset temperature threshold is typically set between -15℃ and -25℃ based on the material characteristics of different pipe sections and historical data to accommodate the specific needs of different areas and pipe sections. When the current ambient temperature is lower than the preset temperature threshold, the health status of the pipeline is assessed.

[0119] For example, taking an oil and gas pipeline as an example, the latest weather warning information obtained from the meteorological department indicates that the area through which the pipeline passes will be affected by a strong cold air mass. Simultaneously, a distributed temperature sensor network deployed along the pipeline (one detection point every 8 kilometers, totaling 120 detection points) continuously transmits real-time temperature data. These two types of data are fused. First, the meteorological warning data is analyzed to determine that the affected area covers the entire pipeline length; then, the sensor data undergoes a quality check, removing three outliers caused by malfunctions; next, spatiotemporal alignment is performed to precisely match the regional temperature predictions from the meteorological warning with the specific coordinates of the pipeline; finally, a weighted fusion algorithm is used, with the meteorological warning data weighted at 0.4 and the real-time sensor data weighted at 0.6, generating complete weather event detection data. Based on this weather event detection data, the current ambient temperature is extracted to be -23.2℃. The preset temperature threshold for this pipeline section is set to -22℃ based on material characteristics and historical data. Comparing the current ambient temperature of -23.2℃ with the preset temperature threshold of -22℃ confirms that the current ambient temperature is below the preset threshold, and the pipeline health assessment begins.

[0120] S3002: The server generates real-time geological data based on surface temperature information, geological exploration disaster report information, and geological permafrost detection information.

[0121] Surface temperature information is derived from satellite remote sensing data. After radiometric calibration and atmospheric correction, a surface temperature distribution map along the pipeline is generated, providing a macroscopic view of the surface thermal conditions. The server collects geological exploration disaster reports, which are obtained through text analysis and sentiment recognition of publicly available information on social media platforms. Each valid report includes precise location, time, phenomenon description, and severity rating, reflecting actual ground-related disasters such as frost heave and cracking.

[0122] Geological and permafrost monitoring information comes from specialized permafrost monitoring stations set up along the pipeline. These stations are equipped with multi-layer soil temperature sensors and displacement detection equipment, and regularly upload data such as permafrost depth, changes in the active layer, and soil displacement. Geological and permafrost monitoring data originates from specialized permafrost monitoring stations located every 20 kilometers along the pipeline. Each monitoring station is equipped with multi-layer soil temperature sensors and precision displacement detection equipment, capable of measuring soil temperature and displacement at different depths. After receiving the raw data regularly uploaded by the monitoring stations, rigorous quality control is implemented: abnormal readings are automatically identified and eliminated, and the data is smoothed to eliminate short-term fluctuations. The final result is specialized permafrost monitoring data containing permafrost depth, changes in the active layer, and soil displacement, providing direct evidence for assessing the impact of permafrost on the pipeline. After acquiring these three types of data, crucial spatiotemporal alignment processing is performed.

[0123] Surface temperature information, geological exploration disaster reports, and frozen soil monitoring information are spatiotemporally aligned to establish a unified time benchmark. Data with different update frequencies (satellite data every 6 hours, social media real-time data, and monitoring station data every 30 minutes) are adjusted to a unified time series using intelligent interpolation methods, ensuring that all data reflect the environmental conditions at the same moment. For coordinate matching, geographic information technology is used to accurately map the raster coordinates of satellite images, the text descriptions in social media reports, and the fixed coordinates of monitoring stations onto the pipeline centerline coordinate system, achieving precise spatial correspondence between data from different sources and generating a unified spatiotemporally aligned dataset. An intelligent fusion algorithm is used to comprehensively process the three types of data. The reliability and timeliness of each type of data are evaluated: satellite remote sensing surface temperature information provides macroscopic temperature distribution but may be affected by cloud cover; social media geological exploration disaster reports reflect real-time disaster conditions but require verification of authenticity; frozen soil monitoring data is the most accurate but has limited coverage. Based on these characteristics, appropriate weights are dynamically allocated, avoiding blind reliance on a single data source. For temperature parameters, a more accurate temperature field distribution is generated by integrating temperature information from three types of data; for permafrost conditions, the degree of permafrost expansion is assessed by combining temperature data and displacement detection results. Through this complementary fusion of multi-source data, geological information containing comprehensive environmental information is ultimately generated.

[0124] For example, consider the extreme cold wave encountered by an oil and gas pipeline: Remote sensing data collected during satellite transit was used, and after radiometric calibration and atmospheric correction, a surface temperature distribution map along the pipeline was generated, showing an average temperature of -32.5℃ and a minimum temperature of -38.2℃. 137 relevant reports were collected from social media, of which 86 were confirmed to be related to the pipeline. BERT model analysis revealed that the keyword "ground cracking" appeared in 73% of the reports, and sentiment analysis showed that 62% of the reports were at the "severe" level. Based on this, disaster analysis data was generated, marking 5 high-risk areas. Data from 12 permafrost monitoring stations along the pipeline showed an average increase in permafrost depth of 15cm, and soil displacement exceeding 5mm was detected at 3 monitoring stations. In the spatiotemporal alignment phase, satellite data (UTC 10:00), social media reports (time span 9:30-10:30), and permafrost monitoring data (most recent 9:45) were unified to the 10:00 time point using linear interpolation. For coordinate matching, descriptive locations such as "5 km east of XX town" in the social media reports were converted into precise coordinates and correlated with pipeline mileage. In the data fusion phase, the weights were calculated as follows: satellite data 0.42 (quality score 0.9, timeliness 0.95), social media data 0.28 (quality 0.8, timeliness 0.85, location relevance 0.9), and permafrost monitoring data 0.30 (quality 1.0, timeliness 0.8). The resulting geological information showed a temperature of -36.8℃ in the pipeline section K125-K130, with an increase in permafrost depth of 18 cm, indicating a high risk.

[0125] This application improves the comprehensiveness and accuracy of perception of extreme cold environments through deep fusion of multi-source heterogeneous data; satellite remote sensing data provides macroscopic temperature distribution, social media reports capture real-time disaster information, and geological monitoring data reflects the state of underground permafrost. The three complement each other to make up for the limitations of a single data source; spatiotemporal alignment and coordinate matching processing ensure the consistency of data from different sources in the spatiotemporal dimension; the dynamic weight allocation fusion algorithm can adaptively adjust according to data quality and timeliness.

[0126] S3003: The server generates real-time pipeline stress data based on the real-time stress detection information of the pipeline.

[0127] Real-time stress detection information is collected by stress detection equipment deployed along the pipeline at an adjusted detection frequency and sensitivity, reflecting the stress state of the pipeline body under the influence of the external environment. The real-time stress detection information undergoes data cleaning and filtering to remove abnormal fluctuations caused by equipment noise or external interference. The raw signals are then converted into stress values ​​expressed in engineering units through a calibration algorithm, generating real-time pipeline stress data that includes axial stress, circumferential stress, shear stress, and their changing trends.

[0128] In this embodiment, when acquiring pipeline inspection data, real-time meteorological data is generated by fusing weather information with real-time temperature monitoring information along the pipeline route. This ensures that the meteorological data possesses both regional macroscopic scope and precise local temperature measurement of the pipeline, improving the environmental adaptability of the data. In generating real-time geological data, surface temperature information, geological exploration disaster report information, and frozen soil monitoring information are integrated. This combines data from three different sources—remote sensing observation, manual reports, and professional inspections—allowing the geological data to corroborate and complement each other, effectively improving the completeness and reliability of the information. For real-time pipeline stress data, it is directly generated based on the pipeline's own real-time stress monitoring information, ensuring the originality and accuracy of the structural response data. Through this categorized and fused acquisition method, this approach provides a high-quality, multi-dimensional data foundation for subsequent analysis.

[0129] In one embodiment, the pipeline health assessment method further includes: S304. The server calculates the pipeline health index based on the pipeline metal embrittlement index and the frozen soil displacement risk value.

[0130] The server employs a weighted fusion algorithm to calculate these two indices. Pre-set weights for the metal embrittlement index and the permafrost displacement risk value, with the sum of these weights equal to 1. These weights can be dynamically adjusted based on the environmental characteristics of the pipeline's location. For example, the weight of the metal embrittlement index can be appropriately increased in extremely cold regions, while the weight of the permafrost displacement risk value can be appropriately increased in areas with active permafrost. The pipeline health index is obtained through weighted summation, and it is also normalized to a range of 0 to 1. This comprehensive index reflects the overall health risk level of the pipeline under the current environment, providing a unified quantitative basis for subsequent health status classification.

[0131] S305. The server compares the pipeline health index with a preset threshold and generates a first identifier.

[0132] The first identifier is used to characterize the risk warning level derived from the pipeline health index assessment.

[0133] The server reads preset thresholds, which include at least one critical value used to classify risk levels. These thresholds are pre-set and stored based on pipeline material characteristics, historical operating data, and industry safety standards. The server compares the pipeline health index with the preset thresholds level by level to determine the numerical range in which the pipeline health index falls. Each range corresponds to a specific risk warning level. Based on the comparison results, the server generates a first identifier, which is presented in the form of a visual icon, color code, text label, or signal code, intuitively reflecting the risk warning level corresponding to the current pipeline health status. The first identifier serves as the final output of the pipeline health assessment, displayed on the operation and maintenance monitoring platform or transmitted to mobile terminals. This helps operation and maintenance personnel quickly identify the pipeline safety status and adopt differentiated inspection strategies, maintenance plans, or emergency response measures for different risk levels, achieving graded early warning and precise control of pipeline risks.

[0134] In one possible implementation, the preset threshold includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold.

[0135] The preset thresholds include a first threshold and a second threshold. The first threshold is greater than the second threshold. The first threshold is used to delineate the boundary between danger and warning, and the second threshold is used to delineate the boundary between warning and safety.

[0136] In one possible implementation, step S305 may include: S3051. When the pipeline health index is greater than the first threshold, a hazard sign is generated.

[0137] When the server compares the pipeline health index with a preset threshold, if the index is determined to be greater than the first threshold, a hazard flag is automatically generated, clearly indicating that the pipeline is in a high-risk state and requires immediate activation of the emergency response procedure. The generation of hazard flags ensures that extreme risks can be quickly identified and responded to, effectively preventing pipeline failures due to risk accumulation.

[0138] S3052. When the pipeline health index is less than or equal to the first threshold and greater than the second threshold, an early warning sign is generated.

[0139] When the server compares the pipeline health index with preset thresholds, if it determines that the pipeline health index is between the first and second thresholds (i.e., less than or equal to the first threshold and greater than the second threshold), it automatically generates a warning indicator. This warning indicator indicates a potential risk in the pipeline, requiring increased attention and preventative measures. Generating warning indicators helps to proactively intervene before risks escalate, reducing the probability of accidents.

[0140] S3053. When the pipeline health index is less than or equal to the second threshold, a safety indicator is generated.

[0141] When the server compares the pipeline health index with a preset threshold, if it determines that the pipeline health index is less than or equal to the second threshold, it will automatically generate a safety indicator. The safety indicator is presented in the form of a green icon or regular text, indicating that the pipeline is currently in a safe operating state. Regular detection and management strategies can be maintained without additional intervention.

[0142] Based on the pipeline health assessment results, the pipeline metal embrittlement index and permafrost displacement risk value are extracted to calculate the pipeline health index. The pipeline health assessment results are a set of quantitative indicators calculated in the previous stage. The pipeline metal embrittlement index reflects the degree of material embrittlement risk of the pipeline under low-temperature conditions, with a value ranging from 0 to 1; a higher value indicates a higher embrittlement risk. The permafrost displacement risk value reflects the degree of displacement risk caused by permafrost expansion, also ranging from 0 to 1; a higher value indicates a higher displacement risk. The pipeline health index is calculated using a weighted average method, with the formula: Pipeline Health Index = α × Metal Embrittlement Index + β × Permafrost Displacement Risk Value, where α and β are weighting coefficients, satisfying α + β = 1. The weighting coefficients can be set according to the characteristics of different pipe sections. For example, in high-altitude and cold regions, α can be appropriately increased to emphasize the metal embrittlement risk; in areas with active permafrost, β can be appropriately increased to highlight the permafrost displacement risk. The pipeline health index is also normalized to the 0-1 range; the closer the value is to 1, the worse the overall health condition of the pipeline and the higher the risk. Based on the pipeline health index, it is compared with a first threshold and a second threshold to generate a first identifier. The first and second thresholds are two pre-defined key judgment points, typically set to 0.7 for the first threshold and 0.4 for the second, with the first threshold being greater than the second. The calculated pipeline health index is precisely compared with these two thresholds, and the pipeline's health status level is determined based on the comparison result. The comparison process uses precise floating-point calculations to avoid misjudgments due to rounding errors. When the pipeline health index is greater than the first threshold, a hazard indicator is generated. This indicator is usually presented in a specific format, such as "Hazard indicator: Pipeline health index = 0.75," clearly indicating that the pipeline faces a serious risk. When the pipeline health index is less than or equal to the first threshold but greater than the second threshold, a warning indicator is generated. The warning indicator indicates that the pipeline has a certain risk and requires attention and preventative measures. This indicator clearly indicates that the current health status has entered the warning zone, such as "Warning indicator: Pipeline health index = 0.55." When the pipeline health index is less than or equal to the second threshold, a safety indicator is generated. The safety indicator indicates that the pipeline is in good condition, such as "Safety indicator: Pipeline health index = 0.30."

[0143] In this embodiment, after obtaining the pipeline metal embrittlement index and the permafrost displacement risk value, the method further integrates the two to calculate a comprehensive pipeline health index. The pipeline metal embrittlement index quantifies the degree of degradation of the pipeline material itself under low-temperature conditions, while the permafrost displacement risk value reflects the displacement threat level posed to the pipeline by external permafrost activity. By integrating these two core indicators reflecting risks from different dimensions, the pipeline health index can comprehensively reflect the superimposed effect of material performance degradation and geological environment, avoiding the one-sidedness of single-indicator assessment. The calculated pipeline health index is compared with a preset threshold. Through this quantitative comparative analysis, the deviation of the current pipeline health status from the preset safety boundary can be objectively judged, providing a unified measurement standard for risk level classification. The final generated first identifier is used to characterize the risk warning level derived from the pipeline health index assessment, transforming complex numerical assessment results into intuitive risk level information. This allows pipeline maintenance personnel to quickly grasp the overall risk status of the pipeline, providing a clear and operable basis for subsequent monitoring, maintenance, or emergency response decisions.

[0144] See Figure 4 This is a flowchart illustrating a method for constructing an association model for pipeline health assessment, provided as an embodiment of this application. Figure 4 As shown, the method for constructing the correlation model for pipeline health assessment can be implemented through the aforementioned server, specifically including the following steps S400~S402.

[0145] S400, the server obtains historical pipeline detection data.

[0146] Historical pipeline inspection data includes historical meteorological data, historical geological data, and historical pipeline stress data. The server obtains historical pipeline inspection data from multiple data sources. Historical meteorological data comes from archived records of meteorological departments and historical data collected by temperature sensors along the pipeline route, including historical meteorological elements such as temperature changes, precipitation, and snow cover. Historical geological data comes from historical satellite remote sensing imagery, geological exploration disaster reports from social media, and historical archived records from permafrost monitoring stations, covering information such as surface temperature evolution, permafrost expansion events, and soil displacement records. Historical pipeline stress data comes from historical stored data of the pipeline stress monitoring system, reflecting the pipeline's stress response process under different environmental and geological conditions.

[0147] S401. The server extracts multiple historical features based on historical pipeline detection data.

[0148] Multiple historical characteristics include at least temperature variation characteristics, permafrost expansion characteristics, pipeline stress characteristics, snow load characteristics, and freeze-thaw cycle characteristics. These historical characteristics comprehensively characterize the environmental effects and structural responses of the pipeline in different historical periods, providing rich sample data for constructing a coupled action model.

[0149] S402, The server builds a knowledge graph model.

[0150] Based on multiple extracted historical features, the server constructs a knowledge graph model, organizing discrete historical features into a semantically related knowledge network, laying the foundation for subsequent relational reasoning.

[0151] In one possible implementation, step S402 includes: S4021, The server creates multiple feature nodes.

[0152] Each feature node corresponds to a type of historical feature. The server creates a corresponding feature node for each type of historical feature. Nodes are the basic units in the knowledge graph, used to store historical data and attribute information for that type of feature. Each feature node has a unique identifier and is bound to the corresponding feature type, laying the foundation for the establishment of subsequent association edges.

[0153] S4022. The server analyzes the coupling relationship between multiple historical features and generates associated edges connecting feature nodes.

[0154] Each association edge represents a quantitative association rule between the two feature nodes it connects. The server performs association analysis on historical feature data to infer the mutual influence relationships between different features. Each association edge is accompanied by and stored quantitative association rules.

[0155] S4023. The server forms a pre-built association model based on multiple feature nodes and associated edges.

[0156] The association model is used to characterize the coupling relationships between multiple historical features. The server combines all created feature nodes and generated association edges to form a complete knowledge graph structure, i.e., the pre-built association model. The server stores this model in a database for later use in evaluation. When real-time features are input, the server can perform inference calculations on this model, propagating the effects along the association edges to achieve a full-process quantitative analysis from temperature changes to pipeline stress.

[0157] In this embodiment, historical pipeline inspection data, including meteorological, geological, and stress data, is acquired to provide comprehensive and realistic foundational information for modeling. Multiple key historical features, such as temperature changes, permafrost expansion, pipeline stress, snow load, and freeze-thaw cycles, are extracted from this historical data, ensuring that subsequent models focus on the core elements affecting pipeline health. When constructing the knowledge graph model, feature nodes are created that correspond one-to-one with each historical feature, giving each environmental or structural element an independent representation unit within the model. By deeply analyzing the coupling relationships between historical features, association edges connecting feature nodes are generated. Each association edge explicitly quantifies the quantitative association rules between the two connected feature nodes, transforming complex multi-factor interactions into explicit structured knowledge. The resulting association model fully characterizes the coupling relationships between multiple historical features, providing an interpretable and mechanistic reasoning basis for pipeline health assessment. This enables assessment methods based on this model to more accurately reflect the influence patterns between the environment and pipeline health.

[0158] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention.

[0159] This application embodiment can divide the pipeline health assessment device into functional modules according to the above method example. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0160] In some embodiments, this application also provides a pipeline health assessment device. This pipeline health assessment device may include one or more functional modules for implementing the pipeline health assessment method described in the above embodiments.

[0161] For example, Figure 5 This is a schematic diagram illustrating the composition of a pipeline health assessment device provided in an embodiment of this application. Figure 5As shown, the pipeline health assessment device 500 includes: a data acquisition module 501, a feature determination module 502, a relationship determination module 503, and a result determination module 504.

[0162] The data acquisition module 501 is used to acquire real-time pipeline inspection data, which includes real-time meteorological data, real-time geological data, and real-time pipeline stress data.

[0163] The feature determination module 502 is used to determine multiple real-time features based on real-time pipeline inspection data. These real-time features include temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features.

[0164] The relationship determination module 503 is used to input multiple real-time features into a pre-built association model to determine the quantitative association relationships between the multiple real-time features. The association model is trained based on historical pipeline detection data and is used to characterize the coupling relationships between the multiple real-time features.

[0165] The result determination module 504 is used to determine the pipeline health assessment results based on quantitative correlation and real-time characteristics. The pipeline health assessment results include the pipeline metal embrittlement index and the frozen soil displacement risk value.

[0166] In one embodiment, the association model is a knowledge graph model, which includes: Multiple feature nodes, each feature node corresponds one-to-one with a type of real-time feature.

[0167] The associated edges connecting feature nodes, each associated edge is associated with a quantitative association rule between the two feature nodes connected.

[0168] In one embodiment, the multiple feature nodes include: temperature change node, permafrost expansion node, pipeline stress node, snow load node, and freeze-thaw cycle node.

[0169] In one embodiment, the relationship determination module 503 is also used to connect the temperature change node and the first associated edge of the permafrost expansion node.

[0170] The second associated edge connects the permafrost expansion node and the snow load node.

[0171] The third associated edge connects the snow load node and the freeze-thaw cycle node.

[0172] The fourth associated edge connects the freeze-thaw cycle node and the pipeline stress node.

[0173] In one embodiment, the result determination module 504 is further configured to determine the frozen soil expansion rate and soil displacement based on temperature change characteristics, frozen soil expansion characteristics, snow load characteristics, freeze-thaw cycle characteristics and pipeline stress characteristics, and in combination with the association rules characterized by the first associated edge, the second associated edge, the third associated edge and the fourth associated edge.

[0174] The embrittlement index of the pipeline metal is determined based on the temperature change characteristics; The risk value of frozen soil displacement is determined based on the amount of soil displacement and the rate of frozen soil expansion.

[0175] In one embodiment, the data acquisition module 501 is further configured to generate real-time meteorological data based on weather information and real-time temperature detection information of the pipeline.

[0176] Real-time geological data is generated based on surface temperature information, geological exploration disaster reports, and frozen soil detection information.

[0177] Real-time pipeline stress data is generated based on real-time pipeline stress detection information.

[0178] In one embodiment, the result determination module 504 is further configured to calculate a pipeline health index based on the pipeline metal embrittlement index and the frozen soil displacement risk value. The pipeline health index is compared with a preset threshold to generate a first identifier, which characterizes the risk warning level derived from the pipeline health index assessment.

[0179] Figure 6 This is a schematic diagram illustrating the composition of an association model construction device for pipeline health assessment, provided in an embodiment of this application. Figure 6 As shown, the correlation model building device 600 for pipeline health assessment includes: a data acquisition module 601, a feature extraction module 602, and a model building module 603.

[0180] The data acquisition module is used to acquire historical pipeline inspection data, which includes historical meteorological data, historical geological data, and historical pipeline stress data. The feature extraction module is used to extract multiple historical features based on historical pipeline inspection data. These features include at least temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. The model building module is used to build knowledge graph models, including: Create multiple feature nodes, each corresponding to a type of historical feature; Based on multiple historical features, the coupling relationship between features is analyzed, and the associated edges connecting feature nodes are generated, where each associated edge represents the quantitative association rule between the two feature nodes it connects. Based on multiple feature nodes and associated edges, a pre-built association model is formed, which is used to characterize the coupling relationship between multiple historical features.

[0181] When implementing the functions of the integrated modules described above in hardware, this embodiment of the invention provides a possible schematic diagram of the electronic device involved in the above embodiments. For example... Figure 7 As shown, the electronic device 700 includes: a processor 702, a communication interface 703, and a bus 704. Optionally, the electronic device 700 may also include a memory 701.

[0182] Processor 702 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 702 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 702 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0183] The communication interface 703 is used to connect to other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0184] The memory 701 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0185] In one possible implementation, the memory 701 can exist independently of the processor 702. The memory 701 can be connected to the processor 702 via a bus 704 and is used to store instructions or program code. When the processor 702 calls and executes the instructions or program code stored in the memory 701, it can implement the pipeline health assessment method provided in this embodiment of the invention.

[0186] In another possible implementation, the memory 701 can also be integrated with the processor 702.

[0187] The 704 bus can be an extended industry standard architecture (EISA) bus, etc. The 704 bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 7 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0188] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0189] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The program can be stored in the aforementioned computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be any of the foregoing embodiments or memory. The aforementioned computer-readable storage medium can also be an external storage device of the aforementioned service invocation device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the aforementioned service invocation device. Further, the aforementioned computer-readable storage medium can include both internal storage units of the aforementioned service invocation device and external storage devices. The aforementioned computer-readable storage medium is used to store the aforementioned computer program and other programs and data required by the aforementioned service invocation device. The aforementioned computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0190] This application also provides computer instructions. All or part of the processes in the above method embodiments can be executed by computer instructions to instruct related hardware (such as computers, processors, network devices, and terminals). The program can be stored in the aforementioned computer-readable storage medium.

[0191] This application also provides a computer program product that, when run on a computer, causes the above-described method embodiments to be executed.

[0192] This application also provides a chip system. The chip system may be composed of chips or may include chips and other discrete devices, without limitation. The chip system includes a processor and a transceiver. All or part of the processes in the above method embodiments can be completed by this chip system, such as the chip system being used to implement the functions performed by the network devices or terminals in the above method embodiments.

[0193] In one possible design, the chip system further includes a memory for storing program instructions and / or data. When the chip system is running, the processor executes the program instructions stored in the memory to enable the chip system to perform the functions performed by the network device or terminal in the above method embodiments.

[0194] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A pipeline health assessment method, characterized in that, include: Acquire real-time pipeline inspection data, which includes real-time meteorological data, real-time geological data, and real-time pipeline stress data; Based on the real-time pipeline detection data, multiple real-time features are determined, including temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. The multiple real-time features are input into a pre-built association model to determine the quantitative association relationship between the multiple real-time features; wherein, the association model is trained based on historical pipeline detection data and is used to characterize the coupling relationship between the multiple real-time features; Based on the quantitative correlation and the real-time characteristics, the pipeline health assessment results are determined, including the pipeline metal embrittlement index and the frozen soil displacement risk value.

2. The method according to claim 1, characterized in that, The association model is a knowledge graph model, which includes: Multiple feature nodes, each of which corresponds one-to-one with a type of real-time feature; The associated edges connecting the feature nodes, and the quantitative association rules between the two feature nodes connected by each associated edge.

3. The method according to claim 2, characterized in that, The multiple feature nodes include: temperature change node, frozen soil expansion node, pipeline stress node, snow load node, and freeze-thaw cycle node.

4. The method according to claim 3, characterized in that, The associated edges include: Connect the temperature change node and the permafrost expansion node with the first associated edge; The second associated edge connects the frozen soil expansion node and the snow load node; The third associated edge connects the snow load node and the freeze-thaw cycle node; The fourth associated edge connects the freeze-thaw cycle node and the pipeline stress node.

5. The method according to claim 4, characterized in that, The process of determining the pipeline health assessment result based on the quantitative correlation and the real-time characteristics includes: Based on the temperature change characteristics, the frozen soil expansion characteristics, the snow load characteristics, the freeze-thaw cycle characteristics, and the pipeline stress characteristics, and in combination with the association rules represented by the first, second, third, and fourth associated edges, the frozen soil expansion rate and soil displacement are determined. The embrittlement index of the pipe metal is determined based on the temperature change characteristics. The risk value of the frozen soil displacement is determined based on the soil displacement amount and the frozen soil expansion rate.

6. The method according to claim 1, characterized in that, The acquisition of real-time pipeline monitoring data includes: The real-time meteorological data is generated based on weather information and real-time temperature detection information of the pipeline; The real-time geological data is generated based on surface temperature information, geological exploration disaster report information, and geological permafrost detection information. The real-time pipeline stress data is generated based on the real-time stress detection information of the pipeline.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: The pipeline health index is calculated based on the pipeline metal embrittlement index and the frozen soil displacement risk value. The pipeline health index is compared with a preset threshold to generate a first identifier, which is used to characterize the risk warning level assessed based on the pipeline health index.

8. A method for constructing an association model for pipeline health assessment, characterized in that, include: Acquire historical pipeline inspection data, which includes historical meteorological data, historical geological data, and historical pipeline stress data; Based on the historical pipeline inspection data, multiple historical features are extracted, including at least temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. Building a knowledge graph model includes: Create multiple feature nodes, each feature node corresponding to a type of one of the historical features; Based on the aforementioned historical features, the coupling relationship between the features is analyzed, and the associated edges connecting the feature nodes are generated, wherein each associated edge represents the quantitative association rule between the two feature nodes it connects. Based on the multiple feature nodes and the associated edges, a pre-constructed association model is formed, which is used to characterize the coupling relationship between the multiple historical features.

9. A pipeline health assessment device, characterized in that, include: The data acquisition module is used to acquire real-time pipeline inspection data, which includes real-time meteorological data, real-time geological data, and real-time pipeline stress data. The feature determination module is used to determine multiple real-time features based on the real-time pipeline detection data. The multiple real-time features include temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. A relationship determination module is used to input the multiple real-time features into a pre-built association model to determine the quantitative association relationship between the multiple real-time features; wherein, the association model is trained based on historical pipeline detection data and is used to characterize the coupling relationship between the multiple real-time features; The result determination module is used to determine the pipeline health assessment result based on the quantitative correlation and the real-time characteristics. The pipeline health assessment result includes the pipeline metal embrittlement index and the frozen soil displacement risk value.

10. A device for constructing a correlation model for pipeline health assessment, characterized in that, include: The data acquisition module is used to acquire historical pipeline inspection data, which includes historical meteorological data, historical geological data, and historical pipeline stress data. The feature extraction module is used to extract multiple historical features based on the historical pipeline inspection data. The multiple historical features include at least temperature change features, permafrost expansion features, pipeline stress features, snow load features, and freeze-thaw cycle features. The model building module is used to build knowledge graph models, including: Create multiple feature nodes, each feature node corresponding to a type of one of the historical features; Based on the aforementioned historical features, the coupling relationship between the features is analyzed, and the associated edges connecting the feature nodes are generated, wherein each associated edge represents the quantitative association rule between the two feature nodes it connects. Based on the multiple feature nodes and the associated edges, a pre-constructed association model is formed, which is used to characterize the coupling relationship between the multiple historical features.

11. An electronic device, characterized in that, The device includes a processor and a memory, the processor being coupled to the memory; the memory is used to store computer instructions, which are loaded and executed by the processor to enable the computer device to implement the pipeline health assessment method as described in any one of claims 1 to 7 or the correlation model construction method for pipeline health assessment as described in claim 8.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes computer-executable instructions that, when executed on a computer, cause the computer to perform the pipeline health assessment method as described in any one of claims 1 to 7 or the correlation model construction method for pipeline health assessment as described in claim 8.

13. A computer program product, characterized in that, The computer program product includes a computer program that, when run on an electronic device, causes the electronic device to perform the pipeline health assessment method as described in any one of claims 1 to 7 or the association model construction method for pipeline health assessment as described in claim 8.

Images