Hydrogen damage grading evaluation method for hydrogen-doped pipeline based on fracture mechanics

By constructing a digital twin model and multiphysics simulation of hydrogen-doped pipelines based on fracture mechanics, and combining it with hydrogen diffusion and accumulation rate models, the problem of insufficient accuracy and predictive ability of hydrogen damage assessment in existing technologies is solved. This enables real-time monitoring and graded assessment of hydrogen damage, ensuring the safe operation of hydrogen energy infrastructure.

CN121051627BActive Publication Date: 2026-02-03HUBEI SHENGXIN SPECIAL EQUIP TESTING CO LTD +1
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202511554010.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-10-29
Publication Date
2026-02-03
Estimated Expiration
2045-10-29

AI Technical Summary

Technical Problem

Existing technologies lack in-depth consideration of the interaction mechanism between hydrogen and materials when assessing hydrogen damage in hydrogen-doped pipelines. This results in limited accuracy and predictive ability of the assessment results, making it difficult to accurately capture the dynamic evolution of hydrogen damage and affecting the safe and reliable operation of hydrogen energy infrastructure and industrial development.

Method used

Based on fracture mechanics, a digital twin model of a hydrogen-doped pipeline is constructed to perform multiphysics field simulation and stress analysis. Combined with hydrogen concentration-toughness experimental data, a hydrogen diffusion and accumulation rate model is established to predict the damage development trajectory. The evaluation model is then optimized through machine learning to achieve real-time monitoring and graded assessment.

Benefits of technology

It enables accurate assessment of hydrogen damage in hydrogen-blended pipelines, providing scientific basis and decision support, improving the accuracy and adaptability of the assessment, and ensuring the safe operation of the pipeline system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121051627B_ABST
    Figure CN121051627B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of pipeline safety, and discloses a hydrogen-doped pipeline hydrogen damage grading evaluation method based on fracture mechanics; comprising: obtaining initial pipeline state data by establishing a hydrogen-doped pipeline digital twin model; then introducing a hydrogen concentration dependent item to correct the J integral expression, and performing stress analysis based thereon; generating a material toughness attenuation curve based on the stress analysis results; obtaining hydrogen diffusion and accumulation rate parameters based on the material toughness attenuation curve; and predicting damage development trajectories in combination with real-time operation data; performing risk assessment based on a dynamic response grading standard system; and finally realizing real-time monitoring and scientific grading of hydrogen damage of hydrogen-doped pipelines. The present application realizes comprehensive dynamic evaluation of hydrogen damage of hydrogen-doped pipelines, and provides scientific decision support for safe operation of pipelines.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of pipeline safety technology, and more specifically, to a method for classifying and assessing hydrogen damage in hydrogen-doped pipelines based on fracture mechanics. Background Technology

[0002] Against the backdrop of global energy transition, hydrogen energy, as a clean energy carrier, is receiving widespread attention. Hydrogen-blended pipeline systems play a crucial role in the field of hydrogen energy transmission. With the continuous expansion of hydrogen energy utilization, pipeline transportation has become a key infrastructure for the large-scale application of hydrogen energy. However, the unique physicochemical properties of hydrogen have posed unprecedented challenges to pipeline materials. In particular, hydrogen damage mechanisms such as hydrogen embrittlement, hydrogen bubbling, and hydrogen-induced cracking seriously threaten the structural integrity and operational safety of hydrogen-blended pipelines, posing potential risks to public safety and economic development.

[0003] Traditional pipeline integrity assessments often rely on experience-based detection methods and simplified mechanical models, lacking in-depth consideration of the interaction mechanism between hydrogen and materials. This results in limited accuracy and predictive ability of the assessment results. With the increase in hydrogen doping ratio and the increasing complexity of the operating environment, traditional methods struggle to accurately capture the dynamic evolution of hydrogen damage and cannot comprehensively assess the safety status of pipeline systems. This situation not only affects the safe and reliable operation of hydrogen energy infrastructure but also hinders the healthy development of the hydrogen energy industry.

[0004] In view of this, the present invention proposes a hydrogen damage classification assessment method for hydrogen-doped pipelines based on fracture mechanics to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution:

[0006] A fracture mechanics-based method for classifying and assessing hydrogen damage in hydrogen-doped pipelines includes:

[0007] Step S1: Based on the pre-collected pipeline feature information, construct a model to obtain the corresponding hydrogen-doped pipeline digital twin model; and perform multiphysics simulation based on it to obtain the corresponding initial pipeline state dataset.

[0008] Step S2: Construct an initial set of calculation parameters based on the initial pipeline state dataset, and modify the traditional J integral expression based on it; perform stress analysis on the hydrogen-doped pipeline according to the modified J integral expression to obtain the corresponding stress distribution and intensity factor dataset;

[0009] Step S3: Construct a hydrogen concentration-toughness experimental data set based on the pipeline stress distribution and strength factor dataset; and obtain the material toughness decay curve by introducing temperature and stress state factors;

[0010] Step S4: Based on the obtained material toughness decay curve, perform crack tip analysis to obtain the corresponding geometric feature data of the plastic zone, and combine the pre-constructed trapping kinetic parameters to perform time series analysis to obtain the corresponding hydrogen diffusion and accumulation rate parameters.

[0011] Step S5: Obtain real-time operating data of the hydrogen-doped pipeline, and combine it with hydrogen diffusion distribution and accumulation rate parameters to predict the hydrogen damage evolution trend and obtain the damage development trajectory prediction results.

[0012] Step S6: Based on the damage development trajectory prediction results, establish a dynamic response damage grading standard system, construct a corresponding damage assessment model based on it, and conduct a risk assessment of the current state of the hydrogen-doped pipeline to obtain a preliminary hydrogen damage level assessment result.

[0013] Step S7: Obtain historical hydrogen damage cases and optimize the parameters of the constructed damage assessment model based on the preliminary hydrogen damage level assessment results to obtain the optimized damage assessment model.

[0014] Step S8: Based on the optimized damage assessment model, the degree of hydrogen damage in the hydrogen-doped pipeline is monitored and graded in real time to obtain the final hydrogen damage grading assessment result.

[0015] Furthermore, the process of obtaining the initial pipeline state dataset includes:

[0016] Data from multiple sources, including pipeline systems, monitoring equipment, and data acquisition devices, is collected in real time from the target hydrogen-doped pipeline environment. This collected data is then processed to obtain corresponding pipeline characteristic information. Simultaneously, an IoT sensor network is deployed to collect data on operating parameters and external environmental parameters within the target hydrogen-doped pipeline, resulting in a multi-source heterogeneous data stream. Feature extraction is performed on this multi-source heterogeneous data stream to obtain various pipeline state characteristic indicators that describe the operational characteristics of the hydrogen-doped pipeline. These indicators consist of multiple dimensions, including temperature, pressure, flow rate, and hydrogen concentration. By leveraging digital twin technology and combining it with pipeline feature information, data mining and characterization of extracted pipeline feature indicators are performed, and the results are visualized in the form of a digital twin model, forming a digital twin model of the hydrogen-doped pipeline that can reflect the current characteristics of the hydrogen-doped pipeline. In this model, different nodes represent physical entities within the hydrogen-doped pipeline, while the physical field distribution represents the physical state and intensity inside the pipeline. By applying a multiphysics coupling analysis method and simulating the interaction and influence between multiple physical fields such as temperature field, stress field, and hydrogen concentration field based on the digital twin model of the hydrogen-doped pipeline, the corresponding initial pipeline state dataset is obtained.

[0017] Furthermore, the process of obtaining the stress distribution and intensity factor dataset includes:

[0018] In this embodiment, by analyzing the hydrogen concentration distribution and geometric features within the initial pipeline state dataset, an initial calculation parameter set is extracted. This initial calculation parameter set includes multiple dimensions such as hydrogen concentration gradient, pipeline geometric dimensions, and material properties. Based on fracture mechanics theory, an energy release rate correction model considering the influence of hydrogen concentration extraction is established. Furthermore, the traditional J-integral formula is modified by introducing a hydrogen concentration dependency term, resulting in a modified J-integral expression. A finite element computational mesh is established based on the J-integral expression and discretized to construct a more complete numerical solution framework. This numerical solution framework more accurately reflects the actual fracture mechanics state of the current hydrogen-doped pipeline environment. Moreover, various nodes and elements within the corresponding mathematical solution framework are endowed with rich physical properties, forming a three-dimensional and dynamic stress distribution and intensity factor dataset.

[0019] Furthermore, the process of obtaining the material toughness degradation curve includes:

[0020] In this embodiment, the pipeline stress distribution and strength factor dataset is combined with the basic fracture toughness value of the corresponding hydrogen-doped pipeline material. An initial toughness reference point is established through global analysis. Simultaneously, the changes in material fracture toughness values ​​under different hydrogen concentrations are tracked and recorded. Through in-depth analysis of the obtained hydrogen concentration-toughness experimental data set, various statistical indicators describing toughness degradation characteristics are extracted, including toughness decay rate, critical hydrogen concentration, saturation degradation value, and other dimensions. Based on mathematical modeling and other techniques, the toughness degradation characteristics are accurately fitted and modeled, establishing a mathematical expression for fracture toughness degradation. By incorporating the nonlinear influence of hydrogen concentration and further introducing temperature and stress state influence factors to correct the toughness degradation model, a multi-parameter coupled toughness degradation model is obtained. This model comprehensively reflects the current toughness degradation status of the pipeline material in a hydrogen environment. Based on the toughness degradation model, a material toughness decay curve is generated, comprehensively describing the changes in fracture toughness with hydrogen concentration, temperature, and stress state, providing a fundamental support for subsequent damage analysis.

[0021] Furthermore, the process of obtaining the hydrogen diffusion and accumulation rate parameters includes:

[0022] In this embodiment, the acquired material toughness decay curve data is combined with elastoplastic fracture mechanics theory to construct a morphological model of the plastic zone at the crack tip. The geometric characteristics of the crack tip region are analyzed, and key information such as the size, shape, and distribution characteristics of the plastic deformation zone are extracted to form a geometric feature dataset of the plastic zone. This dataset provides a basic spatial reference framework for subsequent hydrogen accumulation analysis. Based on this, a hydrogen diffusion control equation considering the influence of stress gradients is constructed, introducing a stress-assisted diffusion term as a driving factor to establish a hydrogen atom migration driving force model. This model reveals the mechanism by which stress gradients promote hydrogen diffusion and explains the physical process of hydrogen atom migration to high-stress regions. Based on the established hydrogen atom migration driving force model, the hydrogen atom trapping mechanism in the crack tip region is further analyzed, considering both reversible and irreversible traps in the material on hydrogen diffusion. The study aimed to extract hydrogen capture kinetic parameters, which accurately describe the capture and release behavior of hydrogen atoms at microscopic defects such as dislocations and grain boundaries, reflecting the regulatory role of material microstructure on the hydrogen diffusion process. By combining the obtained hydrogen capture kinetic parameters with geometric feature data of the plastic zone, a time-dependent model of hydrogen accumulation in the plastic zone was constructed. Based on this model, the evolution of hydrogen concentration in the crack tip region over time was tracked, forming a function of hydrogen accumulation concentration versus time, revealing the dynamic characteristics of hydrogen accumulation behavior. Through numerical simulation and experimental verification, the established function was calibrated and optimized, ultimately obtaining accurate hydrogen diffusion distribution and accumulation rate parameters. These parameters quantitatively describe the spatial distribution and temporal accumulation rate of hydrogen in the crack tip plastic zone, providing crucial input data for subsequent crack propagation prediction analysis.

[0023] Furthermore, the process of obtaining the damage development trajectory prediction results includes:

[0024] In this embodiment, the acquired hydrogen diffusion distribution and accumulation rate parameters are used as basic inputs to construct a hydrogen embrittlement crack propagation rate model. A dual-parameter control mechanism of hydrogen concentration and stress intensity factor is introduced to establish a crack propagation rate function. This function quantitatively describes the mathematical relationship between crack propagation speed and key influencing factors, revealing the mechanism by which hydrogen promotes crack propagation. Based on the deployed IoT sensor terminal network, the operating parameter data of the hydrogen-infused pipeline are collected in real time, and the data is analyzed and processed to extract the fluctuation patterns of key parameters such as hydrogen concentration, pressure, and temperature, forming a characteristic vector of operating condition changes. This characteristic vector comprehensively reflects the dynamic operating state characteristics of the pipeline system. Combining the characteristic vector of operating condition changes and the crack propagation rate function, a state-space model of the hydrogen damage evolution of the pipeline system is constructed. The complex hydrogen damage process is expressed as a series of states and their transition relationships. A damage state transition equation describing the temporal evolution of the system state is established, which provides a mathematical basis for subsequent predictive analysis. Based on the damage state transition equation, a multi-timescale analysis method is introduced to establish a damage accumulation prediction model that can simultaneously take into account short-term fluctuations and long-term trends. The damage accumulation prediction model outputs damage state prediction values ​​at different time points, enabling comprehensive tracking of the damage development process. Through advanced time series analysis and trend extraction techniques, the damage state prediction values ​​are processed in depth to identify key evolutionary inflection points and acceleration periods in the hydrogen damage development process, resulting in complete damage development trajectory prediction results. The damage development trajectory prediction results demonstrate the evolution law of hydrogen damage from the latent period, stable development period to the accelerated development period.

[0025] Furthermore, the process of obtaining the preliminary hydrogen damage level assessment results includes:

[0026] In this embodiment, based on the acquired damage development trajectory prediction results, key damage characteristics are analyzed, core indicators and threshold parameters required for assessment are determined, and a comprehensive assessment indicator system including multiple dimensions such as crack size, propagation rate, and residual strength ratio is constructed. The relative importance relationships between each indicator are established, forming an assessment indicator weight matrix. This weight matrix provides a quantitative basis for multi-indicator comprehensive evaluation. Combining the pipeline system's remaining life estimation and risk tolerance analysis, a grading standard framework based on damage development rate and degree is established, and multi-level early warning thresholds are set, forming a complete damage grading standard system. This system divides hydrogen damage into different levels, each corresponding to a specific risk level and management measures. This system addresses hydrogen-induced cracking, hydrogen bubbling, and hydrogen embrittlement fracture, among other damage types. Differentiated assessment standards are established for different types of hydrogen damage mechanisms, and mechanism-specific assessment sub-models are developed based on them. The development of these sub-models must consider the characteristics and severity of different hydrogen damage mechanisms and provide targeted assessment methods. A structured comprehensive assessment decision tree is constructed by integrating the assessment index weight matrix, damage grading standard framework, and mechanism-specific assessment sub-models. Based on this tree, a pre-selected convolutional neural network is iteratively trained to obtain the damage assessment model. Based on this damage assessment model, a multi-dimensional analysis and risk level determination are performed on the current hydrogen-blended pipeline status, outputting preliminary hydrogen damage level assessment results. These results include damage level, key indicator values, and risk assessment conclusions, providing preliminary decision-making basis for pipeline safety management.

[0027] Furthermore, the process of obtaining the optimized damage assessment model includes:

[0028] In this embodiment, historical damage cases of hydrogen-doped pipelines are systematically collected and organized, and a structured historical damage database is constructed based on these cases. Key information such as damage type, degree, development process, and final result is extracted to form a high-quality dataset for model training. The preliminary hydrogen damage level assessment results are compared and analyzed with the actual historical damage situation to establish a model error analysis framework, quantify and evaluate prediction bias, identify systematic and random errors in the model, and generate model error feature vectors. These model error feature vectors reflect the current performance bottlenecks and improvement directions of the model. Advanced machine learning algorithms such as gradient boosting decision trees are used to construct a parameter optimization learning model for scenarios with large errors. Feature importance analysis is performed to determine the degree of influence of different parameters on the evaluation results, generating a parameter sensitivity matrix. This matrix reveals the key parameters that need to be optimized. Based on the parameter sensitivity matrix, a Bayesian optimization method is introduced to efficiently search for the optimal parameter combination in the parameter space. This adaptively adjusts the key parameters of the damage assessment model, forming a parameter optimization strategy that balances computational efficiency and optimization effectiveness. Based on this optimization strategy, the damage assessment model is iteratively optimized and cross-validated to obtain an optimized damage assessment model. This model has higher prediction accuracy and generalization ability, and can adapt to the hydrogen damage assessment needs under different operating conditions.

[0029] Furthermore, the process for obtaining the final hydrogen damage grading assessment results includes:

[0030] In this embodiment, based on the optimized damage assessment model, a high-performance real-time assessment engine is developed, a data processing pipeline and assessment rule execution mechanism are established, and an online assessment platform supporting real-time data access, rapid processing, and automatic assessment is constructed. Based on IoT sensor terminal devices, pipeline operating parameters and external environmental parameters are continuously collected, and real-time feature extraction and anomaly detection are performed to generate a real-time feature vector reflecting the current state of the pipeline. This real-time feature vector provides the latest data input for damage assessment. Combining the real-time state feature vector and historical trend analysis, hydrogen damage risk is calculated for the current state of the pipeline, generating a risk probability distribution map showing the probability distribution of damage at various locations. This risk probability distribution map intuitively reflects the spatial distribution of risk. Based on the risk probability distribution map, damage level classification and risk area identification algorithms are executed to generate a visual assessment report that graphically displays damage distribution and level information. Combining the visual assessment report and remaining life prediction, differentiated maintenance strategies and detection plans are formulated for different risk areas, outputting the final hydrogen damage grading assessment result. The hydrogen damage grading assessment result includes complete information such as damage level, risk assessment, remaining life prediction, and maintenance recommendations, providing comprehensive scientific support for pipeline system safety management and maintenance decisions.

[0031] The technical effects and advantages of the present invention's fracture mechanics-based method for classifying and assessing hydrogen damage in hydrogen-doped pipelines are as follows:

[0032] This invention utilizes digital twin technology to achieve real-time mapping of pipeline conditions, providing an accurate data foundation for assessment. It innovatively introduces a hydrogen concentration-dependent term to modify the J-integral expression, making fracture mechanics analysis more consistent with the realities of hydrogen-doped environments. A multi-parameter coupled toughness degradation model considering temperature and stress states is constructed to accurately describe the toughness decay law of materials under complex conditions. A hydrogen accumulation model in the crack tip plastic zone is established based on microscopic mechanisms, revealing the spatiotemporal evolution process of hydrogen diffusion and accumulation. A multi-timescale analysis method is employed to predict damage trajectories from short-term fluctuations to long-term trends. A dynamic response grading standard system is established to support differentiated maintenance decisions. Model parameters are optimized through historical cases and machine learning methods, significantly improving assessment accuracy. Ultimately, a complete real-time monitoring and grading assessment solution is formed, providing a scientific basis and decision support for the safe operation of hydrogen-doped pipelines. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the hydrogen damage classification assessment method for hydrogen-doped pipelines based on fracture mechanics, as described in this invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Example 1

[0036] Please see Figure 1 As shown in this embodiment, the hydrogen damage grading assessment method for hydrogen-doped pipelines based on fracture mechanics includes:

[0037] Step S1: Based on the pre-collected pipeline feature information, construct a model to obtain the corresponding hydrogen-doped pipeline digital twin model; and perform multiphysics simulation based on it to obtain the corresponding initial pipeline state dataset.

[0038] In this embodiment, various types of data generated in the target hydrogen-doped pipeline environment are collected in real time from multiple data acquisition channels, including the pipeline system, monitoring equipment, and data acquisition equipment. The collected data is then processed to obtain corresponding pipeline characteristic information. Simultaneously, an IoT sensor terminal network is deployed to collect data on operating parameters and external environmental parameters within the target hydrogen-doped pipeline, obtaining corresponding multi-source heterogeneous data streams. Feature extraction is performed on these multi-source heterogeneous data streams to obtain various pipeline state characteristic indicators that can describe the operating characteristics of the hydrogen-doped pipeline. These pipeline state characteristic indicators consist of multiple dimensions, including temperature, pressure, flow rate, and hydrogen concentration. By using digital twin technology and combining it with pipeline feature information, the extracted pipeline feature indicators are data mined and characterized, and then visualized in the form of a digital twin model to form a digital twin model of the hydrogen-doped pipeline that can be used to reflect the current characteristics of the hydrogen-doped pipeline. In this model, different nodes represent physical entities within the hydrogen-doped pipeline, and the distribution of physical fields represents the physical state and physical intensity inside the pipeline. By applying a multiphysics coupling analysis method and simulating the interaction and influence between multiple physical fields such as temperature field, stress field, and hydrogen concentration field based on the digital twin model of the hydrogen-doped pipeline, the corresponding initial pipeline state dataset is obtained.

[0039] Step S2: Construct an initial set of calculation parameters based on the initial pipeline state dataset, and modify the traditional J integral expression based on it; perform stress analysis on the hydrogen-doped pipeline according to the modified J integral expression to obtain the corresponding stress distribution and intensity factor dataset;

[0040] In this embodiment, by analyzing the hydrogen concentration distribution and geometric features within the initial pipeline state dataset, an initial calculation parameter set is extracted. This initial calculation parameter set includes multiple dimensions such as hydrogen concentration gradient, pipeline geometric dimensions, and material properties. Based on fracture mechanics theory, an energy release rate correction model considering the influence of hydrogen concentration extraction is established. Furthermore, the traditional J-integral formula is modified by introducing a hydrogen concentration dependency term, resulting in a modified J-integral expression. A finite element computational mesh is established based on the J-integral expression and discretized to construct a more complete numerical solution framework. This numerical solution framework more accurately reflects the actual fracture mechanics state of the current hydrogen-doped pipeline environment. Moreover, various nodes and elements within the corresponding mathematical solution framework are endowed with rich physical properties, forming a three-dimensional and dynamic stress distribution and intensity factor dataset.

[0041] Step S3: Construct a hydrogen concentration-toughness experimental data set based on the pipeline stress distribution and strength factor dataset; and obtain the material toughness decay curve by introducing temperature and stress state factors;

[0042] In this embodiment, the pipeline stress distribution and strength factor dataset is combined with the basic fracture toughness value of the corresponding hydrogen-doped pipeline material. An initial toughness reference point is established through global analysis. Simultaneously, the changes in material fracture toughness values ​​under different hydrogen concentrations are tracked and recorded. Through in-depth analysis of the obtained hydrogen concentration-toughness experimental data set, various statistical indicators describing toughness degradation characteristics are extracted, including toughness decay rate, critical hydrogen concentration, saturation degradation value, and other dimensions. Based on mathematical modeling and other techniques, the toughness degradation characteristics are accurately fitted and modeled, establishing a mathematical expression for fracture toughness degradation. By incorporating the nonlinear influence of hydrogen concentration and further introducing temperature and stress state influence factors to correct the toughness degradation model, a multi-parameter coupled toughness degradation model is obtained. This model comprehensively reflects the current toughness degradation status of the pipeline material in a hydrogen environment. Based on the toughness degradation model, a material toughness decay curve is generated, comprehensively describing the changes in fracture toughness with hydrogen concentration, temperature, and stress state, providing a fundamental support for subsequent damage analysis.

[0043] Step S4: Based on the obtained material toughness decay curve, perform crack tip analysis to obtain the corresponding geometric feature data of the plastic zone, and combine the pre-constructed trapping kinetic parameters to perform time series analysis to obtain the corresponding hydrogen diffusion and accumulation rate parameters.

[0044] In this embodiment, the acquired material toughness decay curve data is combined with elastoplastic fracture mechanics theory to construct a morphological model of the plastic zone at the crack tip. The geometric characteristics of the crack tip region are analyzed, and key information such as the size, shape, and distribution characteristics of the plastic deformation zone are extracted to form a geometric feature dataset of the plastic zone. This dataset provides a basic spatial reference framework for subsequent hydrogen accumulation analysis. Based on this, a hydrogen diffusion control equation considering the influence of stress gradients is constructed, introducing a stress-assisted diffusion term as a driving factor to establish a hydrogen atom migration driving force model. This model reveals the mechanism by which stress gradients promote hydrogen diffusion and explains the physical process of hydrogen atom migration to high-stress regions. Based on the established hydrogen atom migration driving force model, the hydrogen atom trapping mechanism in the crack tip region is further analyzed, considering both reversible and irreversible traps in the material on hydrogen diffusion. The study aimed to extract hydrogen capture kinetic parameters, which accurately describe the capture and release behavior of hydrogen atoms at microscopic defects such as dislocations and grain boundaries, reflecting the regulatory role of material microstructure on the hydrogen diffusion process. By combining the obtained hydrogen capture kinetic parameters with geometric feature data of the plastic zone, a time-dependent model of hydrogen accumulation in the plastic zone was constructed. Based on this model, the evolution of hydrogen concentration in the crack tip region over time was tracked, forming a function of hydrogen accumulation concentration versus time, revealing the dynamic characteristics of hydrogen accumulation behavior. Through numerical simulation and experimental verification, the established function was calibrated and optimized, ultimately obtaining accurate hydrogen diffusion distribution and accumulation rate parameters. These parameters quantitatively describe the spatial distribution and temporal accumulation rate of hydrogen in the crack tip plastic zone, providing crucial input data for subsequent crack propagation prediction analysis.

[0045] Step S5: Obtain real-time operating data of the hydrogen-doped pipeline, and combine it with hydrogen diffusion distribution and accumulation rate parameters to predict the hydrogen damage evolution trend and obtain the damage development trajectory prediction results.

[0046] In this embodiment, the acquired hydrogen diffusion distribution and accumulation rate parameters are used as basic inputs to construct a hydrogen embrittlement crack propagation rate model. A dual-parameter control mechanism of hydrogen concentration and stress intensity factor is introduced to establish a crack propagation rate function. This function quantitatively describes the mathematical relationship between crack propagation speed and key influencing factors, revealing the mechanism by which hydrogen promotes crack propagation. Based on the deployed IoT sensor terminal network, the operating parameter data of the hydrogen-infused pipeline are collected in real time, and the data is analyzed and processed to extract the fluctuation patterns of key parameters such as hydrogen concentration, pressure, and temperature, forming a characteristic vector of operating condition changes. This characteristic vector comprehensively reflects the dynamic operating state characteristics of the pipeline system. Combining the characteristic vector of operating condition changes and the crack propagation rate function, a state-space model of the hydrogen damage evolution of the pipeline system is constructed. The complex hydrogen damage process is expressed as a series of states and their transition relationships. A damage state transition equation describing the temporal evolution of the system state is established, which provides a mathematical basis for subsequent predictive analysis. Based on the damage state transition equation, a multi-timescale analysis method is introduced to establish a damage accumulation prediction model that can simultaneously take into account short-term fluctuations and long-term trends. The damage accumulation prediction model outputs damage state prediction values ​​at different time points, enabling comprehensive tracking of the damage development process. Through advanced time series analysis and trend extraction techniques, the damage state prediction values ​​are processed in depth to identify key evolutionary inflection points and acceleration periods in the hydrogen damage development process, resulting in complete damage development trajectory prediction results. The damage development trajectory prediction results demonstrate the evolution law of hydrogen damage from the latent period, stable development period to the accelerated development period.

[0047] Step S6: Based on the damage development trajectory prediction results, establish a dynamic response damage grading standard system, and conduct a risk assessment of the current state of the hydrogen-doped pipeline based on it to obtain a preliminary hydrogen damage level assessment result.

[0048] In this embodiment, based on the acquired damage development trajectory prediction results, key damage characteristics are analyzed, core indicators and threshold parameters required for assessment are determined, and a comprehensive assessment indicator system including multiple dimensions such as crack size, propagation rate, and residual strength ratio is constructed. The relative importance relationships between each indicator are established, forming an assessment indicator weight matrix. This weight matrix provides a quantitative basis for multi-indicator comprehensive evaluation. Combining the pipeline system's remaining life estimation and risk tolerance analysis, a grading standard framework based on damage development rate and degree is established, and multi-level early warning thresholds are set, forming a complete damage grading standard system. This system divides hydrogen damage into different levels, each corresponding to a specific risk level and management measures. This system addresses hydrogen-induced cracking, hydrogen bubbling, and hydrogen embrittlement fracture, among other damage types. Differentiated assessment standards are established for different types of hydrogen damage mechanisms, and mechanism-specific assessment sub-models are developed based on them. The development of these sub-models must consider the characteristics and severity of different hydrogen damage mechanisms and provide targeted assessment methods. A structured comprehensive assessment decision tree is constructed by integrating the assessment index weight matrix, damage grading standard framework, and mechanism-specific assessment sub-models. Based on this tree, a pre-selected convolutional neural network is iteratively trained to obtain the damage assessment model. Based on this damage assessment model, a multi-dimensional analysis and risk level determination are performed on the current hydrogen-blended pipeline status, outputting preliminary hydrogen damage level assessment results. These results include damage level, key indicator values, and risk assessment conclusions, providing preliminary decision-making basis for pipeline safety management.

[0049] Step S7: Obtain historical hydrogen damage cases and optimize the parameters of the constructed damage assessment model based on the preliminary hydrogen damage level assessment results to obtain the optimized damage assessment model.

[0050] In this embodiment, historical damage cases of hydrogen-doped pipelines are systematically collected and organized, and a structured historical damage database is constructed based on these cases. Key information such as damage type, degree, development process, and final result is extracted to form a high-quality dataset for model training. The preliminary hydrogen damage level assessment results are compared and analyzed with the actual historical damage situation to establish a model error analysis framework, quantify and evaluate prediction bias, identify systematic and random errors in the model, and generate model error feature vectors. These model error feature vectors reflect the current performance bottlenecks and improvement directions of the model. Advanced machine learning algorithms such as gradient boosting decision trees are used to construct a parameter optimization learning model for scenarios with large errors. Feature importance analysis is performed to determine the degree of influence of different parameters on the evaluation results, generating a parameter sensitivity matrix. This matrix reveals the key parameters that need to be optimized. Based on the parameter sensitivity matrix, a Bayesian optimization method is introduced to efficiently search for the optimal parameter combination in the parameter space. This adaptively adjusts the key parameters of the damage assessment model, forming a parameter optimization strategy that balances computational efficiency and optimization effectiveness. Based on this optimization strategy, the damage assessment model is iteratively optimized and cross-validated to obtain an optimized damage assessment model. This model has higher prediction accuracy and generalization ability, and can adapt to the hydrogen damage assessment needs under different operating conditions.

[0051] Step S8: Based on the optimized damage assessment model, the degree of hydrogen damage in the hydrogen-doped pipeline is monitored and graded in real time to obtain the final hydrogen damage grading assessment result.

[0052] In this embodiment, based on the optimized damage assessment model, a high-performance real-time assessment engine is developed, a data processing pipeline and assessment rule execution mechanism are established, and an online assessment platform supporting real-time data access, rapid processing, and automatic assessment is constructed. Based on IoT sensor terminal devices, pipeline operating parameters and external environmental parameters are continuously collected, and real-time feature extraction and anomaly detection are performed to generate a real-time feature vector reflecting the current state of the pipeline. This real-time feature vector provides the latest data input for damage assessment. Combining the real-time state feature vector and historical trend analysis, hydrogen damage risk is calculated for the current state of the pipeline, generating a risk probability distribution map showing the probability distribution of damage at various locations. This risk probability distribution map intuitively reflects the spatial distribution of risk. Based on the risk probability distribution map, damage level classification and risk area identification algorithms are executed to generate a visual assessment report that graphically displays damage distribution and level information. Combining the visual assessment report and remaining life prediction, differentiated maintenance strategies and detection plans are formulated for different risk areas, outputting the final hydrogen damage grading assessment result. The hydrogen damage grading assessment result includes complete information such as damage level, risk assessment, remaining life prediction, and maintenance recommendations, providing comprehensive scientific support for pipeline system safety management and maintenance decisions.

[0053] It should be further explained that, in the specific implementation process, step S1 includes the following steps:

[0054] Step S11: Construct a pipeline geometry model and material property database, and build a basic digital twin model of the pipeline based on it to obtain an initial digital twin framework;

[0055] Step S12: Deploy an IoT sensor terminal network and collect real-time data on the operating parameters of the hydrogen-doped pipeline and external environmental parameters to obtain a multi-source heterogeneous sensor data stream.

[0056] Step S13: Perform data cleaning and feature extraction on the multi-source heterogeneous sensor data stream to obtain the pipeline state feature vector;

[0057] Step S14: Based on the pipeline state feature vector and the initial digital twin framework, establish a data-driven pipeline system state mapping mechanism to obtain a real-time updated digital twin model of the hydrogen-doped pipeline.

[0058] Step S15: Perform multiphysics coupling analysis on the hydrogen-doped pipeline based on the digital twin model to obtain the initial pipeline state dataset;

[0059] Specifically, by connecting to data acquisition channels, pipeline characteristic information corresponding to the target hydrogen-doped pipeline is obtained. This pipeline characteristic information includes geometric structure, material properties, and operating conditions. Based on this pipeline characteristic information, a pipeline geometric model and a material property database are constructed. The pipeline geometric model can be used to describe the geometric structure information of the hydrogen-doped pipeline. Geometric structure information includes geometric parameters such as pipeline diameter, wall thickness, and connection method. The material property database contains data on the basic mechanical properties, hydrogen sensitivity, and environmental adaptability of hydrogen-doped pipelines corresponding to different materials. A material behavior model considering the hydrogen environment is constructed using the material property database, and this model is integrated with the constructed pipeline geometric model to obtain the corresponding initial twin framework.

[0060] An IoT sensor terminal network is set up and deployed inside the target hydrogen-doped pipeline. The network collects operational parameters (such as temperature, pressure, flow rate, and hydrogen concentration) and external environmental parameters (such as geological conditions and climate factors) within the pipeline, obtaining corresponding multi-source heterogeneous data streams. The IoT sensor terminal network consists of several types of sensors, including temperature sensors, pressure sensors, flow rate sensors, and hydrogen concentration sensors. The collected multi-source heterogeneous data streams are cleaned, and feature extraction is performed to obtain corresponding pipeline state features, including temperature fluctuation characteristics, pressure change patterns, and hydrogen concentration distribution characteristics. These extracted pipeline state features are then integrated into a structured pipeline state feature distribution. The initial digital twin model is used to obtain the pipeline state feature vectors for subsequent analysis and use. By combining the obtained pipeline state feature vectors with the initial digital twin framework, a data-driven pipeline system state mapping mechanism is established, forming a hydrogen-doped pipeline digital twin model with real-time data synchronization and state updates. The state mapping mechanism uses a combination of Kalman filtering and Bayesian updating to ensure real-time synchronization between the digital model and the physical pipeline state. Based on the hydrogen-doped pipeline digital twin model, the interaction and influence between multiple physical fields such as temperature field, stress field, and hydrogen concentration field are simulated. Multiphysics coupling analysis is performed based on the simulation results, and the initial pipeline state dataset is formed by integrating the analysis results. The initial pipeline state dataset can be used to comprehensively characterize the current state and potential risk areas of the hydrogen-doped pipeline in the pipeline system, providing basic data support for subsequent assessment.

[0061] It should be further explained that, in the specific implementation process, step S2 involves the following steps:

[0062] Step S21: Extract features and perform discrete representation of the computational domain on the initial pipeline state dataset to form an initial computational parameter set;

[0063] Step S22: Construct an energy release rate correction model and modify the traditional J integral formula by introducing a hydrogen concentration dependence term to obtain the modified J integral expression;

[0064] Step S23: Construct a numerical solution framework based on the modified J integral expression and the initial calculation parameter set;

[0065] Step 24: Calculate the stress intensity factor at key locations within the hydrogen-doped pipeline based on the numerical solution framework, and obtain the intensity factor distribution map;

[0066] Step S25: Establish a coupling relationship model between stress, hydrogen concentration, and intensity factor by combining the intensity factor distribution map, and generate stress distribution and intensity factor datasets based on it;

[0067] Specifically, data mining algorithms are used to identify the spatial distribution characteristics and gradient variation patterns of hydrogen concentration in hydrogen-doped pipelines from the initial pipeline state dataset. Simultaneously, the geometric features of the hydrogen-doped pipelines are extracted, including wall thickness variations, welded joints, and bends. The extracted spatial distribution and gradient variation patterns of hydrogen concentration are integrated with the corresponding geometric features and then discretized in the computational domain to form an initial set of computational parameters. This discretization uses finite element analysis to transform the continuous physical space or computational domain into discrete nodes, elements, or meshes for numerical computation. The initial set of computational parameters includes discrete mesh information, material property parameters, and boundary conditions—the fundamental data required for computation—providing a foundation for subsequent basic calculations. Based on elastoplastic fracture mechanics theory, the influence mechanism of hydrogen concentration gradient on the fracture energy release rate of different hydrogen-doped pipeline materials is analyzed, and an energy release rate correction model considering the influence of hydrogen concentration gradient is established.

[0068] The energy release rate correction model introduces a hydrogen concentration dependency term and modifies the traditional J integral formula based on it; the modified J integral expression is obtained; where the modified J integral expression is: In the formula, This represents the modified J-integral expression; This represents a closed integral path around the crack tip; Let represent the length element and area element on the corresponding integration path; s and A represent the length of the integration path and the surrounding area, respectively. This indicates the required energy density correction term, and ; Represents the strain tensor; The strain tensor function is denoted by ; W represents the strain energy intensity. Indicates hydrogen concentration; The traction vector affected by hydrogen concentration; This represents partial derivative operations; The displacement gradient is represented by u, which is used to represent the partial derivative of the displacement with respect to the coordinates. u represents the displacement vector, and x represents the crack propagation direction parallel to the crack surface, which is the horizontal coordinate direction. Represents the hydrogen concentration gradient; hydrostatic pressure stress; The Laplace operator represents hydrostatic pressure stress, and the rate of change of the stress gradient represents the stress gradient. represents a material constant used to characterize the strength of the interaction between hydrogen and the stress field; its value is determined experimentally. n represents the unit outward normal vector on the integration path. This represents a small element perpendicular to the crack propagation direction y. It can be used to represent the hydrogen concentration gradient term; This can be used to represent the hydrogen-stress field interaction term; among which, the energy density correction term, hydrogen concentration gradient term, and hydrogen-stress field interaction term are all introduced hydrogen concentration-dependent terms; based on the modified J integral expression and combined with the boundary conditions in the initial calculation parameter set, a solution strategy is constructed to obtain a numerical solution framework; based on the constructed numerical solution framework, the stress intensity factor is calculated at key locations (e.g., welds, bends, defects, etc.) in the hydrogen-doped pipeline to obtain the stress intensity factor at different locations; and based on this, a corresponding intensity factor distribution map of the hydrogen-doped pipeline is generated, which can be used to display the stress intensity distribution in the entire hydrogen-doped pipeline;

[0069] By combining the strength factor distribution map and the stress concentration region within the hydrogen-doped pipeline, the interrelationship among stress distribution, hydrogen concentration, and stress intensity factor is analyzed, and a coupling model among the three is established. This coupling model can be used to describe how hydrogen concentration affects stress distribution and stress intensity factor, and how stress distribution affects hydrogen diffusion behavior. The coupling model adopts a multivariate function form. Based on the established coupling model, a complete dataset of pipeline stress distribution and strength factor is generated, providing data support for subsequent fracture toughness analysis.

[0070] It should be further explained that, in the specific implementation process, the traditional formula for the J integral is: .

[0071] It should be further explained that, in the specific implementation process, step S3 includes the following steps:

[0072] Step S31: Based on the strength factor data and combined with the material standard test results, conduct a hydrogen-free environment benchmark analysis to obtain a reference dataset, and obtain benchmark toughness data by statistical analysis of the reference dataset;

[0073] Step S32: Design a fracture toughness test, obtain material fracture parameters under different hydrogen concentration conditions based on the fracture toughness test results, and perform data regression processing on the obtained material fracture parameters to obtain a hydrogen concentration-toughness test data point set;

[0074] Step S33: Based on the hydrogen concentration-toughness experimental data set, and combined with the nonlinear regression method, a basic toughness degradation function is constructed. The functional form of the basic toughness degradation function is optimized through physical mechanism analysis to obtain the toughness degradation function.

[0075] Step S34: Construct temperature factor and stress state factor, and introduce them into toughness degradation function to modify the original toughness degradation function, obtain toughness degradation model, and perform model parameter calibration and optimization through verification test to generate material toughness decay curve;

[0076] Specifically, the reference fracture toughness values ​​of the materials corresponding to the hydrogen-doped pipelines in a hydrogen-free environment are measured using standard material testing methods to obtain a reference dataset. For example, the critical J-integral value (JIC) of the material is obtained through three-point bending tests or compact tensile tests, serving as a reference value for the reference fracture toughness. Based on strength factor data and through statistical analysis of the reference datasets corresponding to multiple sets of standard material testing results, a reliable range of values ​​corresponding to the reference fracture toughness is determined, and benchmark toughness data is produced to provide a comparison benchmark for subsequent toughness degradation analysis. Hydrogen embrittlement fracture experiments are conducted on the hydrogen-doped pipeline materials under different hydrogen concentration conditions using standard fracture toughness testing methods (such as CTOD, J-integral, or KIC tests) to obtain the values ​​for different hydrogen concentrations. The fracture toughness value of the material at different hydrogen concentrations was determined. During the hydrogen embrittlement fracture test, the experimental conditions needed to cover the hydrogen concentrations that the corresponding hydrogen-doped pipeline might encounter in its actual service environment. By recording the hydrogen embrittlement test results, the correlation between different hydrogen concentrations and the material's fracture toughness value was extracted from the results, forming a hydrogen concentration-toughness experimental data set. Based on the obtained hydrogen concentration-toughness experimental data set, the nonlinear relationship between hydrogen concentration and the material's fracture toughness value was extracted. A mathematical expression that accurately describes the toughness degradation law was established using a nonlinear regression method, obtaining the corresponding basic toughness degradation function. The obtained basic toughness degradation function was optimized through physical mechanism analysis to obtain the corresponding toughness degradation function. Where KIC(CH) represents the fracture toughness of the material at a hydrogen concentration of CH. Indicates the baseline fracture toughness value; Represents the rate coefficient, and Units are ; Represents the sensitivity coefficient, and Due to material differences, the specific values ​​of the rate coefficient and sensitivity coefficient vary for different materials. Physical mechanism analysis involves optimizing the expression of the basic toughness degradation function based on hydrogen embrittlement theory to better reflect its physical nature. This optimization process also considers the interaction mechanisms between hydrogen atoms and dislocations and grain boundaries. The influence of temperature and stress state on toughness degradation is analyzed, establishing temperature and stress state factors, which are then incorporated into the toughness degradation function to correct the original function. This yields a toughness degradation model considering multi-parameter coupling effects. Verification experiments are conducted to calibrate and optimize the model parameters. Verification experiments involve selecting experimental conditions not involved in model construction, comparing the measured actual fracture toughness values ​​with the model predictions, and obtaining the corresponding prediction errors. Based on these prediction errors, parameter sensitivity analysis and optimization algorithms are used to adjust the model parameters to improve prediction accuracy. Based on the optimized toughness degradation model, material toughness degradation curves are obtained for hydrogen-doped pipelines under different service conditions (different temperatures, stresses, hydrogen concentrations, etc.), accurately reflecting the toughness degradation law of the pipeline material.

[0077] It should be further explained that, in the specific implementation process, step S4 includes the following steps:

[0078] Step S41: Based on the material toughness decay curve and combined with the elastic-plastic fracture mechanics theory, construct a morphological model of the plastic zone at the crack tip, and predict the morphology of the plastic zone based on it to obtain the geometric feature data of the plastic zone.

[0079] Step S42: Establish the hydrogen diffusion control equation and generate a hydrogen atom migration driving force model by introducing a stress-assisted diffusion term;

[0080] Step S43: Identify and extract the hydrogen atom trapping mechanism in the crack tip region based on the hydrogen atom migration driving force model to obtain hydrogen trapping kinetic parameters;

[0081] Step S44: Combining hydrogen capture kinetic parameters and geometric feature data of the plastic zone, establish a time-dependent model of hydrogen accumulation in the plastic zone, thereby obtaining the function of hydrogen accumulation concentration changing with time;

[0082] Step S45: Through numerical simulation and experimental verification, the function of hydrogen accumulation concentration changing with time is calibrated and optimized to obtain the parameters of hydrogen diffusion distribution and accumulation rate.

[0083] Specifically, based on the material toughness decay curve and the theory of elastoplastic fracture mechanics, a geometric model of the plastic zone at the crack tip is constructed. This model describes the geometric deformation region around the crack tip. The theoretical plastic zone morphology corresponding to the crack tip period is predicted based on this model. The shape and size of the plastic zone at the crack tip are calculated under different loading conditions, generating geometric characteristic data of the plastic zone, including its size, shape, and distribution. The yield behavior and work hardening characteristics of the material must be considered during the calculation process. The diffusion process of hydrogen within the pipeline system under stress gradient is analyzed. The hydrogen diffusion mechanism is analyzed, and a hydrogen diffusion control equation considering the influence of stress gradients is established based on the analysis results. Furthermore, a hydrogen atom migration driving force model is generated by incorporating a stress-assisted diffusion term. This model describes the migration behavior of hydrogen atoms under stress gradients. The stress-assisted diffusion term is a mathematical component added to the hydrogen diffusion control equation to explain how stress gradients affect hydrogen movement (e.g., how hydrogen atoms preferentially migrate to regions with high tensile stress). The hydrogen atom migration driving force model obtains the migration path of hydrogen atoms and analyzes the hydrogen atom trapping effect in the crack tip region based on the trapping effect of microscopic defects such as dislocations, vacancies, and grain boundaries. The mechanism of hydrogen atom trapping is established by distinguishing the effects of reversible and irreversible traps. A hydrogen atom trapping kinetic model can be used to describe the transfer process of hydrogen atoms between interstitial and trap sites. Trap density and binding energy distribution are determined through microstructural analysis, generating hydrogen trapping kinetic parameters, including trapping rate and untrap rate. Microstructural analysis employs techniques such as transmission electron microscopy (TEM) and atomic force microscopy (AFM) to observe microscopic defects such as dislocations, grain boundaries, and second-phase particles in the material, determining the trap type and density. Binding energy distribution is measured using thermal desorption spectroscopy (TDS). The hydrogen binding energy of different types of traps was determined; a time-dependent hydrogen accumulation concentration model was established by combining hydrogen capture kinetic parameters and geometric characteristic data of the plastic zone, which can be used to describe the change of hydrogen accumulation concentration over time; by considering the dynamic equilibrium process of hydrogen diffusion, capture, and untrapping, a function of hydrogen accumulation concentration over time was generated; the diffusion and accumulation process of hydrogen in the plastic zone at the crack tip was simulated by numerical simulation, and the accuracy was verified by experimental methods (such as hydrogen permeation experiments, hydrogen content determination, etc.). Based on the verification results, the function of accumulation was calibrated and optimized, and parameters of hydrogen diffusion distribution and accumulation rate were generated based on the optimized function of accumulation.

[0084] It should be further explained that, in the specific implementation process, step S5 includes the following steps:

[0085] Step S51: Establish a hydrogen embrittlement crack propagation rate model based on hydrogen diffusion distribution and accumulation rate parameters, and generate a crack propagation rate function by introducing a dual-parameter control mechanism of hydrogen concentration and stress intensity factor;

[0086] Step S52: Perform fluctuation analysis on the real-time collected pipeline operating parameters to obtain the characteristic vector of operating condition changes;

[0087] Step S53: Establish a state-space model of hydrogen damage evolution based on the characteristic vector of working condition change and crack propagation rate function, form a set of state equations, and solve the state equations using numerical integration method to obtain the damage state transition equations.

[0088] Step S54: Based on the damage state transition equation and combined with multi-timescale analysis methods, establish a damage accumulation prediction model, obtain the cumulative damage curve, identify key evolution inflection points and acceleration periods through trend analysis, obtain damage development trajectory prediction results, and generate damage state prediction values ​​at different time nodes based on them.

[0089] Step S55: Perform time series analysis and trend extraction on the obtained damage state prediction values ​​to obtain the damage development trajectory prediction results.

[0090] Specifically, based on hydrogen diffusion distribution and accumulation rate parameters, a hydrogen-induced crack propagation rate model is constructed, yielding a fundamental rate equation. A dual-parameter control mechanism of hydrogen concentration and stress intensity factor is introduced to establish parameter coupling, and a crack propagation rate function is generated based on this. The construction of the hydrogen-induced crack propagation rate model involves applying fracture mechanics theory and hydrogen embrittlement mechanism to construct a multi-physics coupled mathematical model based on the hydrogen concentration distribution at the crack tip, local stress state, and material microstructure characteristics. This multi-physics coupled mathematical model integrates hydrogen diffusion dynamics, elastoplastic mechanics, and damage mechanics to compare hydrogen-induced crack propagation behavior and critical values ​​under different microscopic mechanisms. Concentration threshold and propagation rate sensitivity are used to assign quantitative contribution weights to each propagation mechanism; the two-parameter control mechanism is based on hydrogen chemical potential gradient analysis and stress intensity field calculation, combined with finite element numerical simulation to identify the main controlling factors of crack propagation, including hydrogen concentration gradient, critical stress intensity factor, plastic zone size and trap density distribution; a piecewise function-based mathematical mapping method is applied to quantify the nonlinear interaction between stress intensity factor and local hydrogen concentration, and the corresponding evaluation results are integrated with the micro-damage accumulation model, and a modified exponential function is applied to construct the hydrogen concentration-stress intensity factor coupling coefficient to form the crack propagation rate function;

[0091] Furthermore, based on the Internet of Things (IoT) sensor terminal network, real-time operating parameters of the target pipeline system are acquired, and parameter fluctuation patterns corresponding to each operating parameter are identified to obtain the corresponding motion feature sequences. Time series analysis is performed on the corresponding operating feature sequences to obtain the corresponding parameter fluctuation characteristic models. Time series analysis refers to training a pre-selected ARIMA (Autoregressive Integral Moving Average) model based on the operating feature sequences to construct a parameter fluctuation characteristic model that can be used to predict future operating parameter change trends. Based on this, corresponding operating condition change feature vectors are generated. Combining the operating condition change feature vectors and the crack propagation rate function, a state-space model of the hydrogen damage evolution of the pipeline system is established. The state-space model can describe the evolution of the pipeline system state over time using a set of first-order differential equations, forming a corresponding set of state equations. The pipeline system state includes key parameters such as crack length, hydrogen concentration, and stress intensity factor.

[0092] The obtained state equations are solved by time discretization using the Runge-Kutta method or the predict-correction method. An error control algorithm is applied to monitor the numerical stability and convergence of the solution process, and the corresponding severance results are verified against physical constraints to construct a complete state-space representation, forming the damage state transition equation. This equation describes the evolution of the system from the current damage state to the next time step damage state under external excitation and can be used to predict the damage accumulation process and critical failure point under different operating conditions. Based on the damage state transition equation, a damage accumulation prediction model is established by combining multi-timescale analysis methods. This model can be used to predict the damage state at different time points (including short-term, medium-term, and long-term). The system generates predicted damage status values ​​for corresponding time points. For example, short-term scales focus on hourly or daily parameter fluctuations, medium-term scales focus on monthly trend changes, and long-term scales focus on annual cumulative effects. Damage status is predicted based on the damage accumulation prediction model, and the obtained damage status prediction values ​​are statistically analyzed to obtain the corresponding cumulative damage curve. Trend analysis is performed on the obtained cumulative damage curve, which involves identifying key stages and turning points in the damage development process using curve fitting methods. The results of the trend analysis are combined to form a complete damage development trajectory prediction result. The damage development trajectory prediction result includes information such as the change curve of damage parameters over time, the division of key evolution stages, and the prediction of critical time points, providing a basis for pipeline safety assessment and maintenance decisions.

[0093] It should be further explained that, in the specific implementation process, step S6 includes the following steps:

[0094] Step S61: Extract key damage indicators based on the damage development trajectory prediction results, and establish a multi-dimensional evaluation indicator system and evaluation indicator weight matrix based on them.

[0095] Step S62: Based on the damage development trajectory prediction results, estimate the remaining life of the pipeline and conduct a risk assessment, and set multi-level early warning thresholds based on them to obtain a damage classification standard framework.

[0096] Step S63: Generate differentiated evaluation criteria for different types of hydrogen damage mechanisms, and construct a model based on the generated evaluation criteria to obtain a mechanism-specific evaluation sub-model;

[0097] Step S64: Construct a damage assessment model based on the assessment index weight matrix, damage grading standard framework, and mechanism-specific assessment sub-model;

[0098] Step S65: Based on the constructed damage assessment model, perform multi-dimensional analysis and risk level determination of the current pipeline status to obtain preliminary hydrogen damage level assessment results.

[0099] Specifically, by extracting the evolution law of key damage features from the obtained damage development prediction trajectory results, a corresponding set of feature parameters is obtained. Key damage features refer to key indicators that can be used to reflect the degree of hydrogen damage, including crack size, propagation rate, and degree of hydrogen accumulation. The relative importance of each damage feature parameter in the corresponding feature parameter set in the overall evaluation process is determined by comprehensively using the analytic hierarchy process (AHP) and fuzzy comprehensive evaluation method. Different levels of threshold parameters are set for each key indicator to establish a multi-dimensional evaluation index system and evaluation index weight matrix. Based on the obtained damage development prediction trajectory results, the remaining life of the target hydrogen-doped pipeline is estimated to obtain the corresponding life estimation result. The life estimation result refers to the time required for the corresponding hydrogen-doped pipeline to go from the current state to the failure state. At the same time, a risk assessment of the hydrogen-doped pipeline is carried out. The risk assessment considers the pipeline failure probability and the comprehensive impact after failure, and is usually represented by a risk matrix. By combining the damage development rate and degree in the damage development prediction trajectory results, a damage analysis framework is constructed to obtain the corresponding initial classification rules. The initial analysis rules classify hydrogen damage into different levels, such as minor damage, moderate damage, severe damage, and critical damage, according to the remaining life and risk level.

[0100] Furthermore, by using expert knowledge and case analysis, early warning thresholds are set for each damage level, forming a complete damage analysis standard framework. The multi-level early warning thresholds include normal operation thresholds, attention thresholds, warning thresholds, and emergency thresholds, which correspond to different risk levels and management measures.

[0101] To obtain specific damage characteristics and evolution patterns corresponding to different types of hydrogen damage mechanisms (such as hydrogen-induced cracking, hydrogen bubbling, and hydrogen embrittlement fracture); for example, hydrogen-induced cracking mainly focuses on crack size and distribution characteristics, hydrogen bubbling focuses on bubbling density and size distribution, and hydrogen embrittlement fracture focuses on material toughness degradation and critical stress intensity factor.

[0102] Differential assessments are conducted for different types of hydrogen damage mechanisms, and corresponding assessment standards are established. Differentiated assessment refers to designing corresponding assessment indicators and standards based on the damage characteristics and evolution patterns of different hydrogen damage mechanisms. Then, based on the assessment standards, a mechanism-specific assessment sub-model is constructed by optimizing the assessment method through case validation. The mechanism-specific assessment sub-model can accurately assess specific damage mechanisms. Case validation selects typical hydrogen damage cases, analyzes them by applying the assessment method, and compares them with the actual damage level to verify the effectiveness of the assessment method. Based on the validation results, the parameters and process of the assessment method are optimized to improve the accuracy of the assessment.

[0103] A comprehensive assessment decision tree is constructed based on the weight matrix of assessment indicators, the damage grading standard framework, and the mechanism-specific assessment sub-model. This comprehensive assessment decision tree can determine the damage level and risk based on the combination of multiple indicators, forming a complete grading assessment algorithm. A damage assessment model is then obtained by iteratively training a pre-selected convolutional neural network using this model. The damage assessment model is then applied to achieve a comprehensive determination of multiple indicators, generating a preliminary hydrogen damage level assessment result. This preliminary hydrogen damage level assessment result includes information such as damage level, key indicator values, dominant mechanism, and risk assessment conclusions, providing a basis for decision-making in pipeline safety management.

[0104] It should be further explained that, in the specific implementation process, step S7 includes the following steps:

[0105] Step S71: Collect and organize historical pipeline damage cases and actual inspection results to obtain the model training dataset;

[0106] Step S72: Based on the model training dataset and the preliminary hydrogen damage level assessment results, establish a model error analysis framework, quantify the assessment bias, and obtain the model error feature vector;

[0107] Step S73: Use the gradient boosting decision tree algorithm to establish a parameter optimization learning model, and combine the model error feature vector to perform feature importance analysis to obtain the parameter sensitivity matrix;

[0108] Step S74: Based on the parameter sensitivity matrix, the key parameters in the damage assessment model are adaptively adjusted using the Bayesian optimization method to obtain the parameter optimization strategy;

[0109] Step S75: Based on the parameter optimization strategy, iteratively optimize and cross-validate the damage assessment model to obtain the optimized damage assessment model.

[0110] Specifically, historical pipeline damage cases are collected and organized, including information on damage type, extent, development process, and final outcome. Simultaneously, actual inspection results are collected, including inspection methods, inspection times, and damage findings. The collected data is cleaned, standardized, and structured to form a training dataset suitable for machine learning models. Based on the training dataset and preliminary hydrogen damage level assessment results, a model error analysis framework is established. This framework quantifies the prediction error of the damage assessment model by comparing the model's predictions with actual damage conditions. Error metrics include mean absolute error (MAE), root mean square error (RMSE), and mean relative error (MRE). Quantitative analysis of prediction errors for different scenarios and damage types is performed to generate a model error feature vector, which describes the model's performance under different conditions. Prediction bias characteristics; Gradient Boosting Decision Tree (GBDT) algorithm is selected to establish a parameter optimization learning model. This model can learn the relationship between error and model parameters. For scenarios with large errors, feature importance analysis is performed to identify key parameters that have a significant impact on the prediction results, generating a parameter sensitivity matrix. This parameter sensitivity matrix describes the sensitivity of each parameter to the prediction error. Based on the parameter sensitivity matrix, Bayesian optimization method is used to adaptively adjust the key parameters of the damage assessment model, finding the parameter combination that minimizes the prediction error, forming a parameter optimization strategy. Based on the parameter optimization strategy, the damage assessment model is iteratively optimized. The optimization effect is evaluated through cross-validation to avoid overfitting. An ensemble learning method is used to integrate multiple optimized models to improve the stability and accuracy of prediction, ultimately obtaining a high-performance optimized damage assessment model.

[0111] It should be further explained that, in the specific implementation process, step S8 includes the following steps:

[0112] Step S81: Based on the optimized damage assessment model, construct a real-time assessment engine, establish a data processing pipeline and assessment rule execution mechanism to obtain an online assessment platform;

[0113] Step S82: Continuously monitor key locations within the pipeline system, collect real-time operating parameters and environmental change data, perform feature extraction and anomaly detection, and obtain a real-time state feature vector;

[0114] Step S83: Based on the real-time state feature vector, and by applying an online evaluation platform, calculate the hydrogen damage risk of the pipeline's current state to obtain a risk probability distribution map;

[0115] Step S84: Based on the risk probability distribution map, classify the damage level and identify the risk area, generate a visual assessment report, and obtain detailed damage distribution results;

[0116] Step S85: Combining detailed damage distribution results and remaining life prediction, develop differentiated maintenance strategies and inspection plans to obtain the final hydrogen damage grading assessment results.

[0117] Based on the optimized damage assessment model, a real-time assessment engine was developed. This engine can receive real-time data and perform rapid processing and assessment. A data processing pipeline was designed, including data reception, cleaning, feature extraction and assessment. An assessment rule execution mechanism was established to ensure the consistency and reliability of the assessment process, forming a complete online assessment platform.

[0118] Continuous monitoring of key components of the pipeline system is achieved through an IoT sensor network, collecting real-time operational parameters such as temperature, pressure, flow rate, and hydrogen concentration. Environmental factors, such as temperature fluctuations and humidity changes, are also monitored. The collected data is processed in real-time, extracting key features and detecting anomalies to generate a real-time state feature vector. Based on this vector and historical trend data, the hydrogen damage risk of the pipeline under its current state is calculated. Considering the combined influence of multiple factors, a risk probability distribution map is generated, showing the risk level distribution at different locations within the system. Based on this map, damage levels are classified, high-risk areas and key areas of concern are identified, and an intuitive, visual assessment report is generated. The report includes detailed damage distribution results. Combining these results with the system's remaining life prediction, differentiated maintenance strategies and inspection plans are developed for areas with different risk levels, forming a complete hydrogen damage grading assessment result. This result provides a scientific basis for pipeline operation and maintenance decisions.

[0119] This invention establishes a digital twin model of a hydrogen-infused pipeline to understand the current pipeline environment and operating status, providing a foundation for subsequent damage assessment. A modified J-integral calculation model considering hydrogen concentration gradients reveals stress distribution characteristics, providing a deeper understanding of pipeline strength factor data. Based on the strength factor data, a fracture toughness degradation function under the influence of hydrogen concentration is constructed to more comprehensively describe the material's toughness decay characteristics. A hydrogen accumulation model in the crack tip plastic zone integrates hydrogen diffusion and accumulation mechanisms, providing a detailed data foundation for subsequent damage assessment. A hydrogen damage evolution trend prediction algorithm is constructed to achieve comprehensive prediction and tracking of damage development trajectories. A dynamic response damage grading standard system is established to analyze damage levels, identify abnormal patterns and potential risks, providing data support for subsequent safety assessments. Machine learning methods are used to optimize assessment model parameters, evaluating the system's predictive ability under different damage conditions to obtain a more accurate assessment model. The optimized assessment model is used to monitor and grade the degree of hydrogen damage in the hydrogen-infused pipeline in real time, discovering potential safety vulnerabilities and abnormal behaviors, and providing improvement directions and strategies for system safety.

[0120] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0121] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0122] In the description of this invention, it should be understood that the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0123] In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0124] In the description of this invention, "several" means one or more, and "a large number" means two or more.

[0125] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0126] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0127] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics, characterized in that, include: Step S1: Based on the pre-collected pipeline feature information, construct a model to obtain the corresponding hydrogen-doped pipeline digital twin model; and perform multiphysics simulation based on it to obtain the corresponding initial pipeline state dataset. Step S2: Construct an initial set of calculation parameters based on the initial pipeline state dataset, and modify the traditional J integral expression based on it; perform stress analysis on the hydrogen-doped pipeline according to the modified J integral expression to obtain the corresponding stress distribution and intensity factor dataset; The specific implementation steps of step S2 are as follows: Step S21: Extract features and perform discrete representation of the computational domain on the initial pipeline state dataset to form an initial computational parameter set; Step S22: Construct an energy release rate correction model and modify the traditional J integral formula by introducing a hydrogen concentration dependence term to obtain the modified J integral expression; Step S23: Construct a numerical solution framework based on the modified J integral expression and the initial computational parameter set; Step S24: Calculate the stress intensity factor at key locations within the hydrogen-doped pipeline based on the numerical solution framework, and obtain the intensity factor distribution map; Step S25: Establish a coupling relationship model between stress, hydrogen concentration, and intensity factor by combining the intensity factor distribution map, and generate stress distribution and intensity factor datasets based on it; Step S3: Construct a hydrogen concentration-toughness experimental data set based on the pipeline stress distribution and strength factor dataset; The material toughness decay curve was obtained by introducing temperature and stress state factors. Step S4: Based on the obtained material toughness decay curve, perform crack tip analysis to obtain the corresponding geometric feature data of the plastic zone, and combine the pre-constructed trapping kinetic parameters to perform time series analysis to obtain the corresponding hydrogen diffusion and accumulation rate parameters. Step S5: Obtain real-time operating data of the hydrogen-doped pipeline, and combine it with hydrogen diffusion distribution and accumulation rate parameters to predict the hydrogen damage evolution trend and obtain the damage development trajectory prediction results. Step S6: Based on the damage development trajectory prediction results, establish a dynamic response damage grading standard system, construct a corresponding damage assessment model based on it, and conduct a risk assessment of the current state of the hydrogen-doped pipeline to obtain a preliminary hydrogen damage level assessment result. Step S7: Obtain historical hydrogen damage cases and optimize the parameters of the constructed damage assessment model based on the preliminary hydrogen damage level assessment results to obtain the optimized damage assessment model. Step S8: Based on the optimized damage assessment model, the degree of hydrogen damage in the hydrogen-doped pipeline is monitored and graded in real time to obtain the final hydrogen damage grading assessment result.

2. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics as described in claim 1, characterized in that, The specific implementation steps of step S1 include: Step S11: Construct a pipeline geometry model and material property database, and build a basic digital twin model of the pipeline based on it to obtain an initial digital twin framework; Step S12: Deploy an IoT sensor terminal network and collect real-time data on the operating parameters of the hydrogen-doped pipeline and external environmental parameters to obtain a multi-source heterogeneous sensor data stream. Step S13: Perform data cleaning and feature extraction on the multi-source heterogeneous sensing data stream to obtain the pipeline state feature vector; Step S14: Establish a data-driven pipeline system state mapping mechanism based on the pipeline state feature vector and the initial digital twin framework to obtain a digital twin model of the hydrogen-doped pipeline; Step S15: Perform multiphysics coupling analysis on the hydrogen-doped pipeline based on the digital twin model to obtain the initial pipeline state dataset.

3. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics as described in claim 1, characterized in that, The specific implementation steps of step S3 include: Step S31: Based on the strength factor data and combined with the material standard test results, conduct a hydrogen-free environment benchmark analysis to obtain a reference dataset, and obtain benchmark toughness data by statistical analysis of the reference dataset; Step S32: Design a fracture toughness test, obtain material fracture parameters under different hydrogen concentration conditions based on the fracture toughness test results, and perform data regression processing on the obtained material fracture parameters to obtain a hydrogen concentration-toughness test data point set; Step S33: Based on the hydrogen concentration-toughness experimental data set, and combined with the nonlinear regression method, a basic toughness degradation function is constructed. The functional form of the basic toughness degradation function is optimized through physical mechanism analysis to obtain the toughness degradation function. Step S34: Construct temperature factor and stress state factor, and introduce them into the toughness degradation function to modify the original toughness degradation function, obtain the toughness degradation model, and perform model parameter calibration and optimization through verification experiments to generate material toughness decay curve.

4. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 3, characterized in that, The specific implementation steps of step S4 include: Step S41: Based on the material toughness decay curve and combined with the elastic-plastic fracture mechanics theory, construct a morphological model of the plastic zone at the crack tip, and predict the morphology of the plastic zone based on it to obtain the geometric feature data of the plastic zone. Step S42: Establish the hydrogen diffusion control equation and generate a hydrogen atom migration driving force model by introducing a stress-assisted diffusion term; Step S43: Based on the hydrogen atom migration driving force model, identify and extract the hydrogen atom trapping mechanism in the crack tip region to obtain hydrogen trapping kinetic parameters; Step S44: Combining the hydrogen capture kinetic parameters and the geometric feature data of the plastic zone, establish a time-dependent model to obtain the function of hydrogen accumulation concentration changing with time; Step S45: The hydrogen accumulation concentration variation function over time is calibrated and optimized through numerical simulation and experimental verification to obtain the hydrogen diffusion distribution and accumulation rate parameters.

5. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 4, characterized in that, The specific implementation steps of step S5 include: Step S51: Establish a hydrogen embrittlement crack propagation rate model based on the hydrogen diffusion distribution and accumulation rate parameters, and generate a crack propagation rate function by introducing a dual-parameter control mechanism of hydrogen concentration and stress intensity factor; Step S52: Perform fluctuation analysis on the real-time collected pipeline operating parameters to obtain the characteristic vector of operating condition changes; Step S53: Based on the operating condition change feature vector and crack propagation rate function, establish a state space model of hydrogen damage evolution, form a set of state equations, and solve the state equations using the numerical integration method to obtain the damage state transition equations. Step S54: Based on the damage state transition equation and combined with the multi-timescale analysis method, establish a damage accumulation prediction model, obtain the cumulative damage curve, identify key evolution inflection points and acceleration periods through trend analysis, obtain damage development trajectory prediction results, and generate damage state prediction values ​​at different time nodes based on them. Step S55: Perform time series analysis and trend extraction on the obtained damage state prediction values ​​to obtain the damage development trajectory prediction results.

6. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 5, characterized in that, The specific implementation steps of step S6 include: Step S61: Extract key damage indicators based on the damage development trajectory prediction results, and establish a multi-dimensional evaluation indicator system and evaluation indicator weight matrix based on them. Step S62: Based on the damage development trajectory prediction results, estimate the remaining life of the pipeline and conduct a risk assessment, and set multi-level early warning thresholds based on them to obtain a damage classification standard framework. Step S63: Generate differentiated evaluation criteria for different types of hydrogen damage mechanisms, and construct a model based on the generated evaluation criteria to obtain a mechanism-specific evaluation sub-model; Step S64: Construct a damage assessment model based on the assessment index weight matrix, damage grading standard framework, and mechanism-specific assessment sub-model; Step S65: Based on the constructed damage assessment model, perform multi-dimensional analysis and risk level determination of the current pipeline status to obtain preliminary hydrogen damage level assessment results.

7. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 6, characterized in that, The specific implementation steps of step S7 include: Step S71: Collect and organize historical pipeline damage cases and actual inspection results to obtain the model training dataset; Step S72: Based on the model training dataset and the preliminary hydrogen damage level assessment results, establish a model error analysis framework, quantify the assessment bias, and obtain the model error feature vector; Step S73: Use the gradient boosting decision tree algorithm to establish a parameter optimization learning model, and combine the model error feature vector to perform feature importance analysis to obtain the parameter sensitivity matrix; Step S74: Based on the parameter sensitivity matrix, the key parameters in the damage assessment model are adaptively adjusted using the Bayesian optimization method to obtain the parameter optimization strategy; Step S75: Based on the parameter optimization strategy, iteratively optimize and cross-validate the damage assessment model to obtain the optimized damage assessment model.

8. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 7, characterized in that, The specific implementation steps of step S8 include: Step S81: Based on the optimized damage assessment model, construct a real-time assessment engine, establish a data processing pipeline and assessment rule execution mechanism, and obtain an online assessment platform; Step S82: Continuously monitor key locations within the pipeline system, collect real-time operating parameters and environmental change data, perform feature extraction and anomaly detection, and obtain a real-time state feature vector; Step S83: Based on the real-time state feature vector, and by applying an online evaluation platform, calculate the hydrogen damage risk of the pipeline's current state to obtain a risk probability distribution map; Step S84: Based on the risk probability distribution map, damage levels are classified and risk areas are identified, a visual assessment report is generated, and detailed damage distribution results are obtained; Step S85: Based on the detailed damage distribution results and remaining life prediction, formulate differentiated maintenance strategies and inspection plans to obtain the final hydrogen damage grading assessment results.

9. The method for graded assessment of hydrogen damage in hydrogen-doped pipelines based on fracture mechanics according to claim 1, characterized in that, The hydrogen concentration dependence term includes the stress density correction term, the hydrogen concentration gradient term, and the hydrogen-stress field interaction term.

Citation Information

Patent Citations

  • Intelligent monitoring method and system for full-life-cycle safety state of oil and gas pipeline

    CN120670967A

  • Evaluation method for predicting crack propagation and residual service life of hydrogen-doped natural gas pipeline

    CN120671393A