<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=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" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Investigation of Reinforced Thermoplastic Pipes (RTPs) Stiffness under Transverse Loading</title>
      <link>https://trid.trb.org/View/2711906</link>
      <description><![CDATA[Reinforced thermoplastic pipes (RTPs) are increasingly used in onshore and offshore oil and gas production industries owing to their lightweight and resistance to corrosion and pressure. Accurate estimation of pipe stiffness (PS) is crucial in onshore engineering. Further research is needed to define the properties of the parameters affecting stiffness. This study investigates the estimation of the stiffness of RTPs under transverse loads, incorporating analytical, numerical, and experimental research methodologies. Additionally, the influence of winding angle, thickness of the reinforced layer, and thicknesses of the liner and cover on the stiffness of the RTP is evaluated and discussed. To achieve this objective, a detailed analysis of composite pipes with external diameters of 90, 200, and 323 mm and wall thicknesses of 14.2, 16.6, and 13.6 mm, respectively, was conducted. These analyses involved the implementation of multiple diameter verification procedures to ensure the accuracy and reliability of the results. In analytical and numerical studies, a layered composite modeling approach has been adopted for RTPs. Consequently, isotropic and anisotropic material properties have been defined for each layer. The studies show that the thickness and winding angle of composite tapes significantly contribute to the stiffness of the pipe, whereas the thickness of the liner and cover is less effective. Additionally, increasing the winding angle to 90° significantly improved the PS. The findings provide a reliable and efficient method for calculating the stiffness of pipes under transverse loading. This method can be utilized during the design stage to simplify the design process for RTPs. A comprehensive evaluation of PS using a layered modeling approach enables researchers to make informed decisions regarding the use of thermoplastic composite tapes for piping applications. This approach can also be used to maximize pipe performance under various stiffness scenarios.]]></description>
      <pubDate>Fri, 28 Aug 2026 13:34:20 GMT</pubDate>
      <guid>https://trid.trb.org/View/2711906</guid>
    </item>
    <item>
      <title>On the development and calibration of a beam-theory based model to calculate the collapse pressure of flexible pipes</title>
      <link>https://trid.trb.org/View/2706407</link>
      <description><![CDATA[As the oil and gas industry operational scenarios move to ultra-deep waters, failure mechanisms in flexible pipes such as instability of the armor layers under compression and hydrostatic collapse are more likely to occur. Therefore, it is important to develop reliable numerical tools to reproduce the failure mechanisms that may occur in flexible pipes. These tools can be used in the design stage or during service-life to assess the structural integrity of pipes under specific operational conditions. This paper presents a methodology to develop simple finite element models capable of reproducing the behavior of structural layers of flexible pipes under external hydrostatic pressure up to collapse. These models use beam elements and, in multi-layer analyses, include nonlinear contact between layers. Because of the material anisotropy induced by the manufacturing process, an inverse method was carried out to estimate the average stress-strain curves of the metallic layers used in the numerical simulations. The simulations are performed for two different configurations: one where the flexible pipe is composed only of the interlocked armor, and another considering interlocked armor and pressure armor. The adequacy of the numerical models is finally evaluated in light of experimental tests on flexible pipes with nominal internal diameters of 101.6 and 152.4 mm (4 and 6 in), proving excellent correlation.]]></description>
      <pubDate>Tue, 18 Aug 2026 14:11:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2706407</guid>
    </item>
    <item>
      <title>Statistical Analysis of Corrosion Losses of the Main Oil Pipelines Linear Part</title>
      <link>https://trid.trb.org/View/2579443</link>
      <description><![CDATA[The article analyzes the recurrence of defects and damages on main oil pipelines, in particular corrosion, and the problem of ensuring the reliability of steel structures and pipelines used in oil and gas complexes. It has been found that the most common and dangerous operational damage to the linear parts of main oil pipelines is corrosion damage, which has the character of cavities with random depth and length, as well as random placement along the length of the pipeline. A reliable means of detecting corrosion damage is in-line diagnostics, which allows measuring the depth of each corrosion cavity with high accuracy and determining its location along the pipeline length. The length of corrosion damage is represented as a random variable with a logarithmically normal distribution law, independent of the random variable of corrosion depth. The preferred probabilistic model is the sequence of maximum corrosion depth values, which provides a probabilistic description of corrosion damage sufficiently accurate for reliability assessment and is consistent with the accepted repair technology, in which entire sections of the pipeline are replaced in severely damaged areas. Using the statistical analysis, the corrosion effects were studied at each of the selected sites as a result of diagnostics inside the pipe, and attention is drawn to the need to research and improve scientific, technical and technological developments in the field of corrosion and mechanical resistance of metal structures. The results obtained are an important contribution to the development of technical and design measures to improve the efficiency and reliability of oil and gas complexes.]]></description>
      <pubDate>Mon, 27 Jul 2026 11:16:41 GMT</pubDate>
      <guid>https://trid.trb.org/View/2579443</guid>
    </item>
    <item>
      <title>The formation of bioavailable Hg in a pipeline: An initial investigation into Hg bioaccumulation resulting from oil and gas decommissioning</title>
      <link>https://trid.trb.org/View/2672175</link>
      <description><![CDATA[Mercury (Hg) released during offshore pipeline decommissioning may pose ecological risks, yet little is known about its chemical form and biological impact. We examined how Hg⁰ can react on and subsequently be released from laboratory-generated steel pipeline material to interact with marine algae, focusing on chemical speciation, transformation, and cellular uptake. Laboratory exposures of the marine algae, Isochrysis galbana, to pipeline-derived Hg showed accumulation up to 109 mg kg⁻¹ dry weight, as determined by cold vapour atomic fluorescence spectrometry (CV-AFS). Single-cell inductively coupled plasma mass spectrometry (SC ICP-MS) confirmed substantial cell-associated Hg, with bimodal distributions suggesting distinct uptake or surface-association pathways. Although classical growth and photosynthetic parameters did not consistently reveal toxicity, Hg exposure altered cell populations and aggregation behaviour, indicating sublethal but ecologically relevant effects. Our findings demonstrate that through interactions with pipeline material Hg⁰ can be transformed into species which have an increased likelihood of bioaccumulation.]]></description>
      <pubDate>Tue, 26 May 2026 11:56:22 GMT</pubDate>
      <guid>https://trid.trb.org/View/2672175</guid>
    </item>
    <item>
      <title>A novel PINN-STAN model for predicting corrosion rate in subsea oil and gas pipelines</title>
      <link>https://trid.trb.org/View/2656500</link>
      <description><![CDATA[This paper presents a novel Physics-Informed Neural Network with Spatio-Temporal Attention Network (PINN-STAN) for predicting corrosion rates in subsea oil and gas pipelines. Traditional data-driven models fail to capture multi-factor coupling effects and lack physical consistency under extreme conditions. PINN-STAN overcomes these limitations by embedding corrosion kinetics and mass conservation laws as trainable constraints. It employs a Bayesian spatio-temporal attention mechanism to dynamically adjust the importance of environmental factors and uses a 1D Convolutional Neural Network (CNN) for short-term events and Neural ODEs for long-term trends. Additionally, the model integrates a partial differential equation (PDE) to describe the physical evolution of corrosion processes, ensuring physical consistency by coupling the PDE solution with deep learning features. Experimental results on the CNOOC dataset demonstrate that PINN-STAN achieves a Root Mean Squared Error (RMSE) of 0.0165mm/a and a coefficient of determination (R²) of 0.9950, showcasing its superior predictive accuracy and stability under extreme conditions. Compared to traditional linear regression and machine learning models, PINN-STAN better incorporates physical constraints and multi-scale prediction capabilities, making it well-suited for corrosion assessment in multi-physical field coupling environments. This framework provides a reliable solution for corrosion prediction in complex subsea environments.]]></description>
      <pubDate>Mon, 13 Apr 2026 09:40:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2656500</guid>
    </item>
    <item>
      <title>Improved AHP and dual-hidden layer adaptive regularized neural network model for the quantitative risk assessment of oil and gas pipelines</title>
      <link>https://trid.trb.org/View/2634938</link>
      <description><![CDATA[Conventional risk assessment of oil and gas pipelines primarily relies on expert judgment, which suffers from limited generalizability, high uncertainty and subjectivity, frequently leading to imprecise outcomes. To address these challenges, this paper proposes a risk evaluation model (ImAHP-DHARN) for oil and gas pipelines based on an improved analytic hierarchy process (AHP) and a dual-hidden-layer adaptive regularized neural network (DHARN) model. First, a dynamic frequency-consequence matrix is employed to enable dynamic updates of risk values using five significant key factors: corrosion factor, structural defect, third party damage, natural disaster, and physical damage. Second, AHP is integrated with a comprehensive weighting framework to determine indicator weights, offering improved accuracy. Third, the DHARN model is developed, incorporating tansig-Softmax activation, L2 regularization, and early stopping to mitigate overfitting. Finally, Shapley Additive Explanations (SHAP) module is implemented to interpret model outputs and quantify the relative importance of indicators, providing insights into risk prioritization. The proposed ImAHP-DHARN model was validated using PHMSA pipeline accident data, achieving an accuracy of 98.3%, outperforming SVM (93.3%) and LVQ (98.0%). In Monte Carlo adversarial testing (e.g., earthquake scenarios superimposed with internal corrosion), the model exhibits a misjudgment rate of only 1.3%, underscoring its robustness. The SHAP analysis further highlights corrosion prevention T₃ as the highest-priority risk factor, providing valuable reference for pipeline risk management. In summary, the ImAHP-DHARN model significantly enhances the objectivity, generalization capability, robustness of pipeline risk assessment. Beyond this domain, this methodology can be extended to other complex industrial systems, providing a robust framework for risk assessment.]]></description>
      <pubDate>Mon, 23 Feb 2026 11:24:29 GMT</pubDate>
      <guid>https://trid.trb.org/View/2634938</guid>
    </item>
    <item>
      <title>Seismic Screening and Requalification of Marine Oil Terminals in California</title>
      <link>https://trid.trb.org/View/2263781</link>
      <description><![CDATA[There is an infrastructure of marine oil terminals in California, with an average age of over 50 years. Historically, there has been no consideration for the increased seismic risk and the possibility of massive amounts of oil to be spilled in California's ports or in San Francisco Bay. Most of these structures were built with little or no consideration for seismic loading and the current state of repair is generally poor. Additionally, the petroleum pipelines have never been analyzed for the displacement motion associated with the seismic demand. As these structures were inspected above the water line for many years by the California State Lands Commission (CSLC), operators/owners were in no rush to make any improvements. As a result of the 1994 Northridge Earthquake, a federal hazard mitigation grant was given to the CSLC to develop standards to mitigate future damage to port/harbor pile-supported structures. The code was completed in 2006, and is primarily for marine oil terminals, but is generally applicable to port/harbor pile supported wharves or piers. This new set of standards is Section 31F of the California Building Code and is now enforceable. An initial group of 10 "high risk" terminals have submitted their initial audits, documenting the structural condition, above and below the water line, along with plans for seismic rehabilitation. A fitness-for-purpose criterion not only provides inspection results, but also provides a structural assessment, as to whether or not the facility will meet the performance standard of two levels of seismic demand. These initial 10 terminals are in high seismic zones, and the seismic demand for a Level 1 earthquake is 50% probability of excedance in 50 years, and for a Level 2, the standard is a 10% probability of excedance in 50 years. This paper will provide summary results of some of these terminals, providing a general description of how they were modified to meet the seismic criteria, with minimum down-time.]]></description>
      <pubDate>Fri, 06 Feb 2026 13:53:32 GMT</pubDate>
      <guid>https://trid.trb.org/View/2263781</guid>
    </item>
    <item>
      <title>Analyses of the mechanical characteristics of long-distance crude oil pipeline support assembly on “L”-type pipeline</title>
      <link>https://trid.trb.org/View/2627187</link>
      <description><![CDATA[Trenched crude oil pipelines rely on anchor blocks and pipe supports for stability, however mechanical interaction remains unclear, limiting design safety. This study employs fluid-solid-thermal coupling methods to establish a three-dimensional finite element calculation model of an “L”-type crude oil pipeline based on an equivalent model of the end-side displacement of the anchor block and pipe section. The mechanical response patterns of anchor blocks and pipe supports to the isolated action of “L"-type pipes were investigated separately. After evaluating the optimal design position of pipe supports to balance internal forces within the pipe, we further investigated the synergistic optimisation mechanism of anchor blocks and pipe supports on the mechanical characteristics of the pipe. Results indicate that the anchor block provides limited protection to the structural safety of pipeline. Under 300 mm end-side displacement imposed by anchor block, the maximum deformation of elbow decreases by 2.31 mm. However, increasing the end-side displacement from 500 mm to 600 mm reduces the maximum deformation by only 0.05 %. A symmetrical distribution of pipe supports delivers the optimal solution, lowering the maximum stress of the pipeline to 78.049 MPa, the 25.55 % reduction. Additionally, the combined use of anchor blocks and pipe supports demonstrates a synergistic optimisation effect on pipeline safety, reducing elbow deformation by approximately 85.14 % (to 33.121 mm). These findings provide a theoretical basis for optimising the working conditions of pipeline support assemblies in long-distance crude oil pipeline networks.]]></description>
      <pubDate>Thu, 29 Jan 2026 17:01:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2627187</guid>
    </item>
    <item>
      <title>Long-term performance and optimization of two-phase closed thermosyphons for buried large-scale warm-oil pipelines under rising oil temperature</title>
      <link>https://trid.trb.org/View/2597155</link>
      <description><![CDATA[The China–Russia crude oil pipeline (CRCOP) serves as a critical energy supply route for both countries. The pipeline during its operation inevitably dissipates heat to the surrounding permafrost, leading to thaw settlement issues. A full-scale monitoring system has been deployed in key areas affected by thaw settlement along the CRCOP to guarantee the safe operation of the pipeline. This system provides real-time alerts and verifies the cooling mechanism of the vertical two-phase closed thermosyphons (TPCTs) used for mitigating thaw settlement. However, on-site monitoring results indicate that the TPCTs alone cannot maintain the permafrost temperature in the vicinity of the CRCOP, suggesting the need for optimized TPCT arrangements. This study proposes three design schemes to optimize the configuration of TPCTs based on in-situ observations. Numerical simulation results indicate that the composite approach — combining TPCTs with a thermal insulation layer — performs better than the standalone TPCT configuration. Using this composite solution, the artificial permafrost table (APT) can be stably maintained at approximately 2.5 m depth. This strategy is recommended for engineering applications in permafrost environments. Furthermore, the findings of this study provide valuable guidance for the CRCOP and other similar pipeline projects in permafrost regions.]]></description>
      <pubDate>Fri, 07 Nov 2025 11:30:15 GMT</pubDate>
      <guid>https://trid.trb.org/View/2597155</guid>
    </item>
    <item>
      <title>Identifying Hotspots in Natural Gas and Oil Pipeline Networks and Understanding Incident Risk Factors: An Integrated Spatial Analysis and Machine Learning Approach</title>
      <link>https://trid.trb.org/View/2578860</link>
      <description><![CDATA[Pipeline networks are crucial for transporting natural gas and oil but are vulnerable to significant safety and environmental risks. Previous studies have primarily focused on identifying overall incident causes without examining how spatial variations contribute to incident risk and have largely overlooked the differences in risk factors between natural gas and oil pipelines pipeline networks. A few studies often address broader geographic regions, such as states or counties, overlooking intricate spatial dynamics within pipeline networks. This study aims to bridge this gap by identifying hotspot areas within pipeline networks, areas with a high concentration of incidents compared to surroundings, using incident data from the Pipeline and Hazardous Materials Safety Administration and pipeline network data from the US Energy Information Administration. Leveraging spatial analysis techniques such as global Moran’s 𝐼, Getis-Ord Gi*, Anselin local Moran’s 𝐼, and spatiotemporal analysis, this study identified statistically significant pipeline hot spots, prominently in California and Ohio. After identifying hot spots using spatial analysis, different machine learning algorithms, such as random forest (RF), were employed to assess the factors contributing to these hotspots. The RF model achieved the highest accuracy, with 82% on the testing data for natural gas pipelines and 94% for oil pipelines, with standard deviations of 0.010 and 0.012, respectively, from K-fold cross validation to ensure robustness and accuracy. The analysis revealed distinct geographical variations driven by spatial and operational factors, such as pipeline intersections and operating pressure. Natural gas pipelines are more prone to operational errors, while oil pipelines are more susceptible to corrosion. This study enhances the detection of localized incident clusters, enabling more effective interventions compared to broader area-based analyses.]]></description>
      <pubDate>Fri, 24 Oct 2025 16:53:54 GMT</pubDate>
      <guid>https://trid.trb.org/View/2578860</guid>
    </item>
    <item>
      <title>Numerical Analysis and Structural Optimization of a Jet Pig Using a Rifling Structure</title>
      <link>https://trid.trb.org/View/2587129</link>
      <description><![CDATA[Since pigging technology is commonly used to reduce blockage or clogging during the pipeline transportation process, jet pig optimization is essential. This paper proposed a method for incorporating a rifling structure into the inner jet hole wall of a jet pig. Two types of jet pigs with differing jet hole configurations were created using modeling software, one with rifling and one without. The computational fluid dynamics (CFD) method was utilized for the analysis of the flow distribution changes in crude oil pipeline pigs. The contour maps and change curves of the fluid velocity, pressure, and turbulent kinetic energy (TKE) of the two types of pigs were obtained at different bypass fractions and inlet flow velocities. The results demonstrate that a constant inlet fluid velocity and a higher bypass fraction decrease the flow velocity and TKE of the pigs of two types, while the differential pressure values across the pigs decline gradually. The flow velocity, pressure difference, and TKE of the rifling pig exceed those of the nonrifling pig. When the bypass fraction is set at 3%, the rifling pig demonstrates about a 3.4% increase in exit flow velocity compared to its nonrifling pig, while the TKE at the outlet increases by over 13.5%. This indicates that using a rifling structure on the inner jet hole wall of the jet pig improves its dispersal ability.]]></description>
      <pubDate>Fri, 24 Oct 2025 16:53:54 GMT</pubDate>
      <guid>https://trid.trb.org/View/2587129</guid>
    </item>
    <item>
      <title>The application of deep learning in pipeline inspection: current status and challenges</title>
      <link>https://trid.trb.org/View/2582679</link>
      <description><![CDATA[Pipelines are a primary route of transportation for essential energy sources such as oil and gas, and the inspection and assessment of their operational status is vital to the health of the industry. In addition to traditional manual inspection techniques, emerging Deep Learning (DL) methods have promoted the development of intelligent pipeline inspection. Using DL techniques, oil and gas pipelines can be automatically, efficiently and accurately inspected and evaluated, which is important for improving pipeline safety and reducing accident risks. This paper reviews the application of DL to damage detection, identification and classification of pipelines. Firstly, a review of commonly used DL methods is given, and then the main application scenarios of current DL in pipeline inspection are discussed. Finally, the advantages and limitations of the existing detection methods are given.]]></description>
      <pubDate>Fri, 17 Oct 2025 16:49:29 GMT</pubDate>
      <guid>https://trid.trb.org/View/2582679</guid>
    </item>
    <item>
      <title>Properties of ground surface temperature within the Engineering Corridor traversing the Xing'an permafrost region in Northern Northeast China</title>
      <link>https://trid.trb.org/View/2590298</link>
      <description><![CDATA[Permafrost degradation emerges as a critical threat to sustainable infrastructure development in cold regions. Engineering activities across Northern Northeast China have been shown to modify ground surface temperature (GST), thereby profoundly altering the thermal stability of underlying permafrost layers. To this end, GSTs at different latitudes were investigated at sites along the Engineering Corridor covering the National Highway from Jiagedaqi to Mohe and China-Russia crude oil pipelines. Electrical resistivity tomography was also conducted to examine permafrost distribution and influencing factors. Results indicated that adverse snows on the surface facilitated higher GSTs associated with frost numbers less than 0.5, further accelerating Xing'an permafrost degradation. The insulation provided by snow layers contributed to progressively warmer temperatures at northern latitudes, enhancing the mean annual GST from 2.28 °C to 7.08 °C and advancing the timing of the coldest temperatures. Heat absorption in the gravel land in summer and heat preservation by snow cover in winter promoted the talik in the ambient topsoil and surrounding pipelines. Snowfall intensified the surface offset, with the highest recorded in January (25.1 °C) and the lowest in April (2.2 °C) in gravel lands, consistently demonstrating a positive monthly difference. Additionally, accumulated water significantly increased the thawing depth and lowered the load-bearing capacity in slopes, further triggering cracks and tilting. These findings indicated that long-term monitoring should be carried out to determine the spatiotemporal variations of GST for validating complex permafrost distribution and facilitating a better understanding of infrastructure-induced permafrost response and consequent degradation.]]></description>
      <pubDate>Thu, 16 Oct 2025 17:02:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2590298</guid>
    </item>
    <item>
      <title>A performance-based NaTech risk assessment methodology for hydrocarbon pipelines subjected to landslides</title>
      <link>https://trid.trb.org/View/2602403</link>
      <description><![CDATA[Distributed gas and oil pipelines often cross areas prone to landslides. According to the Eu Directive 2012/18/UE, known also as Seveso III, the risk to people and community served by the pipeline should be assessed. If the need is clear, the state of the art on available methods to solve this peculiar problem is limited. Methods to calculate NaTech risk according to a performance-based approach have already been proposed, especially for hazardous plants like refineries and oil & gas plant. Differently, a few similar approaches have been used for estimating the risk of consequences generated by a hydrocarbon material release from a pipeline subjected to landslide. In this paper a performance-based methodology to assess the mean annual frequency of release from pipelines for the transportation of hydrocarbons subjected to a landslide is proposed. This methodology employs the hazard curves of landslides, and the fragility curves of the pipelines defined in terms of permanent ground deformation. The content release is evaluated by using mechanical response quantities as local deformation as suggested in the literature. The proposed approach is validated through a representative case study of a pipeline for the transportation of oil crossing in a mountain region.]]></description>
      <pubDate>Mon, 13 Oct 2025 08:48:01 GMT</pubDate>
      <guid>https://trid.trb.org/View/2602403</guid>
    </item>
    <item>
      <title>Rapid Prediction Models for Oil Temperature and Surrounding Environment Temperature Fields in Buried Hot Crude Oil Pipelines</title>
      <link>https://trid.trb.org/View/2552267</link>
      <description><![CDATA[Wax-rich or high-viscosity crude oil requires heating for long-distance transportation in buried pipelines. Monitoring the temperature distributions and variations of both oil and surrounding environment is crucial to ensuring safety and economic efficiency. This study proposes two models for rapidly predicting the steady-state oil temperature and surrounding environment temperature fields based on the Fourier neural operator (FNO) network and U-shaped network (UNet), respectively. These models leverage numerical results as training data, incorporating boundary conditions and environment grid coordinates at the pipeline cross section as inputs to predict the temperature distributions of both oil and the surrounding environment along the buried pipeline. With optimized hyperparameters, the models achieve accurate and efficient predictions. The FNO and UNet models had average RMS errors (RMSEs) in environment temperature field prediction of 2.68×10⁻³ and 5.49×10⁻³ at the pipeline cross section, respectively. For oil temperature predictions, the FNO model had an average relative error of 1.49×10⁻⁴, compared with 2.55×10⁻⁴ for the UNet model, with average absolute error values of 5.32×10⁻³ and 7.24×10⁻³, respectively. Moreover, both models exhibited strong generalization, with an average RMSE in the environment temperature field prediction of less than 3.5×10⁻² and an average relative error in oil temperature predictions of less than 2.1×10⁻³ across different data sets. Comparatively, the FNO and UNet model had slightly higher prediction accuracy than the UNet model. In terms of computational efficiency for a 100-km pipeline, these models offer improvements of at least 116.25× over the numerical simulation method, with a maximum improvement of 1,775.03× as the number of simultaneously predicted pipeline cross sections increases.]]></description>
      <pubDate>Wed, 24 Sep 2025 15:24:34 GMT</pubDate>
      <guid>https://trid.trb.org/View/2552267</guid>
    </item>
  </channel>
</rss>