<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>Frequency domain viscoelastic analysis of thermoplastic composite pipes under axisymmetric loading</title>
      <link>https://trid.trb.org/View/2710083</link>
      <description><![CDATA[Thermoplastic Composite Pipes (TCP) have emerged as an alternative for deepwater oil & gas transportation due to their high strength, excellent fatigue life, and corrosion resistance. However, TCP exhibit creep and stress relaxation behaviors, which are significantly influenced by temperature. In this work, the TCP viscoelastic mathematical model is presented by combining the laminate periodic microstructure model and the macroscale pipe model, and a general solution is given for the frequency domain viscoelastic analysis of TCP under axisymmetric loadings. The methodology begins with analyzing the steady-state temperature distribution and establishing an axisymmetric mechanical model. The laminate stiffness tensor is derived based on the fiber and matrix properties, and the viscoelastic behavior of the polymer and laminate layers employs Maxwell elements and Prony series with temperature shift factor. Furthermore, the pipe tension-axial strain, tension-torsion hysteresis loops, equivalent axial stiffness and loss factor under harmonic loading at various frequencies are analyzed. The results show that smaller loading frequencies yield larger hysteresis loop areas and larger maximum and minimum strains. Equivalent axial stiffness increases with increasing frequency, and higher temperature leads to lower equivalent axial stiffness. The loss factors are affected by the frequency and temperature simultaneously.]]></description>
      <pubDate>Tue, 09 Jun 2026 14:36:49 GMT</pubDate>
      <guid>https://trid.trb.org/View/2710083</guid>
    </item>
    <item>
      <title>An advanced frequency-domain approach for stochastic response analysis of floating bridges subjected to inhomogeneous wave loads</title>
      <link>https://trid.trb.org/View/2676412</link>
      <description><![CDATA[Floating bridges are large, slender structures supported by floating foundations, providing a cost-effective solution for crossing wide and deep waters. In code-based design, response analysis faces challenges due to hydroelastic effects, multi-body hydrodynamic interactions, and computational efforts. This study proposes a frequency-domain hydroelastic analysis framework for floating bridges under coastal wave fields characterized by spatial inhomogeneity, low coherence and short-crestedness. A spectral density matrix approach is developed to rigorously model wave energy distribution, coherence, and directional spreading, integrated with first- and second-order wave excitation formulations through linear and mean-drift force transfer functions. Extreme values of the girder internal loads are further estimated applying a Gaussian-based analytical method. The approach, validated against time-domain simulations, is applied to a conceptual Bjørnafjord crossing case study. Based on field measurement of the wave condition, a sensitivity study is conducted to investigate the effects of each individual factor governing the inhomogeneity, including wave spectral parameters and coherence functions. Specific suggestions are proposed for the determination of wave conditions, during practical applications when the relevant data are insufficient to accurately model the inhomogeneity. It is highlighted that a homogeneous assumption with the largest significant wave height yields conservative bending moments, but still underestimates the extreme axial force at certain locations, which is crucial for the ultimate buckling design. The proposed frequency-domain approach demonstrates a substantial computational efficiency advantage over nonlinear time-domain simulations, reducing the overall analysis time from several hours to a few minutes for a complete extreme response evaluation. The study also shows the need for information about site-specific wave inhomogeneity and coherence modeling for the safe design of floating bridges.]]></description>
      <pubDate>Mon, 08 Jun 2026 08:38:12 GMT</pubDate>
      <guid>https://trid.trb.org/View/2676412</guid>
    </item>
    <item>
      <title>Multi-Sea-State Joint Frequency-Domain Identification of hydrodynamic parameters for float-over vessels</title>
      <link>https://trid.trb.org/View/2706862</link>
      <description><![CDATA[Identification of hydrodynamic parameters is critical for the safety of float-over installations but is often compromised by uncertainties in traditional methods. This study proposes a Multi-Sea-State Joint Frequency-Domain Identification (MSS-JFI) framework. By formulating the problem in the frequency domain, the framework leverages a joint optimization strategy to integrate data from varying wave conditions, effectively resolving the parameter coupling ambiguities inherent in single-sea-state identification. Physical consistency is ensured by parameterizing radiation forces via State-Space Models (SSM) to enforce Kramers–Kronig relations, supplemented by a frequency masking mechanism to enhance noise robustness. Pitch motion is selected as a representative proof-of-concept to validate the method against complex frequency-dependent radiation and stochastic excitation. Numerical validations demonstrate high accuracy, achieving relative errors below 3% for key scalar parameters. Furthermore, a full-scale experiment on the semi-submersible heavy-lift vessel Xiang He Kou confirms the engineering applicability of the framework, yielding response predictions with quantified uncertainty, where the in-situ measurements are well contained within the predicted 95% empirical prediction intervals.]]></description>
      <pubDate>Mon, 01 Jun 2026 09:14:40 GMT</pubDate>
      <guid>https://trid.trb.org/View/2706862</guid>
    </item>
    <item>
      <title>Numerical prediction of wave loads and motions for ships with forward speed by a frequency-domain harmonic polynomial method</title>
      <link>https://trid.trb.org/View/2707655</link>
      <description><![CDATA[This paper presents a three-dimensional frequency-domain Harmonic Polynomial Method (HPM) for predicting wave loads and motions of ships advancing with constant forward speed. The method extends the two-dimensional high-order harmonic polynomial approach for the Laplace equation, originally proposed by Wang et al. (2020), to handle complex three-dimensional hydrodynamic problems involving ship-wave interactions. The framework accounts for the influence of steady waves generated by a ship moving in calm water, which impacts unsteady wave loads. Within the potential flow theory, the method solves the Laplace equation using a locally refined Cartesian grid system composed of three components: hull-surface grids, free-surface grids, and flow-field grids. To achieve stable and accurate implementation of boundary conditions and discretization of the Laplace equation near complex boundaries, irregular cells are employed. The numerical approach is applied to compute radiation wave loads, wave excitation forces, and resultant ship motions expressed via Response Amplitude Operators (RAOs). Comprehensive verification and validation are performed through comparisons with benchmark experimental data (including the Wigley, Series 60 and S175 hull forms) as well as established numerical solutions. The results demonstrate that the present method can efficiently and accurately predict wave-induced loads and motions for ships with forward speed.]]></description>
      <pubDate>Fri, 29 May 2026 15:37:09 GMT</pubDate>
      <guid>https://trid.trb.org/View/2707655</guid>
    </item>
    <item>
      <title>Quantification of Tire-Pavement Interaction Noise Using Frequency Analysis</title>
      <link>https://trid.trb.org/View/2668441</link>
      <description><![CDATA[Tire-pavement interaction noise (TPIN) significantly contributes to overall vehicular noise, particularly at higher speeds, as it arises from the dynamic interaction between vehicle tires and road surfaces. Accurately quantifying TPIN is critical for developing noise mitigation strategies, and researchers have employed various measurement techniques to assess it under different conditions and pavement types. This study proposes a frequency analysis-based approach for TPIN quantification and compares it with the widely used logarithmic subtraction method. Controlled pass-by tests were conducted on both asphalt and cement concrete pavements across multiple speeds and frequency ranges. Results show that TPIN increases with speed for all pavement types, and frequency-specific crossovers between engine noise and TPIN are clearly observed using the frequency method but not in the subtraction method. Furthermore, the subtraction method consistently overestimates TPIN, as it does not isolate frequency-dependent noise characteristics. To address this limitation, empirical relationships have been developed to estimate frequency-based TPIN using values obtained from the subtraction method, enabling more accurate predictions in scenarios where frequency analysis tools are unavailable. The findings offer valuable insights into pavement acoustic performance and provide a foundation for more effective TPIN reduction measures and improved transportation noise management practices.]]></description>
      <pubDate>Tue, 26 May 2026 09:40:59 GMT</pubDate>
      <guid>https://trid.trb.org/View/2668441</guid>
    </item>
    <item>
      <title>The Impact of Fluctuations in Global Energy Prices on Maritime Transportation: A Frequency-Based Approach</title>
      <link>https://trid.trb.org/View/2698314</link>
      <description><![CDATA[This study examines the volatility spillovers between oil prices (WTI, Brent, Dubai) and shipping indices (BDI, BCTI, BDTI) from January 02, 2015, to July 31, 2024. Using the methods of Diebold and Yilmaz (2012) and Baruník and Krehlík (2018), the volatility relationships between energy and shipping markets are analysed in directional and frequency terms. The findings show that oil prices have a strong volatility effect on shipping indices, which is particularly pronounced in the long run. The long-term impact of Brent oil prices on shipping indices suggests that shocks in energy markets can have lasting effects on global transportation costs. Moreover, global events, such as the COVID-19 pandemic and the Russia-Ukraine war, have further amplified the spillover of energy market volatility to shipping indices. The study results emphasise the need for firms operating in energy and shipping markets to develop stronger risk management strategies against volatility. Future research should examine the effects of larger data sets and macroeconomic factors on these relationships. © 2026, Faculty of Maritime Studies. All rights reserved.]]></description>
      <pubDate>Wed, 20 May 2026 09:10:38 GMT</pubDate>
      <guid>https://trid.trb.org/View/2698314</guid>
    </item>
    <item>
      <title>Comparison of time and frequency domain analysis of propeller loading noise</title>
      <link>https://trid.trb.org/View/2700550</link>
      <description><![CDATA[This study presents a unified analytical framework for investigating the acoustic pressure generated by rotating point forces, with particular emphasis on comparing time and frequency domain formulations. While both approaches yield consistent results, each offers distinct benefits: the time domain provides intuitive insight into the contributions of steady and unsteady loading, whereas the frequency domain facilitates modal decomposition and simplifies the analysis of periodic noise. The model represents uniformly spaced blades as point forces with axial, drag, and radial components. The force magnitude varies in time according to a Gaussian pulse, enabling investigation into the influence of time scale on propeller noise. This formulation is relevant to scenarios such as rotor systems experiencing periodic loading or isolated gust encounters. Both periodic and aperiodic cases are analysed in the time and frequency domains, allowing conditions for constructive and destructive inter-blade interference to be identified. The study also examines the advantages of each formulation. Time domain analysis provides a more intuitive interpretation of aperiodic phenomena but requires careful treatment of retarded time effects. In contrast, the frequency domain is well-suited to periodic problems and naturally aligns with modal acoustic models, though it may obscure certain inter-blade interaction mechanisms. Collectively, these insights advance the understanding of rotor noise generation and its spatial characteristics in installed configurations.]]></description>
      <pubDate>Wed, 20 May 2026 09:10:37 GMT</pubDate>
      <guid>https://trid.trb.org/View/2700550</guid>
    </item>
    <item>
      <title>A semantic‐enhanced transformer with adaptive fusion for road damage detection</title>
      <link>https://trid.trb.org/View/2646693</link>
      <description><![CDATA[Road damage detection faces significant challenges including extreme scale variations, complex visual interference from road textures, diverse orientational patterns, and irregular boundaries. This paper proposes a semantic-enhanced and adaptive fusion detection transformer to address these domain-specific challenges through two synergistic innovations. The semantic enhancement attention module exploits distinctive frequency-domain characteristics of road damages through learnable spectral processing, where damaged regions exhibit 50.5% higher high-frequency energy, compared to intact surfaces, enabling effective discrimination between structural defects and background interference. The adaptive information fusion module implements a three-stage progressive architecture: loss-less transmission establishes information integrity across extreme scales through amplitude-aware upsampling and attention-driven fusion; omnidirectional pattern capture via multi-directional convolutions addresses diverse damage orientations; dual-path processing optimizes computational efficiency. Comprehensive evaluation across four datasets demonstrates state-of-the-art performance with significant improvements: 83.4% mean average precision at intersection over union threshold 0.5 on UAV-PDD2023 (+3.4% over previous best), 31.2% on CNRDD (+1.3%), 61.9% on RDD2020 (+3.0%), and 90.2% on nighttime NPD (+0.6%), while achieving superior efficiency with 62 giga floating-point operations, 20 million parameters, and 51 frames per second inference speed for real-time processing.]]></description>
      <pubDate>Mon, 18 May 2026 16:36:42 GMT</pubDate>
      <guid>https://trid.trb.org/View/2646693</guid>
    </item>
    <item>
      <title>Stochastic Seismic Analysis of Structures under Multidimensional and Multisupport Ground Motions Using an FFT-Based Absolute Displacement Method</title>
      <link>https://trid.trb.org/View/2582540</link>
      <description><![CDATA[Due to the spatial effect in seismic wave propagation, it is necessary to consider the nonconsistency of ground motion in seismic analysis of long-span structures. A highly efficient frequency-domain method was developed for addressing the response power spectral analysis of long-span structures under multidimensional and multisupport ground motions by the absolute displacement method and Fourier analysis. The explicit formulations of response power spectrum of structures are first derived based on the absolute displacement method. Then, the unit impulse response of structures with respect to the ground acceleration is solved by a few transient dynamic analyses, the corresponding number of which is much less than that required by the traditional method, which is the pseudo-excitation method (PEM). Finally, the explicit formulations are calculated by fast Fourier transform (FFT) or inverse fast Fourier transform, which can also execute dimension-reduced analysis to only focus on certain responses of interest. Therefore, the proposed method possesses a considerably high efficiency. Furthermore, it can account for stationary and nonstationary excitations with arbitrary forms of power spectrum, even of the nonseparable one. Numerical examples are given to study the stochastic seismic response of a suspension bridge considering the spatial effect of ground motion, and the correctness and effectiveness of the proposed method are verified by comparison with the traditional PEM.]]></description>
      <pubDate>Tue, 28 Apr 2026 11:20:01 GMT</pubDate>
      <guid>https://trid.trb.org/View/2582540</guid>
    </item>
    <item>
      <title>Self-Interaction Dynamic Graph Convolutional Network With Multiscale Time-Frequency Fusion for Vehicle Trajectory Prediction</title>
      <link>https://trid.trb.org/View/2617704</link>
      <description><![CDATA[The ability of autonomous vehicles to accurately predict future trajectories is crucial for ensuring safe driving in complex traffic environments. However, current methods exhibit limitations in capturing the dynamic dependencies inherent in vehicle social interactions and lack effective integration of multiscale time-frequency features. To address these issues, we propose a self-interactive dynamic graph convolutional network with multiscale time-frequency fusion. It introduces a self-interaction graph convolution module designed to capture dynamic high order features of social interactions and employs a hybrid attention mechanism to fuse multiscale time-frequency features, thereby enhancing the representation of vehicle trajectories. In addition, the spatiotemporal attention module is adopted to achieve dynamic weight update of key information focusing on the interaction of the target vehicle with surrounding vehicles. Finally, a comparison and analysis of the proposed method was conducted based on datasets of NGSIM and HighD. The findings suggest that the suggested model demonstrates enhanced accuracy in trajectory prediction in comparison to leading techniques, achieving average root mean square errors of 1.54 m and 0.69 m across the two datasets when predicting for a duration of 5 seconds,and the average ADE and FDE indicators were 17.33% and 26.61% higher than the best performance.]]></description>
      <pubDate>Wed, 25 Mar 2026 17:11:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617704</guid>
    </item>
    <item>
      <title>Driving Behavior Classification Method Based on Fourier Transform Multimodal Fusion</title>
      <link>https://trid.trb.org/View/2617697</link>
      <description><![CDATA[Driving behavior classification is an important component of advanced driver assistance system (ADAS) and plays a pivotal role in enhancing driving safety and economy. Existing driving behavior classification methods primarily rely on time-domain features or multimodal fusion techniques. However, these methods often involve many parameters and complex training processes, making them difficult to be directly applied in real-world scenarios. This study proposes a lightweight driving behavior classification method that integrates both time-domain and frequency-domain feature. The proposed method comprises two parallel branches. The first branch extracts time-domain features using a combination of Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory Network (BiLSTM). Instead of the conventional fully connected layers, the Kolmogorov-Arnold Network (KAN) is employed to reduce modeling complexity and enhance performance. The second branch captures frequency-domain features through the Discrete Fourier Transform (DFT) method. An adaptive filtering block removes high-frequency noise, and the frequency-domain features are then fused adaptively using two learnable filters—global and denoising. The time-frequency domain multimodal feature fusion module performs the final weighted fusion of these features. Extensive experiments are conducted to evaluate the proposed method based on the UAH-DriveSet. Experimental results show that our method achieves an F1 score of 98.74%, surpassing the current state-of-the-art. Moreover, to assess the robustness of our method, experiments are conducted on the Ford Stay Alert Challenge dataset, obtaining an F1 score of 97.83%. Furthermore, our method is highly efficient, with only 3.97M parameters and 0.38G FLOPS, significantly lower than existing methods. The inference speed reaches an impressive 1,257 FPS, meeting the real-time requirements in resource-constrained environments.]]></description>
      <pubDate>Wed, 25 Mar 2026 17:11:07 GMT</pubDate>
      <guid>https://trid.trb.org/View/2617697</guid>
    </item>
    <item>
      <title>ABCDWaveNet: Advancing Robust Road Ponding Detection in Fog Through Dynamic Frequency-Spatial Synergy</title>
      <link>https://trid.trb.org/View/2679326</link>
      <description><![CDATA[Road ponding presents a substantial threat to vehicular safety, particularly in foggy conditions where reliable detection continues to be a major challenge for advanced driver assistance systems (ADASs). To address this issue, we propose an aggregation-broadcast-coupling dynamic wavelet network (ABCDWaveNet), a novel deep learning framework specifically designed to achieve robust ponding detection in fog-affected environments. The central architecture of ABCDWaveNet improves detection performance by utilizing dynamic convolution for adaptive feature extraction under reduced visibility, together with a wavelet-based module that improves feature representation across both spatial and frequency domains, thereby effectively alleviating fog-related interference. In addition, ABCDWaveNet incorporates multi-scale structural and contextual information and employs an adaptive attention coupling gate to dynamically integrate global and local features, leading to improved detection accuracy. For realistic evaluations under compounded adverse weather conditions, we introduce the Foggy Low-Light Puddle dataset. Comprehensive experiments confirmed that ABCDWaveNet attained state-of-the-art results, with notable intersection over union gains of 3.51%, 1.75%, and 1.03% on the Foggy-Puddle, Puddle-1000, and Foggy Low-Light Puddle datasets, respectively. Furthermore, with an inference speed (FPS) of 25.48 on the NVIDIA Jetson AGX Orin, the proposed framework demonstrates strong suitability for development in ADAS applications. These results highlight the effectiveness of ABCDWaveNet, presenting valuable advancements for proactive road safety under challenging weather conditions.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:56 GMT</pubDate>
      <guid>https://trid.trb.org/View/2679326</guid>
    </item>
    <item>
      <title>Extremely short-term ship motion prediction using real-sea motion datasets based on the VMD-LSTM-FDR network</title>
      <link>https://trid.trb.org/View/2676538</link>
      <description><![CDATA[Ship motion in real sea is inherently random and uncertain, significantly impacting operational safety. The study proposes an extremely short-term ship motion prediction model integrating Variational Mode Decomposition (VMD), LSTM networks, and a Frequency Domain Rectification (FDR) algorithm. The motion time series is decomposed into intrinsic mode functions (IMFs) using VMD. Then, these IMFs are used as the input of a multi-input multi-output LSTM network for extremely short-term prediction. In this process, the FDR algorithm is introduced to rectify amplitude predictions. Moreover, the optimal advance prediction time is discussed. The prediction model applied to sea-trial motion data achieves 80%-90% accuracy in extremely short-term motion prediction. To verify prediction performance of the prediction model in different types of ships, it is applied to ships of different tonnages: an offshore supply ship (20900t), a crane ship (51000t), and a fishing vessel (50t) operating in nearshore or open-sea area. The extremely short-term motion prediction achieved an average accuracy exceeding 80%. It is found that the prediction accuracy rate in nearshore area is generally higher than that in open-sea area. Increasing the data sampling rate is an effective approach to enhance the accuracy of extremely short-term ship motion prediction, while higher sampling rate results in larger training efforts. Prediction performance is affected by the length of training datasets, with training dataset length at ∼10² timesteps balancing accuracy and computational cost. The model has good generalizability for the same ship under different conditions, while its performance drops sharply when directly applied without pretraining to different ships.]]></description>
      <pubDate>Wed, 25 Mar 2026 11:44:53 GMT</pubDate>
      <guid>https://trid.trb.org/View/2676538</guid>
    </item>
    <item>
      <title>Impact of floating solar island size on wave loads and frequency domain responses</title>
      <link>https://trid.trb.org/View/2637915</link>
      <description><![CDATA[Floating Photovoltaic (fpv) has widespread use in inland water bodies and it was only recently that small fpv island demonstrators were deployed offshore. The mooring system is a key design aspect of fpv islands, which typically feature thousands of interconnected modules that can be supported by floats or pontoons. Numerical modelling of design loads and responses for large pontoon-type fpv islands is to date unattainable with existing software, and early-stage mooring design is generally limited to analytical models that neglect wave dynamics. One of the main challenges remains the efficient evaluation of hydrodynamic interactions and connection forces with reduced computational burden. In this study, we implement the hierarchical interaction theory by Kashiwagi (2000) to expedite calculation of potential-flow hydrodynamic loads, which we combine with steady wind loads and hinge constraints in a frequency domain model of a fpv island. We use this model to evaluate wave drift loads and mooring fairlead responses for varying fpv island sizes, ranging from a few hundred to thousands of floats. We assess the evaluation of horizontal drift forces with approximate methods, including Maruo’s formula (Maruo, 1960) with full wave reflection, and we show that drift force estimates based on isolated modules may be non-conservative. As in previous studies, we find that responses are overestimated when hydrodynamic interactions among floats are neglected. We show that the maximum island size for which the minimum breaking load of the mooring line is not exceeded is significantly dependent on the modelling approach.]]></description>
      <pubDate>Tue, 10 Mar 2026 09:55:43 GMT</pubDate>
      <guid>https://trid.trb.org/View/2637915</guid>
    </item>
    <item>
      <title>Iris time-frequency map visual feature-based cluster matching: A universal domain adaptation method for propulsion shafting fault diagnosis</title>
      <link>https://trid.trb.org/View/2664969</link>
      <description><![CDATA[The propulsion shafting is a vital component of ship power systems. Timely and accurate fault diagnosis is essential for ensuring navigational safety. Domain adaptation techniques have been widely applied in intelligent fault diagnosis. However, most existing methods overlook the critical impact of input representation quality on diagnostic performance and are confined to specific domain adaptation scenarios. In practical engineering, the label space relationships between domains are often unavailable, limiting the applicability of these methods. To address these issues, this study proposes a universal domain adaptation (UniDA) method, termed the source domain category anchor-guided cluster matching network. Specifically, the network utilizes iris time-frequency maps as input, which enhances the readability of the information. A similarity criterion is formulated to cluster features of the same type, subsequently matching them to the corresponding category anchors. Moreover, an inter-class representation decoupling constraint is designed to shape a more globally discriminative feature space. Further, a distance-based detection strategy is proposed to build reliable decision boundaries between common and private categories. Experimental results on the propulsion shafting dataset validate the effectiveness of the proposed method in handling diagnostic tasks involving domain and category shifts, outperforming other state-of-the-art methods. Additionally, visualization via gradient-weighted class activation mapping indicates that the network's decision-making is grounded in physically meaningful evidence, revealing the complementarity between interpretability and transferability.]]></description>
      <pubDate>Mon, 09 Feb 2026 08:42:26 GMT</pubDate>
      <guid>https://trid.trb.org/View/2664969</guid>
    </item>
  </channel>
</rss>