2026-07-13T10:43:48+03:30 http://ijmt.ir/browse.php?mag_id=50&slc_lang=en&sid=1

50-898 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 Zobeir Raisi zobeir.raisi@cmu.ac.ir Esmaeil Sarani sarani@cmu.ac.ir Rasoul Damani damani@cmu.ac.ir Valimohammad Nazarzehi Had v.nazarzehi@cmu.ac.ir Maritime vessel detection in satellite imagery is essential for coastal monitoring, traffic regulation, and maritime security. Vessels along the southeastern coast of Iran exhibit unique structural and geometric characteristics; they are small and overlapped, differing substantially from international benchmarks. Detecting small and overlapping vessels presents additional challenges due to the loss of fine-grained features and ambiguous object boundaries in conventional deep learning pipelines. Consequently, existing pre-trained models, trained primarily on global datasets, often fail to generalize effectively to this region. To address this, in our study, we provide the first systematic investigation of ship detection for southeastern Iran, supported by a curated dataset of high-resolution satellite imagery from its major ports. We, then, propose a flexible Hybrid Attention Fusion (HAF) module that can be seamlessly integrated into multiple segmentation architectures, including FPN, Mask R-CNN, U-Net, and DeepLab. The module sequentially applies channel and spatial attention mechanisms to adaptively recalibrate multi-scale features, enhancing the representation of subtle and occluded instances. Experimental results demonstrate that HAF-augmented models significantly outperform their baseline counterparts across all architectures. For semantic segmentation, U-Net+HAF and DeepLabv3+HAF achieve mean IoU improvements of 4.5% and 4.4%, respectively, reaching 83.8% and 85.5% mIoU. For instance segmentation, Mask R-CNN+HAF demonstrates the most substantial improvement in small object detection, with Average Precision for small objects (APs) increasing from 42.3% to 50.6%—an 8.3-point improvement. Qualitative analysis confirms superior capability in detecting missed small instances, separating overlapping vessels, and producing more precise boundaries compared to baseline models. Satellite images ship detection and segmentation deep learning South East Iran Attention Mechanism 2026 3 01 1 11 http://ijmt.ir/article-1-898-en.pdf

50-886 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 Pedram Edalat edalat@put.ac.ir Erfan Rezaei erfanrezaei1380@gmail.com Alireza Abyari Bidgoli abyari1381@gmail.com This paper proposes a probabilistic model based on machine learning algorithms to estimate the risk associated with different levels of damage (per DNV-RP-F101) due to a dropped-object impact on subsea pipelines. The model is generalized by considering a wide range of pipeline geometric and mechanical specifications, corrosion conditions, and various possible impact scenarios. Multiple machine learning algorithms—including Linear Regression, Decision Tree, Random Forest, K-Nearest Neighbors, Support Vector Machine, and Gradient Boosting—were evaluated, with Random Forest demonstrating the highest accuracy. The analysis of how pipeline characteristics influence the probability of different damage levels provides a basis for decision-making on implementing preventive measures to reduce damage probability during the pipeline design stage Submarine pipeline dropped object Machine Learning (ML) Monte Carlo Simulation (MCS) Risk assessment pitting corrosion 2026 3 01 12 28 http://ijmt.ir/article-1-886-en.pdf

50-884 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 Hamid Hajirahimi Kashani hajirahimi.hamid@mail.um.ac.ir Seyed Alireza Seyedin seyedin@um.ac.ir Synthetic aperture sonar (SAS) achieves high-resolution imaging by coherently integrating echoes along the platform trajectory, making it inherently sensitive to six-degree-of-freedom (6-DOF) motion. Sub-wavelength perturbations distort phase, broaden the main lobe, and raise sidelobes. This paper introduces a geometry-centric array optimization framework that: (i) operationalizes a Phase Sensitivity Index (PSI) with translational and rotational components to quantify array-dependent motion sensitivity; and (ii) calibrates PSLR/ISLR to image signatures (main-lobe width and sidelobe proximity) against a back-projection (BP) reference, ensuring physics-consistent visual interpretation. Three receiver arrays—2D planar, 3D cubic, and 3D hemispherical (36 elements each)—are evaluated under six motions (surge, sway, heave, roll, pitch, yaw) and three imaging states (ideal, motion-degraded, corrected), yielding 54 scenarios. Motion parameters reflect realistic autonomous underwater vehicle conditions: translational RMS 2.0–3.2 mm, rotational RMS 0.25°–0.35°, platform speed 2.5 m/s, and dominant 0.2–2 Hz content. The hemispherical array attains the lowest normalized PSI and consistently superior sidelobe metrics, outperforming the planar array by approximately 6.5 dB in PSLR and 5.6 dB in ISLR on updated corrected global means, and the cubic array by approximately 2.9 dB in both PSLR and ISLR. Yaw and Heave cause the greatest degradation; Roll is least harmful. Motion correction improves PSLR by ≥4 dB and ISLR by ≥3 dB relative to degraded images, while the residual gap to ideal remains ≤4 dB, by design. The proposed framework enables quantitative array selection and supports robust SAS imaging in embedded AUVs without external navigation sensors. Synthetic Aperture Sonar (SAS) 6-DoF motion Phase Sensitivity Index (PSI) Hydrophone Array Optimization Back-Projection 2026 3 01 29 47 http://ijmt.ir/article-1-884-en.pdf

50-903 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 Kimia Nazarizadeh k.nazarizadeh@outlook.com Hashem Nowruzi h.nowruzi@nit.ac.ir Accurate forecasting of operational parameters is essential for predictive maintenance and digital twinning of offshore wind turbines. Using a unique dataset from the Levenmouth 7MW demonstration turbine, we compare a purely data-driven stacked ensemble model (StackedRidge) with a novel physics-informed neural network (GET-PINN) that incorporates the Energy Gradient (K) parameter from Energy Gradient Theory (GET). The StackedRidge model achieves superior predictive accuracy (RMSE = 0.2976, R² = 0.9731) for barometric pressure signals. In contrast, the GET-PINN provides valuable physics-aware diagnostics by jointly estimating the flow instability parameter K, supporting the detection of phenomena such as vortex-induced vibrations (VIV), albeit with higher forecasting error. These results highlight the complementary strengths of the two approaches: the stacked ensemble for high-fidelity point forecasting and the GET-PINN for interpretable, physics-guided maintenance decision support in operational wind farm digital twins.   Floating Offshore Wind Turbine (FWOT) Physics-Informed Neural Networks (PINN) Gradient Energy Theory (GET) Digital Twin Hybrid ML 2026 3 01 48 59 http://ijmt.ir/article-1-903-en.pdf

50-902 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 Sina Mohammadi sinamohammadi7456@gmail.com Reza Dezvareh rdezvareh@nit.ac.ir Fatigue damage is the predominant failure mechanism for offshore wind turbine support structures, with tubular joints of jacket foundations being particularly vulnerable due to stress concentrations under myriad cyclic loadings. This paper presents a rigorous and detailed methodology for assessing and enhancing the fatigue life of a 5 MW jacket-supported offshore wind turbine (JOWT) through the implementation of an advanced vibration control system. A Semi-Active Liquid Column Gas Damper (SALCGD) is employed to mitigate dynamic responses, thereby reducing stress ranges at critical hotspots. The study employs a multi-step, integrated numerical framework: a high-fidelity finite element model of the JOWT is developed; dynamic analyses are conducted under 29 distinct North Sea wave and wind conditions for three structural configurations—uncontrolled, with a passive damper (TLCGD), and with the semi-active damper (SALCGD); stress time histories at all tubular joints are extracted; critical members are identified based on stress severity; and finally, cumulative fatigue damage is calculated using the Rainflow counting method, relevant S-N curves for welded details, and Palmgren-Miner's rule. The results demonstrate a transformative improvement in fatigue performance. The SALCGD dramatically reduces stress ranges at the hotspots, leading to a remarkable increase in the fatigue life of the most critical structural joints by a factor of 2.5 to 3.2 compared to the uncontrolled case. Furthermore, it outperforms the passive TLCGD, providing up to 1.5 times greater life extension. The analysis reveals that in the uncontrolled scenario, several critical joints, particularly in the wave-dominant lower stories and wind-dominant upper stories, possess a fatigue life shorter than the 20-year design life. The SALCGD successfully elevates the life of all these joints well beyond this critical threshold. This study quantitatively establishes that semi-active vibration control is not merely a serviceability tool but a powerful and essential strategy for significantly extending the operational lifespan, ensuring structural integrity, and improving the economic sustainability of offshore wind infrastructure. Offshore Wind Turbine Jacket Structure Fatigue Damage Tubular Joints S-N Curve Rainflow Counting Semi-Active Damper Life Extension Structural Integrity. 2026 3 01 60 72 http://ijmt.ir/article-1-902-en.pdf

50-893 2026-07-13 10.1002

International Journal of Maritime Technology ijmt 2345-6000 2476-5333 10.66224/ijmt 2026 22 2 roshanak khosrojerdi khosrojerdi.rosha@ut.ac.ir farhoud kalateh fkalateh@tabrizu.ac.ir Oil spills represent a critical source of marine pollution with profound environmental, economic, and social ramifications. This numerical study investigates the interaction between sea waves and oil slicks using ANSYS AQWA software to elucidate the dynamics of oil pollution distribution under wave action. The research examines three pollution mass models with varying dimensions under different wave approach angles (0°, 30°, and 60°) utilizing second-order Stokes wave theory. The investigation reveals that displacement of the pollution mass center decreases with increasing wave angle. Maximum displacement and mobility increase proportionally with expanding oil slick dimensions at constant mass. Wave-induced forces on pollution masses exhibit inverse correlation with wave angle. Forces from irregular waves exceed those from regular Stokes waves by approximately 20% for larger pollution masses. For expanded oil slicks, irregular wave forces can reach up to three times the magnitude of regular wave forces. These findings provide crucial insights for oil spill response planning, containment system design, and environmental impact assessment in marine environments.     oil spill modeling wave-pollution interaction numerical simulation ANSYS AQWA marine pollution dynamics hydrodynamic analysis 2026 3 01 73 82 http://ijmt.ir/article-1-893-en.pdf