Enter your email address
Submit
Write your message
[Articles In Press]
A Flexible Hybrid Attention Mechanism for Multi-Architecture Segmentation of Small Maritime Targets in South East Region of Iran
P. 1-11
Abstract
(116 Views) |
Full-Text (PDF)
(41 Downloads)
| Graphical Abstract
| Highlights
Highlights
| Audio file of the article abstract [MP3] (8 Download)
- Introduces a Hybrid Attention Fusion (HAF) module combining channel and spatial attention for improved small‑object segmentation.
- Demonstrates 4–9% accuracy gains across multiple architectures (U‑Net, DeepLabv3, FPN, Mask R‑CNN).
- Achieves 67.9% overall AP and 50.6% APs for small maritime targets.
- Provides a flexible, architecture‑agnostic module adaptable to various segmentation models.
- Employs a hybrid loss (λ = 0.6) to optimize detection of small, overlapping objects in satellite imagery.
Risk Assessment of Dropped Objects on Corroded Submarine Pipelines Using Machine Learning Algorithms
P. 12-28
Abstract
(117 Views) |
Full-Text (PDF)
(56 Downloads)
| Graphical Abstract
| Highlights
Highlights:
| Audio file of the article abstract [MP3] (6 Download)
- A probability-based framework for evaluating the damage levels in submarine pipelines based on machine learning algorithms is presented.
- The framework employs Monte Carlo Simulation (MCS) to generate a generalized data set based on uncertainties and a wide range of pipelines with their geometric and mechanical specifications, corrosion conditions, and various possible conditions of object impact.
- Using Machine Learning (ML)-based error methods to study the efficiency of different ML algorithms in the predictive model of damage level probability assessment.
- The study offers an actionable tool for implementing preventive measures in the pipeline design phase.
Optimizing Hydrophone Array Configurations for 6-DoF Motion-Induced Phase Error Correction in Synthetic Aperture Sonar Imaging
P. 29-47
Abstract
(1494 Views) |
Full-Text (PDF)
(182 Downloads)
| Graphical Abstract
| Highlights
Article Highlights
A geometry-centric framework is proposed to optimize hydrophone array configurations for robust SAS imaging under 6-DoF motion.
A normalized Phase Sensitivity Index (PSI) is introduced, with translational and rotational components, to quantify array-dependent motion sensitivity.
PSI is calibrated against BP-based PSLR and ISLR, linking phase-error sensitivity directly to interpretable image-quality degradation.
Across 54 scenarios, the 3D hemispherical array consistently outperforms cubic and planar arrays, achieving the best corrected sidelobe performance.
First-order sensitivity analysis shows that mutual coupling, tolerances, and multipath mainly shift absolute metrics but do not change the hemispherical array’s overall superiority.
| Abstract audio file [MP3] (140 Download)
A geometry-centric framework is proposed to optimize hydrophone array configurations for robust SAS imaging under 6-DoF motion.
A normalized Phase Sensitivity Index (PSI) is introduced, with translational and rotational components, to quantify array-dependent motion sensitivity.
PSI is calibrated against BP-based PSLR and ISLR, linking phase-error sensitivity directly to interpretable image-quality degradation.
Across 54 scenarios, the 3D hemispherical array consistently outperforms cubic and planar arrays, achieving the best corrected sidelobe performance.
First-order sensitivity analysis shows that mutual coupling, tolerances, and multipath mainly shift absolute metrics but do not change the hemispherical array’s overall superiority.
Operational Forecasting for an Offshore Wind Turbine: Benchmarking Data-Driven against Physics-Informed Machine Learning
P. 48-59
Abstract
(104 Views) |
Full-Text (PDF)
(56 Downloads)
| Graphical Abstract
| Highlights
Highlights
(1) StackedRidge forecasts pressure with RMSE=0.298 and R²=0.973.
(2) GET-PINN jointly predicts target and flow instability K.
(3) K exceeding 385 signals vortex-induced vibration risk.
(4) Data-driven ensemble outperforms PINN for autocorrelated signals.
(5) Hybrid model merges high accuracy with physics interpretability.
| Audio file of the article abstract [MP3] (13 Download)
(1) StackedRidge forecasts pressure with RMSE=0.298 and R²=0.973.
(2) GET-PINN jointly predicts target and flow instability K.
(3) K exceeding 385 signals vortex-induced vibration risk.
(4) Data-driven ensemble outperforms PINN for autocorrelated signals.
(5) Hybrid model merges high accuracy with physics interpretability.
A Comprehensive Fatigue Damage Assessment and Life Extension Strategy for Jacket-Type Offshore Wind Turbines Using a Semi-Active Control Damper
P. 60-72
Abstract
(100 Views) |
Full-Text (PDF)
(25 Downloads)
| Graphical Abstract
| Highlights
Highlights
| Audio file of the article abstract [MP3] (6 Download)
- Life Extension Factor: The SALCGD extended the fatigue life of critical joints by a factor of 2.5–3.2 versus the uncontrolled case, and up to 1.5 times compared to the passive TLCGD.
- Design-Life Compliance: It elevated the predicted life of all vulnerable joints from below the 20-year design threshold to well above it, ensuring structural integrity.
- Load-Zone Differentiation: Life extension was approximately 3.1 times in wave-dominated lower stories versus roughly 2.5 times in wind-dominated upper stories, reflecting the damper's higher efficacy against wave-induced global bending modes.
- Adaptive Robustness: Unlike the detuning-prone passive damper, the semi-active controller maintained superior damage suppression across all 29 sea states through real-time property adaptation.
- Rigorous Methodology: The study systematically translated global response mitigation into component-level fatigue life, establishing SALCGD as a quantifiable durability and life-extension strategy for offshore wind jackets.
Abstract
(67 Views) |
Full-Text (PDF)
(12 Downloads)
| Graphical Abstract
| Highlights
| Audio file of the article abstract [MP3] (4 Download)
Highlights
- A comprehensive numerical framework was developed to investigate wave–oil slick interactions using ANSYS AQWA under regular and irregular wave conditions.
- Wave approach angle significantly controls oil slick displacement and hydrodynamic loading, with displacement decreasing as the incident angle increases from 0° to 60°.
- Larger oil slicks exhibit greater mobility and experience substantially higher wave-induced forces compared with smaller pollution masses.
- Irregular (JONSWAP) waves generate considerably larger hydrodynamic forces than regular Stokes waves, reaching up to three times higher values for large slicks.
5. The obtained results provide practical insights for oil spill trajectory prediction, containment system design, and marine environmental impact assessment.
