Digital Twin (DT) technology is increasingly recognized for its potential to enhance product lifecycle management, operational efficiency, and data-driven decision-making across multiple industrial sectors. Although its adoption is growing in fields, such as aerospace and shipbuilding, its application within the Naval Auxiliary Industry—comprising the systems and suppliers that support core naval platforms—remains limited. This paper presents key insights from the TWIN NAVAUX project, which promotes the integration of digital twin technologies within this sector in the Galicia–North Portugal Euroregion. Based on demonstrator and real-world use cases, the study identifies major technical and organizational challenges, including system integration, data interoperability, and the inherent complexity of modelling ships as systems-of-systems (SoS). The work also outlines the essential building blocks required for successful DT deployment, such as standardized communication protocols, robust data acquisition and storage infrastructures, and advanced modelling and simulation techniques. Overall, the findings provide a structured roadmap to support and guide future digital twin implementations in the naval auxiliary domain.
Understanding the motion and deformation behavior of floating photovoltaic (FPV) is crucial for achieving a reliable structural-mooring system design, optimum energy output, and minimizing maintenance works. This study performs two-dimensional experiment to investigate structural motions and tension on the connection between floaters of modular FPV under regular wave in relatively shallow depth with vertical pin connector and taut mooring system. The 1:10 scale model is developed in accordance with the Froude–Cauchy similitude law and made of composite materials to capture structural stiffness. The bottom profiles are varied to investigate the effect of the bottom change to the incoming waves. The results show that pitch has direct correlation with structural tension, with higher pitch results higher tension. Smaller wave periods result in higher pitch response and tension on connection between floaters. The heave and pitch response at the bow and stern area are limited by the pre-tensioned of taut moorings, but wave over-wash occurs in this area. With the same incoming wave heights and periods, sloping bottom profile affects the wave steepness. The local waves within the empty gaps of the model and wave over-wash at the bow and stern area have affected both local and global responses of the FPV model.
Advancements in technology have resulted in the widespread use of AUVs worldwide to monitor underwater regions for applications, such as environmental sensing, underwater exploration, and underwater inspection. However, providing dependable communication among cooperating AUVs is challenging because of the high delays in acoustic signal propagation (attributed to slow relative speed), limited bandwidth, dynamic topologies, packet contention, and limited energy. The proposed method represents an adaptive energy-efficient routing and congestion-control framework with intelligence (AERCCF) to enhance dependable communication among cooperating vehicles during ocean-monitoring missions. AERCCF has congestion-aware traffic handling with flexible multi-parameter routing strategies and an intelligence-based decision-making system that aids in jointly optimizing communication reliability, energy consumption, and network performance. The routing solution focuses on using a combination of parameters rather than individual considerations of residual energy, link reliability, distance from the sink, and node speed. The congestion-control mechanism attempts to prevent packet congestion and queuing by using queue-aware rerouting. The simulation results show better performance incorporating all parameters with a Packet Delivery Ratio (PDR) of 95.8
This study experimentally investigates the hydrodynamic performance of a pile-supported, partially immersed semicircular breakwater with a concave seaward face. The two-dimensional model was suspended between two rows of circular piles and tested under regular wave conditions. Key geometric parameters included relative immersion depths (d/h = 0.25–0.5) and relative pile gaps (P = 0.5–0.83), while wave conditions covered relative water depths (h/L = 0.12–0.36) and wave steepness (Hi/gT2 = 0.0007–0.0076). Structural efficiency was quantified using transmission (Kt), reflection (Kr), and energy dissipation (Kd) coefficients. Results indicate that increasing immersion depth significantly enhances wave attenuation by reducing Kt and increasing Kr and Kd. In contrast, larger pile gaps moderately increase transmission due to a higher relative pile gap. For d/h = 0.25 and P = 0.83, Kt remained below 0.30, while Kr and Kd exceeded 0.52 and 0.62, respectively, demonstrating strong wave-blocking capability. To support preliminary design applications, empirical predictive equations were developed using Multiple Polynomial Regression (MPR). The models showed strong agreement with the experimental data, with normalized root-mean-square errors ranging from 0.019 to 0.048. A comparative evaluation with previous breakwater configurations confirms that the proposed pile-supported semicircular design provides enhanced wave attenuation, characterized by a significant reduction in the transmission coefficient and a higher capacity for wave energy dissipation under a wide range of wave steepness conditions.
Marine biofouling presents significant challenges in the maritime industry, including increased drag, fuel consumption, and maintenance costs. Traditional inspection and mitigation methods are labour-intensive and time-consuming, highlighting the need for automated approaches of biofouling detection and analysis. This study aims to bridge the existing literature gap by introducing an enhanced attention U-Net architecture specifically optimised for the semantic segmentation of marine biofouling in real-world conditions. Our model incorporates spatial attention gates within the skip connections and squeeze-and-excitation modules within each convolution block of a traditional U-Net framework. It was trained and tested on an annotated dataset of 504 in-water biofouling images collected from multiple ship hull surveys, provided by diving companies, classification societies, and NTUA’s archive. It contains images captured under various environmental conditions, which enables better model generalisation. The proposed architecture achieves a test set macro Dice coefficient of 0.775 and a macro accuracy of 0.795, outperforming a simple U-Net baseline across all segmentation metrics. Promising implications arise for deployment in automated inspection systems, potentially enhancing the efficiency of hull and offshore structure inspections by reducing manual effort and improving detection accuracy.
