To balance NOX emissions and performance in ammonia-fueled marine engines, this study investigates exhaust gas recirculation (EGR) effects on combustion, performance, and emissions, and proposes a full-load EGR strategy. Based on a validated CFD model of a methanol/diesel dual-fuel engine and an ammonia spray model, an ammonia/diesel dual-fuel engine model was developed. Simulations at four typical loads were conducted with various EGR rates. Results show that higher EGR slows ammonia combustion, prolongs combustion duration, intensifies after-burning, and increases ammonia emissions. Heat release rate is load-dependent, with peaks regulated by EGR. Engine performance varies with load: the gross indicated mean effective pressure (IMEPg) increases with EGR at low-to-medium loads, but first rises then decreases at high loads. EGR effectively reduces NOX emissions at all loads, with maximum reduction ranging from 71.5 % to 90.2 %. Fuel NOX shows higher sensitivity to EGR than thermal NOX. However, EGR also increases ammonia and N2O emissions—marginal at low loads but significant at medium-to-high loads, with a distinct surge threshold. Recommended EGR rates are 50 %, 40 %, 30 %, and 10 % at 25 %, 50 %, 75 %, and 100 % loads, respectively. This strategy enables compliance with IMO Tier III NOX regulations, achieves low ammonia and N2O emissions, and improves IMEPg.
Numerical studies on parametric roll have mainly focused on head and following seas, whereas parametric roll in oblique head seas remains less fully understood due to the coupled effects of roll, sway, yaw, and the propulsion-steering system. In this study, a CFD-based numerical framework is developed for predicting parametric roll of a self-propelled KCS ship in oblique head seas, considering the coupled interactions among the hull, rudder, and propeller. The effects of model speed and wave heading on parametric-roll responses are investigated, and two mitigation strategies are examined, namely passive roll reduction using a bilge keel and active avoidance through speed control. The results show that parametric roll in oblique head seas is highly sensitive to model speed and wave heading. The roll response becomes pronounced when the encounter condition falls within the parametric-roll-sensitive range, whereas it is significantly weakened when the ship moves away from this range. Wave headings closer to head seas correspond to a more critical parametric-roll region. The bilge keel suppresses roll growth by increasing roll damping and reducing the roll moment, while speed control mitigates the nonlinear response by adjusting the encounter frequency. The present study demonstrates that CFD can be used not only to identify parametric-roll occurrence, but also to support the development of passive mitigation and active operational guidance for self-propelled ships in oblique head seas.
This study investigates parametric rolling of a container vessel in regular head waves using a hierarchical modelling framework combining a nonlinear time domain motion solver, a blended Ikeda-based roll damping module, and high-fidelity CFD simulations employing single-mesh and overset mesh strategies. The selected case corresponds to the classical resonance regime for first order parametric instability. All numerical approaches consistently reproduce the fundamental resonance mechanism, including the characteristic frequency relationship between pitch and roll and the exponential growth phase. A key result is the close agreement in instability onset time among the reduced order model and both CFD configurations, demonstrating robust capture of the excitation/restoring interaction governing parametric resonance. Experimental measurements exhibit an earlier visible onset, attributed to unavoidable perturbations and slight irregularities in the basin wave record, highlighting the sensitivity of parametric instability to initial disturbance levels. Differences in saturation amplitude are primarily associated with viscous dissipation modelling and nonlinear free surface phenomena, including green-water events observed experimentally but intentionally excluded from the present CFD setup. The results confirm that accurate representation of the resonance mechanism is achievable across modelling levels, while amplitude prediction remains sensitive to nonlinear damping and deck interaction effects. The developed framework provides a validated and computationally efficient basis for further investigations of damping thresholds, irregular wave excitation, green-water modelling, and coupled propulsion–roll interaction.
Stiffened panels are the fundamental elements of ship hulls, composing main structures, such as decks and bulkheads. Due to their key role in hull dynamics, especially in the range of machinery-induced vibrations propagating throughout the structure, reducing their mobility is expected to contribute to an overall decrease in the vessel's acoustic footprint. An emerging technology with high potential for reducing structural vibrations is the Acoustic Black Hole. When integrated into the ship hull, these devices act as energy wells for flexural waves. Over the past two decades, Acoustic Black Holes have been extensively investigated through analytical, numerical, and laboratory-scale experiments. Yet, their practical integration into ship-representative structures remains unexplored. In this paper, two Acoustic Black Holes were integrated into the plating of a mock-up representative of a typical ship stiffened panel. Their placement was defined based on the structural intensity approach; an experimental campaign was conducted to demonstrate the vibration mitigation effect obtained. The results show the suppression of targeted modes and the reduction of vibrational velocity obtained on the stiffened panel plating. This research aims to be the first step towards implementing this technology onboard ships, paving the way for its future application in full-scale marine structures.
Wave impact loads, such as green water loading, are a critical concern for the structural safety of ships operating in severe sea conditions. Wave impact loads are commonly described as loads that increase rapidly over a short time; however, the criteria used to define wave impact loading differ between hydrodynamic descriptions and structural response considerations, and the connection between these two viewpoints remains unclear. To address this limitation, the present study investigates wave impact loading from a structural response perspective, with a focus on the roles of load impulse and loading rate. A simplified single-degree-of-freedom (SDOF) model, representing the local bending behavior of hull plating, is used to evaluate displacement and strain rate responses under idealized wave impact loading. The results indicate that displacement is mainly governed by load impulse, whereas strain rate response is strongly affected by the rate of load application. Even for wave impact loads with the same impulse, short-duration loading induces pronounced dynamic responses with high strain rates, while longer-duration loading results in quasi-static-like behavior. Based on these findings, a loading-rate-based impact index, Ip, is applied to green water loading and interpreted from a structural response perspective. Analysis of measured green water pressures shows that impact-type and quasi-static loads can be distinguished at approximately Ip ≈ 500, which is proposed as a practical threshold for classifying wave impact loading.
This paper focuses on trajectory tracking control problem in unmanned surface vehicles (USVs) operating under complex ocean disturbances. To accurately characterize the marine environment and evaluate control performance under realistic operating conditions, a three-dimensional short-crested irregular wave force model with randomly distributed frequencies and propagation directions is established. This model incorporates stochastic and multidirectional wave loading into the USV dynamics and control framework, providing a more realistic representation of wave disturbances for trajectory tracking analysis. Building upon this disturbance modeling, an adaptive predefined time nonsingular terminal sliding mode controller (adaptive PT-NTSMC) is proposed. First, a novel time-varying function is introduced to construct a variable-gain global sliding mode manifold, based on which a predefined time controller is developed to ensure predefined time convergence. Second, an adaptive law is established to independently estimate the upper bound of external disturbances, thereby reducing the reliance of controller design on prior disturbance-bound information. The global and predefined time stability of the closed-loop system is rigorously verified using Lyapunov theory. Simulation and comparative results under short-crested wave conditions validate the effectiveness and improved performance of the proposed control strategy.
Enhancing the environmental perception of Unmanned Surface Vehicles (USVs) in complex waters is among the primary approaches to ensuring the safety of autonomous navigation. This paper proposes a robust stereo-vision system that integrates ship detection and localization for maritime scenarios. In terms of detection, GS-YOLO is proposed to address the problems of large model parameters and low detection accuracy for small target ships. Based on YOLO11n, the Global-Local Spatial Attention (GLSA) module, Bidirectional Feature Pyramid Network (BiFPN), and SIoU loss function are introduced to improve detection performance while maintaining model lightness. For localization, a cascaded filtering localization algorithm is proposed to address the instability of distance measurement caused by dynamic interference. The algorithm takes depth data generated by RAFT-Stereo as input and applies temporal smoothing by sequentially combining median filtering and Kalman filtering. This significantly enhances robustness against dynamic interference. Experimental results show that GS-YOLO achieves a mean average precision of 93.5 % on the Mcships dataset while reducing parameters by 17.8 % compared with YOLO11n. It achieves an optimal balance between detection accuracy and model lightweighting. Additionally, compared with traditional methods, the cascaded filtering localization algorithm significantly reduces positioning error. Within the range of 50 m, the standard deviation of the error is reduced by 89.3 %; within an 80 m range, the positioning error is maintained below 2.3 %. These results demonstrate that the proposed stereo-vision system provides accurate and stable perception data for the autonomous navigation of USVs.
This study aims to design a novel experimental system for replicating the bearing capacity behavior of layered seabed foundations during the process of leg piling and penetration. A test apparatus for a jack-up offshore platform was developed, with a telescopic pile foundation hydraulic cylinder mounted at the base of platform leg to achieve experimental investigation of penetration behavior through layered seabed soils. A servo-hydraulic system model was developed by incorporating the Proportional-Integral-Derivative (PID) control algorithm into the proportional valve control system, thereby realizing dynamic simulation of various penetration depths and soil bearing capacities through precise hydraulic cylinder pressure regulation. A series of simulation analyses and experimental tests were conducted to delve into the variation in bearing capacity during the piling process, with a focus on the influence of soil shear strength, layer thickness, and backfill characteristics. The findings indicate that the experimental system exhibits a high level of accuracy in simulating the bearing capacity of complex layered seabed foundations, with an overall error maintained within 15 %. This study provides an experimental methodology and technical approach for simulating the process of pile leg penetration in the investigation targeting the structural integrity and stability of offshore platforms.
This study proposes a numerical approach for identifying the propeller characteristics that achieve maximum open water efficiency for a specific ship, considering its propulsion characteristics and defined operating conditions. The proposed method combines an artificial neural network (ANN) with an optimization procedure based on the genetic algorithm. The ANN is trained using experimentally obtained open water characteristics of 143 propellers, enabling accurate prediction of thrust and torque coefficients as well as the open water efficiency as functions of propeller geometric parameters. The optimal ANN achieved an R2 of 0.95 and RMSE of 0.20 on the validation set. Once trained, the ANN is integrated into the optimization procedure to explore the design space and identify the optimal propeller, while satisfying the imposed constraints. The approach is validated on several benchmark ships. The obtained results show good agreement with those from literature, despite the relatively small training dataset used in the present work. The obtained open water efficiencies are higher than those of the original propellers for all ships considered. It is demonstrated that the required propulsion characteristics used as input parameters can be obtained from different sources, including numerical simulations, experimental data, and empirical prediction methods such as the approach proposed by Holtrop and Mennen. For practical implementation, a standalone application was developed in MATLAB, integrating the trained ANN and genetic algorithm (GA) optimization procedure into a user-friendly environment.
Wind-assisted ship propulsion (WASP) technologies are increasingly considered a viable option for reducing fuel consumption and greenhouse gas emissions in maritime transport. Reliable prediction of their performance is essential for supporting design, retrofitting, and operational decisions, yet existing simulation approaches remain fragmented in terms of assumptions, fidelity, validation practices, and practical applicability. This paper presents a comprehensive review of simulation models used to predict the efficiency of WASP systems, including Computational Fluid Dynamics (CFD), empirical and hybrid formulations, experimental validation methods, and Velocity Prediction Programs (VPPs). Beyond the literature review, this work contributes three original elements. First, it proposes a conceptual taxonomy that classifies WASP simulation models according to model fidelity, operational applicability, and level of technological integration. Second, it introduces a structured and weighted decision-support framework to guide model selection across different project phases and data availability levels. Third, it identifies key research gaps related to validation, hull–sail interaction modeling, and operational integration, and outlines directions for future developments. By linking detailed simulation methods with practical engineering requirements, this review provides researchers and practitioners with a clear framework for selecting and applying WASP simulation tools, supporting more robust efficiency predictions and facilitating the effective deployment of wind-assisted propulsion in commercial shipping.
With the advancement of next generation information technologies, Maritime Autonomous Surface Ships (MASS) are progressively advancing. However, the dynamic uncertainties arising from multi-ship interactions in complex maritime traffic environments significantly constrain their capabilities for risk identification and adaptive switching between Mode(s) of Operation (MoO). To address this challenge, this study proposes a navigation energy field model for risk assessment that integrates multi-ship interaction features. First, maritime traffic complexity is quantified based on intrinsic ship attributes and the Potential Risk Ship Domain (PRSD) framework. Second, to address the inadequacy of conventional field theory in capturing dynamic coupling relationships between ships, a navigation energy field model is developed that incorporating multi-ship interaction characteristics, guided by quantified traffic complexity. Finally, applying the ALARP (As Low As Reasonably Practicable) principle, navigation scenarios are classified, providing a quantitative foundation to support adaptive MoO switching. The results demonstrate that the proposed method effectively reveals the risk evolutionary patterns of collective ship behaviors in multi-ship convergence and high-density traffic environments, thereby enhancing the ability of MASS to identify risks. This research provides theoretical and practical support for risk assessment and adaptive MoO management in MASS, contributing to improved navigational safety under dynamic and complex navigation situations.
Continuous ship steering control is a highly nonlinear and complex task, as it is subject to wave and wind disturbances. It is also crucial for timely obstacle avoidance and effective vessel maneuvering. Reinforcement learning (RL) combined with deep neural networks (DNNs) has demonstrated significant potential in controlling systems with nonlinear dynamics, making it well-suited for decision-making and planning in such complex scenarios. However, existing research struggles to ensure optimal control performance. To address this limitation, this paper proposes an improved deep reinforcement learning approach based on the Pathwise Derivative Policy Gradient (PDPG) algorithm to enable intelligent collision avoidance for continuous ship steering. The proposed method leverages the MMG model as the foundation for learning a steering control strategy using DNNs, comprehensively considers various control actions, and evaluates steering performance through a dedicated evaluation network. To enhance the policy network's representational capacity and balance exploration and exploitation, the PDPG algorithm's policy network structure is optimized. Additionally, an adaptive exploration rate and a dynamic balancing algorithm for random strategies are introduced to fine-tune the exploration-exploitation trade-off. The improved method's performance is verified through simulations of continuous ship steering control.
Hybrid propulsion systems increase ship energy efficiency by allowing the sharing of power between diesel engines and battery energy storage systems. However, the longterm efficiency of these types of systems depends on accurately estimating the Remaining Useful Life (RUL) of lithium-ion batteries to allow effective charge scheduling, maintenance planning, and reliable navigation. This study uses nine data-driven algorithms, including ensemble methods, recurrent neural networks, and linear models, to examine the RUL of a lithium-ion battery pack installed on a hybrid cargo ship. A 5-fold cross-validation structure was used to preprocess, normalize, and analyze actual operational data gathered during the vessel's service life. To improve the accuracy of predictions, hyperparameter optimization was performed out. Long Short-Term Memory (LSTM), which reduced MAE from 2.87 to 1.46 and RMSE from 12.57 to 6.34 after optimization while retaining a high coefficient of determination (R2 = 0.9999), performed the best among the models that were evaluated. The results obtained indicate that condition-based maintenance and energy utilization methods on hybrid ships can be effectively supported by data-driven RUL estimation. In order to enhance generalization and assess integration with real-time propulsion control systems, future research will expand the analysis to multi-vessel datasets.
To enhance the resilience of ship course-keeping and large-angle collision-avoidance maneuvers under cyberattacks, this study proposes a cyber-resilient course control methodology based on an innovative feedback-switching architecture. A dualfeedback automatic switching mechanism that integrates both positive and negative feedback modes is proposed. In this framework, sign inversion of measurement signals induced by simulated cyberattacks are interpreted as positive feedback, enabling automatic switching between positive and negative feedback. This transforms the large-angle collision avoidance task into a small-angle course deviation problem. The core design employs a second-order closed-loop gain-shaping algorithm to synthesize a robust linear controller, which is further augmented with a sine-based nonlinear compensator and implemented via zero-order hold. Compared with conventional designs, the combined approach reduces actuator amplitude by 41.7% reduction and rudder actuation frequency by 33.2%. Closed-loop stability is established through description-function analysis and verified via the Nyquist stability criterion. Full-scale simulations on the "Yupeng" training vessel under standard sea conditions demonstrate that the integrated system maintains course-keeping accuracy within 0.8 degrees during signal-inversion attacks while reducing steering energy consumption by 18.6%. The proposed automatic switching architecture effectively mitigates cyberattack effects during large-angle evasive maneuvers, reduces actuator wear and energy consumption, and enhances the safety and reliability of intelligent ships.
Accurate prediction of ship resistance remains a major challenge in Computational Fluid Dynamics (CFD), particularly when translating results from model to full scale. This study investigates the prediction of total resistance for the historic vessel Lucy Ashton using CFD across six model scales and full scale. Experimental resistance data were harmonized using third-order polynomial fits, enabling consistent comparison with CFD results. Two full-scale approaches were evaluated: Setup 1 with prescribed inflow and Setup 2 incorporating surge motion with applied thrust to emulate deckmounted jets used during sea trials. Across all scales, CFD predictions showed strong agreement with experiments, with deviations typically within +/- 5%, consistent with accepted validation standards. Dynamic motions (heave and pitch) were also examined, and both setups produced nearly identical trends, with absolute differences negligible for resistance assessment. The results demonstrate that both CFD methodologies provide reliable full-scale resistance estimates.
In this study, a floating offshore wind turbine with a foundation constructed from steel-UHPC (ultrahigh performance concrete) is investigated. The structural design and strength performance of the floating foundation are evaluated through a combination of numerical simulation and wave tank experiments. Load assessment and feasibility analyses, which reveal that the proposed steel-UHPC foundation satisfies structural safety requirements while exhibiting significant potential for material cost reduction, are conducted. With respect to the numerical simulation, a coupling of computational fluid dynamics (CFD) and finite element analysis (FEA) solvers is employed to address the fluid-structure interaction (FSI) problem. External hydrodynamic pressure obtained from the CFD solver is used to derive the structural response in the FEA solver. Given that the deformation of the steel-UHPC structure has a negligible effect on the surrounding flow field, one-way CFD-FEA coupling, in which fluid loads are transferred to the structural model without feedback of structural deformation to the fluid solver, is used. Wave tank experiments are conducted to validate the accuracy and reliability of the proposed one-way coupling methodology. Furthermore, an equivalent constitutive model for steel-UHPC is implemented within the FEA solver. The corresponding physical and mechanical properties are derived, and key structural design parameters of the floating foundation are determined accordingly.
Collaborative optimization with relaxation factor is proposed for the lines design of an underwater vehicle. The hydrodynamic performances and energy consumption are considered in optimization framework. Hydrodynamic performances include the resistance, sway force and yaw moment. The efficient power of the propeller is selected to reflect the energy consumption. Analytic hierarchy process (AHP) combined with Delphi method is used to allocate the weights of disciplines in the objective function at the top level. A gradient-based algorithm, sequential quadratic programming (SQP) in combination with an intelligent-based algorithm, the multi-island genetic algorithm (MIGA) is taken into account as the optimization algorithm. To increase the efficiency of optimization, an approximate model based on optimal Latin hypercube and radial basis function (RBF) is introduced to replace the time-consuming discipline analysis model. Full-appendage SUBOFF model is used to test the proposed optimization scheme. The optimization results show that the drag of the underwater vehicle is reduced by 2.05 %, the lateral force by 6.38 %, the yaw moment by 5.90 %, and the energy consumption by 2.15 %. Compared with a single algorithm (e.g., PSO), the proposed hybrid algorithm (MIGA-SQP) reduces the value of the comprehensive objective function by 2.5-4.8 %.The innovations of this paper are as follows: 1. The Delphi-AHP method is combined with the cooperative optimization of relaxation factors to improve the objectivity of weights; 2. An OLH-RBF surrogate model is constructed, which increases the CFD calculation efficiency by 4 times.
This study examines emissions from the rapidly growing nautical tourism sector in Croatia, a major hub for charter and leisure boat tourism, focusing on the significant emissions caused by vessels that lack advanced emission control systems. Despite intensive petrol and diesel consumption, emissions from this sector remain under-researched; this study addresses that gap using data from the Green Sail Association's Ecological Footprint Calculation Platform, collected during a 2023-24 pilot project in selected marinas in the Sibenik-Knin and Split-Dalmatia counties. The sample includes sailboats, yachts, and catamarans 10 to 20 m in length. Operational data (engine hours, fuel consumption, technical specifications) were gathered from charter companies and skippers. Emissions of CO2, NOX, CO, and PM were calculated using EMEP/EEA Tier 1 methodology based on fuel consumption and emission factors. Emissions rise with vessel length and are notably higher for catamarans due to greater fuel consumption. Though only 35 % of the sample, catamarans contributed nearly 60 % of total emissions. Based on the findings on the sample of 160 vessels, extrapolated to Croatia's fleet (similar to 4,500 vessels), seasonal emissions are estimated at 30,000 to 40,000 t CO2, 300 to 400 t NOX, 100 to 150 t CO, and 15 to 25 t PM. This study represents the first large-scale emissions estimate for charter vessels and recommends further research with a more representative sample, broader geographical coverage and advanced methods such as Tier 3 calculations, collection of generator fuel consumption data and real-world emissions measurements to improve emission factors.
Ship motion prediction is essential in marine engineering, but missing data caused by sensor faults or signal interruptions often degrades the accuracy of long short-term memory (LSTM) models. This study investigates how different missing data rates and imputation methods affect LSTM prediction performance. A ship-motion dataset under various speeds and wave conditions was used to examine model feasibility and hyperparameter sensitivity. Traditional filling strategies, including zero and mean filling, were compared under missing data scenarios. Results show that data loss significantly reduces prediction accuracy. The mean-filling method generally performs better than zero-filling, though its effectiveness decreases with higher data diversity. Proper data clustering can effectively enhance its performance.
This paper offers a comprehensive optimisation tool for the design and assessment of hybrid maritime power systems that combine internal combustion engines, fuel cells, and battery energy storage systems. Using a surrogate-assisted NSGA-II algorithm, the framework concurrently reduces operational expenditure, CO2 emissions, and life cycle cost assessment. Under constant technical criteria, including system weight and volume, with and without waste heat recovery, four fuel pathways-diesel, LNG, methanol, and ammonia-are evaluated. The results reveal considerable economic and environmental differences compared to the diesel baseline: LNG increases LCCA by 0.5 % (& euro;1.2M) and global warming potential (GWP) by 2 % (1752 kg), while acidification potential (AP) and aerosol formation potential (AFP) decrease by 91 % (914 kg and 1118 kg, respectively). Methanol reduces LCCA by 14.3 % (& euro;35.3M), GWP by 36 % (35,540 kg), and AP/AFP by 81 %, offering a cost-effective and environmentally balanced solution. Ammonia eliminates GWP, AP, and AFP, though with a 10.7 % (& euro;60M) increase in LCCA, demonstrating its potential for long-term decarbonisation. The findings show clear Pareto fronts for every fuel, suggesting that the possible design area is significantly influenced by fuel type. The framework offers practical guidance for designing energy-efficient, low-emission vessels, aiding in sustainable marine energy transitions.
