The examination of the temperature distribution aboard ships holds significant relevance in ship structural design, serving to mitigate the occurrence of thermal stress and forming a fundamental aspect of infrared stealth applications. Leveraging a thermal model tailored for large vessels, we developed a three-dimensional finite element model within the COMSOL simulation software. The effects of time, season, latitude, ship speed and model thickness on the thermal characteristics of ship were analyzed. The results show that time affects solar azimuth angle, there are temperature differences between ship structures, The temperature stress at the connecting part of the chimney and deck is large, which is easy to produce deformation. The solar altitude angle is primarily influenced by the season and latitude. When latitude rises from 20°N to 60°N, the temperature of upper surface of the ship decreases by 19 K and the ship’s long-wave thermal emission level is therefore expected to decrease, which may reduce infrared detectability under comparable background and observation conditions. Higher ship speed enhances convective heat transfer, provides thermal boundary conditions for infrared signature assessment. While increasing hull thickness also provides thermal boundary conditions for infrared signature assessment, hull structure and cost must be considered. The conclusions drawn offer a theoretical foundation for enhancing energy efficiency and ensuring thermal safety during the navigation of large-scale ships.
Underwater gliders, as key autonomous subsea observation platforms, are widely used in ocean environmental monitoring owing to their long endurance, ease of deployment, and wide-area coverage. The coupled effects of time-varying ocean currents and progressive biofouling, compounded by restricted underwater communication, constrain glider observations. The same coupling also leaves signatures in the onboard records, which can be used to reconstruct environmental fields and assess glider health. This paper proposes a physics-informed, adjoint-based dynamics inversion framework. A six-dimensional lumped disturbance vector is introduced to reformulate disturbance identification as a full-window optimal-tracking problem, and an adjoint-based windowed variational estimator is constructed to recover physically consistent environmental perturbations, thereby inferring quantities beyond those directly measured by the native sensor suite. Validation on operational glider data demonstrates that the framework reconstructs sparse, impulsive disturbances correlated with non-coordinated maneuvers, whose spectral and intermittency signatures are consistent with Kolmogorov inertial-subrange scaling and She–Lévêque anomalous scaling (ζ6/ζ3=1.77 versus the predicted 1.78). These disturbances also survive surrogate-data, noise-only, and model-error falsification tests of the full inversion pipeline. By coupling the identified disturbances with a viscous-moment inversion model, depth-resolved current retrieval is demonstrated without acoustic Doppler hardware. The depth-averaged component is cross-checked against the glide-angle method and two independent ocean reanalyses under an explicit uncertainty budget. Multi-profile analysis further reveals a statistically significant, monotonic drift in the glider’s effective hydrodynamics over the deployment. The framework enables multi-dimensional environmental sensing and glider health assessment from low-cost onboard sensor data.
Hogging and sagging deformations are fundamental indicators of the longitudinal structural integrity and navigational safety of ships. While camera-based measurement systems offer advantages such as non-contact operation, large measurement range, and rapid deployment, their adoption in ship structural monitoring has been constrained by the shipboard environment, which renders conventional orthogonal front-view layouts infeasible. This paper proposes an oblique CCD moiré super-resolution method for measuring hogging and sagging deformation on operational ships. In this approach, the camera is mounted on elevated superstructures (e.g., the island superstructure), with the deformation carrier plane positioned on the deck surface, forming an oblique optical axis that avoids interference with routine deck operations. A rigorous geometric model is developed to characterize the mapping between vertical deck deformation and image-plane displacement under oblique imaging conditions. To overcome the inherent reduction in displacement sensitivity caused by the oblique configuration—the core technical challenge of this work—a novel CCD moiré super-resolution strategy employing a non-uniform grating carrier is introduced, substantially enhancing measurement sensitivity. Validation experiments demonstrate that the proposed method achieves a measurement resolution on the order of 10⁻⁵ of the observation field of view, markedly surpassing corner detection, digital image correlation (DIC), and conventional fringe analysis methods. Scaled ship model tests confirm a maximum measurement error of 0.022 mm within a 1 mm deformation range, validating the method's high-precision performance and engineering readiness. The proposed approach provides a practical solution for structural deformation monitoring of ships and advances the engineering deployment of ship structural health monitoring technology.
Solid fluidized method is a novel and promising technique for submarine methane hydrate exploitation, which involves the formation and decomposition of methane hydrate in submarine muds. However, the effects of complex flow conditions on solid fluidized exploitation are still unclear. This study employed real South China Sea sediments to simulate submarine muds and considered different flow conditions, including pressures (7.19, 7.79, and 8.14 MPa), rates (250, 300, and 400 mL/min), and water contents (50, 70, and 90 wt%). Based on the nonlinear logarithmic collaborative change stage of pressure and temperature during formation or decomposition process, non-equilibrium decomposition curve (NEDC) and minimum formation limitation (MFL) of methane hydrate in submarine muds under different flow conditions are quantitatively obtained. Experimental results indicate that the NEDC of methane hydrate in flowing submarine mud is only affected by fluidized water contents, rather than pressure and rate, which is mainly due to the heat and mass transfer obstruction of water phase surrounding hydrate crystals in decomposition process. In addition, it is found that the presence of MFL means the soil-in-water particles, around which the internal pressure of water can be decreased, are the attached media of hydrate formation in muds and are mainly affected by pressure, rather than fluidized rate and water contents. This study reveals the basic thermodynamic characteristics of methane hydrates in flowing muds, which are significant for the process control of submarine solid fluidized exploitation.
The reliable estimation of ship hydrodynamic performance remains a key requirement for vessel design and optimization. Traditional approaches such as potential flow-based boundary element methods (BEM) and strip theory provide useful solutions but face challenges in balancing computational efficiency and accuracy, particularly for forward-speed scenarios and optimization applications. This study introduces the application of the Harmonic Polynomial Cell (HPC) method, integrated with a boundary-fitted overlapping grid (BFOG) strategy, to improve the computational accuracy and stability of two-dimensional hull section hydrodynamics. Compared with the IBM strategy, the advantage of the BFOG strategy lies in the mathematically strict satisfaction of Neumann boundary conditions at the solid boundary nodes, while compared with direct background grid computation, it exhibits higher stability when dealing with multiple boundary geometries. The proposed method is validated through numerical simulations of U-shaped, triangular, and semicircular hull sections, where added mass and damping coefficients are computed and compared with both BEM predictions and experimental data. The hydrodynamic coefficients based on 2D hull sections are combined with the STF (Salvesen-Tuck-Faltinsen) method to calculate the three-dimensional ship hydrodynamics. The results indicate that the BFOG-HPC method provides hydrodynamic predictions comparable to those obtained by BEM for the cases considered, while maintaining stable performance in the high-frequency range. In addition, at the same level of computational accuracy, the BFOG-HPC method also exhibits the advantage of lower computational complexity compared with the BEM. These findings confirm the potential of the BFOG-HPC method as an efficient and reliable alternative for ship hydrodynamic calculations, providing a promising method for practical engineering applications in seakeeping and hull form optimization.
High-dimensional ocean data assimilation (DA) remains challenging when dynamics are strongly nonlinear, observations are sparse or irregular, or error statistics deviate from Gaussianity. Ensemble Kalman filters (EnKFs) are efficient and widely used, but they rely on approximate Gaussianity and can degrade when strongly non-Gaussian posterior distributions arise. Particle filters (PFs) offer a fully Bayesian alternative, yet they are hampered by weight degeneracy in high-dimensional geophysical applications. Here we introduce and implement a q-likelihood PF (q-PF) in the Parallel Data Assimilation Framework (PDAF), where the Gaussian likelihood is replaced by a q-Gaussian/Tsallis likelihood to temper the influence of outliers and mitigate weight degeneracy. The PDAF-MITgcm system uses PDAF offline coupling in a file-based forecast-analysis-restart workflow with cycled ensemble forecast priors. We evaluate q-PF against a localized ensemble transform Kalman filter (LETKF) and a baseline PF in (i) Lorenz-96 experiments spanning weak to strong nonlinearity and (ii) a 33-day winter regional case over the northern South China Sea and adjacent western Pacific assimilating satellite sea surface height (SSH) and sea surface temperature (SST). For ensemble size N = 40, q-PF yields lower period-mean continuous ranked probability score (CRPS) for both SSH and SST and reduces coherent SSH departure patches in eddy-active corridors. Eddy census and shelf-break diagnostics further show that the DA-cycled SSH output supports organized eddy genesis and propagation patterns consistent with energetic variability near the Luzon Strait and the continental slope.
Radar plays a crucial role in vessel traffic service (VTS) centers by monitoring ship positions and velocities. However, in inland bridge-area radar imagery, a single ship may appear as several disconnected components because of non-uniform scattering responses, bridge-induced occlusion or interference, strong structural echoes, and local background variations. This target fragmentation can cause duplicate tracks, identity switches, and degraded tracking performance. To address this problem, this work proposes a tracking-by-detection framework for robust ship tracking and fragment reconstruction using radar image sequences, without prior knowledge of ship scattering models, clutter models, or detailed environmental parameters. The framework contains two main stages. In the target candidate detection stage, a multi-orientation greatest-of CFAR detector (MOGO-CFAR) is developed to generate a high-recall candidate set, where weak-signal enhancement and orientation-selective background estimation are used to preserve fragmented ship components and small target candidates. In the target tracking and fragment reconstruction stage, motion prediction and data association are combined with a spatio-temporal buoy prior and track-guided fragment hypothesis selection to maintain consistent ship identities and reconstruct fragmented observations. Experiments on 288 radar frames collected from the Nanjing Yangtze River Bridge and its surrounding waters show that the proposed detection method achieves the highest recall of 0.9007 and the complete framework obtains a multiple object tracking accuracy (MOTA) of 0.8389, outperforming joint probabilistic data association (JPDA), multiple hypothesis tracking (MHT), nearest-neighbor (NN) tracking, Simple Online and Realtime Tracking (SORT), and random-matrix extended target tracking (RM-ETT). The results indicate improved robustness in complex inland bridge-area radar scenes.
Synthetic aperture sonar (SAS) imaging has emerged as a powerful tool for seabed mapping, underwater navigation, and marine-environment monitoring in ocean engineering. However, when deployed on unmanned underwater vehicles (UUVs) or compact sonar platforms, SAS systems face severe constraints on sensor aperture, data volume, and phase coherence caused by underwater propagation and platform motion. To address these challenges, this study proposes a compressive sensing discrete cosine transform–based synthetic aperture sonar (CS-DCT-SAS) imaging framework that introduces a domain-transformed sparse representation in the magnitude domain.The proposed method adopts a magnitude–phase separated formulation, treating phase as a structured distortion term while representing the physically real-valued magnitude sparsely in the DCT domain. To ensure practical feasibility on resource-limited underwater platforms, the framework further incorporates two-dimensional basis reduction and random down-sampling strategies, enabling efficient reconstruction under limited sampling and communication bandwidth.Simulation and real underwater experimental results demonstrate that CS-DCT-SAS consistently outperforms conventional CS-based SAS approaches in terms of structural image quality and robustness. In particular, CS-DCT-SAS achieves higher structural similarity (SSIM) while maintaining competitive target-to-background ratio (TBR), indicating improved preservation of extended target shape and texture alongside effective reverberation suppression. These results suggest that the proposed framework provides a balanced and practical solution for high-resolution SAS imaging on compact autonomous underwater platforms.
Particle degradation during long–distance hydraulic lifting can reduce nodule recovery, increase blockage risk, and generate fine particles that affect downstream handling and discharge control in deep–sea mining systems. However, direct prediction over kilometer–scale riser distances remains difficult. Resolved computational fluid dynamics–discrete element method (CFD–DEM) simulations are limited to short pipe sections, while empirical models provide limited insight into collision–energy mechanisms. This study develops a probabilistic collision–energy framework to predict nodule degradation in deep–sea mining risers. The model combines particle–wall impact accessibility, concentration–dependent shielding, impact–velocity probability distribution, fatigue–controlled fragmentation, and abrasion–induced surface loss in a unified formulation. Abrasion and fragmentation are treated as distinct mass–transfer pathways, linking fines generation and daughter–fragment redistribution to collision–energy exposure. Key collision statistics are assessed using CFD–DEM data, and the full model is evaluated against long–distance experimental particle–size distribution (PSD) data. Results show that particle–particle contacts are more frequent than particle–wall impacts but have lower wall–equivalent destructive energy, mainly contributing to cumulative abrasion and fine generation. Wall impacts are more relevant to fatigue damage and primary fragmentation. Pipe enlargement suppresses wall–impact fragmentation, whereas increasing solid concentration shifts degradation toward particle–particle abrasion. The resulting regime map distinguishes abrasion–dominated and fragmentation–prone conditions, providing a practical basis for degradation–risk assessment in deep–sea hydraulic lifting.
This paper investigates the trajectory tracking control problem for an air cushion vehicle (ACV) subject to model uncertainties and actuator faults. A funnel control strategy based on a novel tangent-type Barrier Lyapunov Function is proposed, which constrains the ACV’s position tracking error within a predefined performance funnel. To preserve tracking performance amidst actuator failures, an adaptive fault-tolerant control scheme with an event-triggered mechanism is devised. This control strategy utilizes a fuzzy logic system to estimate unknown actuator fault parameters and hydrodynamic parameters, thereby addressing system model uncertainties. Compared to existing studies, the introduced event-triggered mechanism effectively reduces communication load and mechanical wear between the controller and the actuators. Simulation results demonstrate the effectiveness of the proposed trajectory tracking control scheme.
Stable non-uniform density stratification is ubiquitous in the physical ocean and significantly modulates the evolution of vehicle wakes and the propagation characteristics of internal waves by suppressing vertical motions. Large-eddy simulations are employed to investigate the modulation effects of varying pycnocline thickness on wake–internal wave coupling in a hyperbolic-tangent density stratification. The results show that decreasing pycnocline thickness promotes pronounced wake flattening and drives the flow toward a quasi-two-dimensional state. The pycnocline thickness exerts a non-monotonic influence on vortex coherence and wake geometric scales, with intermediate thicknesses significantly enhancing the persistence of coherent vortical structures. The presence of a finite-thickness pycnocline induces focusing/defocusing and phase shifts of lee waves during propagation, while variations in pycnocline thickness nonlinearly modulate the defect velocity, geometric scales, and periodic characteristics of the mean flow. The onset location and intensity of wake-induced internal waves are also strongly controlled by the pycnocline thickness. This study reveals the mechanism by which non-uniform buoyancy-frequency profiles alter wake energy evolution by modifying the propagation conditions of internal waves.
As offshore wind power expands into ice-prone regions such as the Bohai Sea, the applicability of ice-breaking cones (IBCs) to large monopile-supported wind turbines has become an important engineering issue. This study compares the ice-induced vibration responses of a 15 MW monopile offshore wind turbine with and without an IBC under representative Bohai Sea ice and environmental conditions. Rather than enforcing identical ice-failure mechanisms across the two configurations, the comparison is carried out under the representative adverse ice-load scenarios relevant to each structural form. The results indicate that the static ice-load reduction effect of the IBC is strongly dependent on ice thickness; under thick-ice conditions, the extreme static ice force may even exceed that of the structure without an IBC. Dynamic analyses show that, for large and relatively flexible structures, the periodic ice-breaking process induced by the IBC may produce a narrow-band excitation with a dominant frequency close to the low-order natural frequencies of the turbine, which may lead to resonance-sensitive responses under certain extreme conditions. Under typical combined wind–ice conditions in the Bohai Sea, wind loading dominates the structural response in this range, and the difference between the two configurations is therefore limited. These results suggest that, under the present assumptions and loading conditions, the benefit of installing an IBC on a large monopile offshore wind turbine in the Bohai Sea is limited. This study provides a scenario-based assessment for Bohai Sea ice-resistant design rather than a universal conclusion on the necessity of IBCs in all ice-prone regions.
Flapping hydrofoil propulsion (FHP) systems offer a promising solution for wave powered unmanned surface vehicles (USVs), yet their performance is constrained under low frequency wave excitation due to limited pitch response and suboptimal phase characteristics. To address this limitation, this study introduces a passive FHP system incorporating a nonlinear pitch-stiffness mechanism aimed at enhancing low-frequency energy extraction. A surge-heave-pitch dynamic model is established based on Theodorsen’s unsteady hydrodynamic theory, and comparative simulations are performed against conventional linear stiffness configuration. Results indicate that the nonlinear stiffness leads to significant improvements in thrust, forward speed, and energy conversion efficiency. These improvements arise from increased pitch amplitude and a phase difference maintained closer to the optimal value of π/2, enabling more effective transfer of wave energy to propulsive work. The proposed nonlinear stiffness mechanism demonstrates clear advantages for wave powered propulsion and provides a theoretical basis for designing high-efficiency marine energy harvesters.
Comprehensive methods have been established for analyzing the shock wave energy and bubble pulsation energy of underwater explosions in the free-field. However, the research on the energy distribution of underwater explosion near the water-soil interface remains lacking. To investigate the energy coupling characteristics of the underwater explosion near the water-soil interface, a series of underwater explosion centrifuge model tests were designed and conducted with various explosive heights above the water-soil surface. Images of the bubble pulsation process, water pressure and pore-water pressure in the soil were recorded. By analyzing the bubble pulsation period, water pressure and pore-water pressure response, a calculation method for determining the explosion energy in water near the water-soil interface was proposed theoretically. It is indicated that as the scaled height increases, both the shock wave energy and bubble pulsation energy in water rise. Furthermore, the equivalent coefficient of impulse for underwater explosion near the water-soil interface is developed to analyze the ground shock energy characteristics induced by underwater explosion near the water-soil interface. As the scaled height increases, compared to the bottom-charge underwater explosion, the equivalent coefficients of impulse exhibits different variation across different phases. Finally, empirical formulas for calculating the equivalent coefficient of impulse under different scaled heights are derived. The research findings can provide a theoretical basis for the damage assessment and protective design of underwater structures such as subsea tunnels and submarine oil or gas pipelines.
Improving the energy efficiency of ships is a key pathway towards meeting the greenhouse gas reduction targets set by the International Maritime Organization. Voyage optimization is widely used to reduce fuel consumption and emissions, but its reliability depends critically on the ship performance model. However, uncertainties are often associated with today’s ship performance models, especially when many planning variables are considered. This study aims to investigate and clarify the impact of uncertainties in ship performance models on voyage optimization. A genetic algorithm (GA) is implemented in deep neural networks (DNN) to model engine shaft power in terms of its operational and metocean weather conditions. This GA-DNN model improved predictive performance R² by 50% over the semi empirical. The specific fuel oil consumption (SFOC) used to estimate fuel consumption from engine shaft power is proposed and described as a stochastic model, due to its strong dependence on engine settings related control variables that are often unknown in advance for voyage planning. The Gaussian process regression (GPR) method is utilized to establish the stochastic SFOC model, achieving an average RMSE of 2.91 g/kWh while also providing confidence intervals for uncertainty quantification. The Three-Dimensional Dijkstra Algorithm (3DDA) is used to investigate uncertainties of voyage optimization due to the uncertain SFOC. Finally, Monte Carlo simulation is employed to examine fuel consumption uncertainty from stochastic SFOC in voyage optimization and assess sensitivity of voyage optimization using different objective functions, such as deterministic power consumption, or stochastic fuel consumption with varying expected time of arrival (ETA). The proposed stochastic fuel consumption prediction model can provide more valuable information for the decision support of practical voyage planning, in terms of varying ETA and energy efficiency.
Suppression of whirling vibration is essential for the safe and stable operation of ship propulsion shafting systems. However, the mechanisms by which the axial positions of bearings influence the whirling vibration characteristics of ship propulsion shafting have not yet been fully elucidated. Therefore, using a ship propulsion shafting test bench as the study object, this research investigates the effects of the axial positions of different bearings on the whirling vibration characteristics of the shafting through numerical analysis. Furthermore, whirling vibration experiments are conducted using position-adjustable bearing pedestal assemblies with the bearings positioned at various axial locations, allowing validation of the numerical results. The results indicate that, while having a negligible effect on the natural frequencies and critical speeds, appropriately displacing the front stern bearing and the intermediate bearing toward the stern end can effectively reduce the whirling vibration response of the shafting under steady-state operating conditions. Moreover, adjustment of the front stern bearing produces a more pronounced vibration suppression effect than adjustment of the intermediate bearing. These findings provide valuable guidance for the control of whirling vibration and the optimal arrangement of bearings in ship propulsion shafting.
Vortex-induced vibration (VIV) poses a primary threat to the fatigue life and structural integrity of deepwater risers. Current offshore design practices and VIV prediction models predominantly rely on idealized uniform or linear shear flows. However, extreme typhoon environments often induce exponential shear flows featuring abrupt spanwise variations in the velocity gradient. Consequently, the VIV response and underlying fluid–structure interaction (FSI) mechanisms under these realistic and extreme flow conditions remain largely unresolved. This paper investigates the VIV behavior of a 3000 m full-scale deepwater drilling riser under typhoon-induced exponential shear flows, using a validated coupled numerical framework that combines the strip-theory-based Discrete Vortex Method (SDVM) for hydrodynamics and the Finite Element Method (FEM) for structural dynamics. Results demonstrate that traditional uniform and linear shear flow assumptions fail to capture the distinct broadband, low-wavenumber-dominated VIV response induced by realistic typhoon shear profiles. A critical “short-input/long-dissipation” FSI energy transfer mechanism is revealed: positive energy input is spatially confined to a high-velocity segment spanning less than 10% of the riser span, while the majority low-velocity section acts as the dominant hydrodynamic damping zone. Furthermore, contrary to conventional engineering understanding, extending the typhoon return period paradoxically suppresses global riser VIV amplitudes, a phenomenon attributed to intensified shear rates that significantly enhance damping effects in non-excitation regions. This work enriches the theoretical framework for VIV analysis of deepwater flexible structures under extreme ocean environments, and the identified localized energy input mechanism provides direct guidance for the anti-typhoon design of deepwater riser systems.
Engineering strategies for fluid locomotion typically aim to minimize drag, whereas biological systems often exploit it for propulsion. Conventional drag-minimization methods face challenges in further reducing hull drag, which remains a dominant source of resistance. When threatened, basilisk lizards slap the water surface with their hind feet, generating a reaction force that lifts their body out of the water and reduces resistance—a strategy we adapt here for surface vehicles. This study presents a surface vehicle equipped with a tandem flapping foil array that employs asymmetric kinematics (55°–85° downstroke, 5°–40° upstroke) to redirect foil drag for hull lift. This design enables a transition from a hull-borne to a foil-borne mode, sustaining water-running for over one hour. The foil motion converts foil drag into hull lift, which lifts the hull out of the water, thereby reducing hydrodynamic drag on the hull. Under the same 2 kg loading condition, compared to the Hull-borne mode (hull floating on water), the proposed Foil-borne mode reduces energy consumption per meter from 331 J/m to 165 J/m and increases maximum speed from 0.77 m/s to 1.65 m/s. This approach thus achieves drag reduction by harnessing foil drag, demonstrating the concept of drag utilization. The results offer a potential design perspective for surface propulsion and related engineering applications.
To address the domain shift problem in cross-station significant wave height prediction in South China Sea island and reef waters, this paper systematically proposes and evaluates three novel domain shift mitigation methods. Under the unified framework of a single-factor LSTM predictor, an evaluation framework encompassing "overall error distribution-single station accuracy-computational efficiency" was constructed to systematically compare the baseline method, the proposed "heavyweight" adaptation methods BO-FTTA (Bayesian Optimization Full-model Test-time Adaptation), BO-FBAT (Bayesian Optimization Freezing BN-layer Adaptive Test-time Tuning) and innovative "lightweight" adaptation method TDNA (Target Domain Normalization Adaptation). The study finds that all three methods significantly enhance cross-station prediction accuracy, but with markedly different computational overhead. Compared with the baseline method, BO-FTTA and BO-FBAT reduce the overall error standard deviation by 52 %–53 %, and achieve >50 % NMAE reduction in high-discrepancy cross-station scenarios. In contrast, TDNA obtains comparable performance improvement, with its calculation time merely 0.9–1.2 times that of the baseline method, keeping similar computational efficiency. Accordingly, the research clarifies the applicable scenarios for different methods: the "heavyweight" optimization methods are suitable for offline high-precision optimization, while the "lightweight" adaptation method provides a feasible solution for online, real-time deployment scenarios. This study provides quantitative references for method selection in cross-station wave height prediction of South China Sea island and reef waters, and also offers valuable insights for other marine time-series cross-domain adaptation issues.
This paper proposes an adaptive neural-network-based fast nonsingular terminal sliding mode controller (ANN-FNTSMC) for blade pitch regulation of floating offshore wind turbines (FOWTs) operating in the above-rated region. To facilitate controller design, a control-oriented reduced-order FOWT model is first established, and a nominal feedback linearization model is derived to expose the pitch-control channel. Based on the resulting dynamics, a fast nonsingular terminal sliding mode controller is developed to improve transient response and finite-time regulation performance. To reduce the dependence on an accurate prior bound of model uncertainty, an adaptive radial basis function neural network (RBFNN) is further introduced to estimate the dominant lumped uncertainty online and compensate it within the control law, thereby alleviating chattering and enhancing robustness to modeling mismatch and external disturbances. Lyapunov-based analysis is used to derive the adaptive update laws and to establish bounded closed-loop behavior together with finite-time convergence of the sliding dynamics under the adopted assumptions. Comparative simulations under combined turbulent-wind and Jonswap irregular-wave excitation show that the proposed ANN-FNTSMC strategy achieves improved rotor-speed regulation and reduced output-power fluctuation compared with GSPI, SMC, FNTSMC, and ANN-SMC. The results also indicate that the proposed method maintains reasonable platform-motion and blade-load responses under the considered operating scenario.
