As a new kind of large observation platform, the three-anchor buoy system can effectively realize multifunctional ocean observation, e.g., ocean profiling and autonomous underwater vehicle docking. In order to understand effects of different anchor chain arrangements on the motion response of the three-anchor buoy system under the coupling effect of wind, wave, and current loads, a hydrodynamic model of the buoy system was developed. Wave-period-dependent characteristics of added mass, radiation damping, and the motion response amplitude operator (RAO) were analyzed to derive their response curves; the effects of adding additional viscous damping on RAO performance were investigated. Subsequently, frequency domain and time domain analyses were conducted on five three-anchor buoy systems with distinct anchor chain arrangements to investigate the variation patterns of 6-DOF motion response amplitudes, top-chain tension characteristics, and submarine anchor chain length alterations under combined wind, wave, and current loading conditions. The results show that under the same environmental load, when the three anchor chains are evenly distributed at 120°, the 6-DOF motion response amplitude of the buoy system is the smallest, the top-chain tension and the submarine anchor chain length are more in line with the design requirements, and the comprehensive performance is better.

High-precision wave data serve as a foundation for investigating the wave characteristics of the East China Sea (ECS) and wave energy development. Based on the simulating waves nearshore (SWAN) model, this study uses the ERA5 (ECMWF Reanalysis v5) reanalysis wind field data and ETOPO1 bathymetric data to perform high-precision simulations at a resolution of 0.05° × 0.05° for the waves in the area of 25–35° N and 120–130° E in the ECS from 2009 to 2023. The simulation results indicate that the application of the whitecapping dissipation parameter Komen and the bottom friction parameter Collins yields an average RMSE of 0.374 m and 0.369 m when compared to satellite-measured data, demonstrating its superior suitability for wave simulation in shallow waters such as the ESC over the other whitecapping dissipation parameter, Westhuysen, and the other two bottom friction parameters, Jonswap and Madsen, in the SWAN model. The monthly average significant wave height (SWH) ranges from 0 to 3 m, exhibiting a trend that it is more important in autumn and winter than in spring and summer and gradually increases from the northwest to the southeast. Due to the influence of the Kuroshio current, topography, and events such as typhoons, areas with significant wave heights are found in the northwest of the Ryukyu Islands and north of the Taiwan Strait. The wave energy flux density in most areas of the ECS is >2 kW/m, particularly in the north of the Ryukyu Islands, where the annual average value remains above 8 kW/m. Because of the influence of climate events such as El Niño and extreme heatwaves, the wave energy flux density decreased significantly in some years (a 21% decrease in 2015). The coefficient of variation of wave energy in the East China Sea exhibits pronounced regional heterogeneity, which can be categorized into four distinct patterns: high mean wave energy with high variation coefficient, high mean wave energy with low variation coefficient, low mean wave energy with high variation coefficient, and low mean wave energy with low variation coefficient. This classification fundamentally reflects the intrinsic differences in dynamic environments across various maritime regions. These high-precision numerical simulation results provide methodological and theoretical support for exploring the spatiotemporal variation laws of waves in the ECS region, the development and utilization of wave resources, and marine engineering construction.

In marine hydrodynamics, the core of the boundary element method (BEM) lies in the numerical calculation of the free-surface Green’s function. With the rise of artificial intelligence, using neural networks to fit Green’s function has become a new trend, yet most existing studies are confined to fitting Green’s function in infinite water depth. In this paper, a neural network fitting method for a finite-depth Green’s function is proposed. The classical Multilayer Perceptron (MLP) network and the emerging Kolmogorov–Arnold Network (KAN) are employed to conduct global and partition-based fitting experiments. Experiments indicate that the partition-based KAN fitting model achieves higher fitting accuracy, with most regions reaching 4D fitting precision. For large-scale data input, the average time for the model to calculate a single Green’s function value is 0.0868 microseconds, which is significantly faster than the 0.1120 s required by the traditional numerical integration method. These results demonstrate that the KAN can serve as an accurate and efficient model for finite-depth Green’s functions. The proposed KAN-based fitting method not only reduces the computational cost of numerical evaluation of Green’s functions but also maintains high prediction precision, providing an alternative approach to accelerate BEM calculations for floating body hydrodynamic analysis.

Profiling floats are important platforms for oceanic profile observations, yet they are prone to positional drift and grounding when deployed in shallow-sea environments. In order to address these issues, an aluminum alloy-based propulsion-enabled station-keeping anchoring system (PESKAS) is designed in this paper. The PESKAS comprises anchor wings, thrusters, a steering connector, support frames, and an upper connection flange, which allows easy installation to the bottom of conventional profiling floats. Three anchor wings, with a cone angle of 40° and a length of 0.12 m, enable the attached profiling float to anchor to the seabed under ocean currents of up to 0.5 m/s when fully penetrating the sediment. Numerical simulation results show that achieving full penetration into clay, clayey silt, and silty sand requires thrust forces of 80–100 N, 100–120 N, and 160 N, respectively. To achieve full sediment penetration, the PESKAS employs a redundant quadruple-thruster configuration (total thrust 200 N) with an effective actuation duration of approximately 1 s. It ascends from the seabed via a thruster-generated upward force during the ascent of the profiling float, effectively avoiding grounding. Over a complete operational cycle (descent and ascent), the PESKAS consumes approximately 0.65–1.84 kJ of energy. Compared to the energy consumption of PROVOR profiling float motors (10.25 kJ) and sensors (8.33 kJ), the additional energy requirement for the PESKAS does not have a significant effect on the endurance of profiling floats. According to the results of the simulation experiment of the PESKAS, the system successfully achieves its design objectives of full penetration into and ascending from sediments. PESKAS is a cost-effective solution for the positional drift and grounding of profiling floats, which enables stable long-term profile observations in shallow-sea environments and has broad application prospects.