Strong typhoon winds enhance turbulent mixing, which induces sediment to resuspend and to promote chlorophyll-a (Chl-a) blooms in the continental shelf areas. In this study, we find limited Chl-a responses to three late autumn typhoons (typhoon Nesat, Mujigae and Khanun) in the northwestern South China Sea (NWSCS) using satellite observations. In climatology, the Chl-a and total suspended sediment (TSS) concentrations are high all year round with higher value in autumn in the offshore area of the NWSCS. After the typhoon passage, the Chl-a concentration increases slightly (23%), while even TSS enhances by 280% on the wide continental shelf of the NWSCS. However, in the southern area, located approximately 100 km from the typhoon tracks, both TSS and Chl-a concentrations increase 160% and 150% after typhoon passage, respectively. In the deeper area, the increased TSS concentration is responsible for the considerable increase of the Chl-a. An empirical analysis is applied to the data, which reveals the TSS and Chl-a processes during typhoon events. The results of this study suggest a different mechanism for Chl-a concentration increase and thus contribute toward further evaluation of typhoon-induced biological responses.

The stability performance of the buoy is an important parameter that should be taken into account when designing marine buoys. This paper introduces a theoretical and numerical analysis method to examine the stability of marine buoys, including analysis of the initial stability and large inclination stability by calculating the natural period, metacentric height, static stability, and dynamic stability, deriving the calculation process of the static stability lever in detail to obtain the righting moment, and so on, showing that the designed buoy in this paper has sufficient stability performance with theoretically excellent resistance performance to the wind and waves. Additionally, the hydrodynamic performance of the buoy at different heights is also further analyzed for structural optimization, which concluded that the buoy would have a more balanced stability performance when the buoy’s width-to-height ratio is 0.375–0.5, hoping that the computational model and ideas used in this paper can provide a reference for the theoretical stability analysis and buoy design of other types of buoys.

The study of stabilisation performance is a crucial consideration in the design of offshore floating platforms. For large floating structures, incorporating passive anti-roll tanks is a common technique for roll reduction. To investigate the feasibility of using ballast tanks for roll attenuation on the “Guo Hai Shi 1” platform, this study employs the internal tank theory to analyze the influence of four independent empty tanks, located around the platform, acting as anti-roll tanks with varying ballast water volumes. The results indicate that: (1) Different ballast water volumes within a single tank do not significantly affect the static stability parameters of the platform. (2) In regular wave simulations, the ballast tanks show limited effectiveness in reducing pitch motion along the wave incidence direction but effectively suppress coupled responses in other degrees of freedom. In the resonance case (Case 3), the minimum pitch occurs in Condition 1 at 10.56°, while the maximum pitch reaches 11.45° in Condition 2. Nevertheless, a 40% reduction in roll motion is achieved (3.36° in Condition 4 vs. 5.60° in Condition 1), along with a 24.5% reduction in yaw motion (39.22° in Condition 4 vs. 51.94° in Condition 1). (3) In irregular wave simulations, the ballast tanks effectively reduce the heave amplitude by up to 8.34% in sea state level 4 and 6.06% in sea state level 8, thereby enhancing its wave-following performance in the heave degree of freedom. (4) A CNN_BiLSTM_Attention algorithm is developed using hydrodynamic analysis generated datasets to predict the pitch motion time series of the platform under different ballast water conditions and sea states, while the model has a superior prediction performance (R² = 0.9658, RMSE = 0.5343, MAE = 0.3188, representing a 4.82% increase in R² and 30.31% reduction in RMSE compared to the original model). Future work will further explore the application of ballast tanks on floating platforms, with a focus on performance optimization and the development of advanced neural network models capable of predicting motion responses under various ballast configurations. Moreover, appropriate evaluation metrics will be established to assess the effectiveness of ballast tank designs. Efforts will also be directed towards integrating time-domain motion prediction using neural networks with control theories aimed at dynamically regulating ballast water volume to enhance platform stability.

Accurate motion prediction of floating platforms is critical for ensuring operational safety in offshore engineering applications or marine equipment testing. However, the strong nonlinearity and non-stationary characteristics induced by complex marine environments pose significant challenges to conventional prediction models. This study proposes a reinforced hybrid neural network (CNN-BiLSTM-Attention) integrated with advanced signal processing techniques to address these challenges. The methodology combines complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) for multi-scale signal analysis, coupled with temporal feature engineering through sliding window optimization. And the architecture innovatively integrates convolutional neural networks for spatial pattern extraction, bidirectional long short-term memory networks for temporal dependency modeling, and attention mechanisms for dynamic feature weighting. By analyzing datasets generated via hydrodynamic simulations, this study elucidates the model’s physical interpretability and establishes a closed-loop validation framework between data-driven methods and physics-based models. Finally, the predictive performance of the model is evaluated using motion datasets of the proportional platform in the water pool test under different working conditions, demonstrating its broad applicability and transferability by assessed using a dual-stage EWMA control line. Overall, the proposed CNN-BiLSTM-Attention model and its data-physics integrated validation method provide a reliable, interpretable and transferable solution for floating platform motion prediction, which can break through the limitations of single analysis methods, and provide a new research idea for integrating data-driven and physics-based methods in the field of ocean engineering.