Suction anchors are large cylindrical anchors and well-suited for both catenary and taut mooring systems. The determination of horizontal and vertical load components hinges upon factors such as inclined loading angle and the padeye position, significantly influencing its failure mechanism. This study undertook a series of geotechnical centrifuge tests to scrutinize the behavior of suction anchor under lateral loading. Precise measurements were taken for variables including the pore pressure at the anchor tip, soil pressure along the anchor, anchor capacity, as well as the displacement and rotation around X, Y, and Z axes. The anchor failure mechanisms and interaction coefficients within a dissipation function were determined from the experimental results. The failure angle β serves as an indicator of anchor rotation in the translational failure mechanism, and α_r describes anchor rotation in backward and forward movements. Under the inclined loading angle (θ_m) of 35°, the anchor failed in translational mechanism, with its ultimate bearing capacity surpassing that of the anchor failing through backward and forward mechanisms. Upon elevating the padeye location moved upward to 4L/7 to 2L/3, the anchor failed in a forward mechanism, accompanied by a significant decrease in soil pressure in the passive side. This separation of the soil plug from the anchor, driven by accumulated displacement and rotation, resulted in 25% decrease in ultimate bearing capacity. Conversely, when the inclined loading reduced from 35° to 20° at padeye location, the anchor failed in a backward mechanism, with no observed separation of the soil plug from the anchor.

Offshore monopile foundations endure complex loads during service. They bear vertical loads from the superstructure’s self-weight and lateral cyclic loads (e.g., wind, waves), while near-coastal seabeds are usually sloped. To resolve these issues, model pile tests under combined vertical and lateral cyclic loads were carried out to study how slope angle and vertical load affect monopile deformation. A numerical model was validated via comparison with test results and then used to reveal the development of cumulative deformation in offshore sloped monopiles under the above combined loads. The results show that, below 1000 cycles, the cumulative displacement of the pile increases logarithmically with the number of cycles. As the slope angle increases, the cumulative deformation increases. In these tests, the cumulative deformation of the pile increases by 114% compared to the results of the flat site, but it decreases with an increase in vertical load. As such, when a 3 N vertical load is applied to the pile top, its deformation at the flat site decreases by 20%, but its deformation at a 20° slope site decreases by 10%. Finally, a predicted formula is proposed for offshore monopiles with the effect of slope angle, and this formula can provide a preliminary assessment method of cumulative.

This study presents the design and performance assessment of suction anchor foundations for an integrated offshore wind–solar–aquaculture system located in Jiangsu Sheyang, China. The project represents one of the first practical demonstrations of coupling renewable energy production with large-scale marine aquaculture on a shared floating platform. Using three-dimensional numerical simulations in FLAC3D and ABAQUS, the study evaluates the anchors’ bearing capacity, structural safety, and fatigue performance under ultimate (ULS), accidental (ALS), and fatigue (FLS) limit states. The analysis incorporates site-specific geotechnical conditions, seabed scour, and installation deviations, providing a realistic framework for foundation design in layered coastal sediments. Results confirm that the suction anchor system meets international safety requirements (DNV, CCS) and maintains robust performance throughout its service life. The findings demonstrate that scour depth and installation accuracy are critical factors governing anchor reliability and offer practical insights for updating offshore foundation design standards in future multifunctional renewable–aquaculture developments.

The integration of wave energy converters with offshore wind foundations offers a potential route to improving the utilization of offshore renewable energy infrastructure. This study numerically investigates the power-load characteristics of fixed oscillating water column (OWC) chambers intended for potential installation near offshore wind jacket foundations. A preliminary jacket comparison is first used to delimit the scope, after which the main parametric study is performed on isolated OWC chambers so that pneumatic response and local chamber loads can be compared consistently. The simulations are conducted under regular waves with a wave height of H = 0.05 m, a water depth of h = 1.6 m, and wave periods of T = 0.9–1.7 s. Three baseline geometries, namely cylindrical, sandglass-shaped, and bottle-shaped OWCs, are first screened in order to identify the most suitable reference chamber family. The cylindrical chamber is then retained as the reference configuration for subsequent local parameter studies of the frustum-contraction parameter D₂ and the front-wall draft d₂. The results indicate that the geometric effect is strongly dependent on the incident-wave period. The sandglass-shaped and bottle-shaped chambers can enhance short-period pneumatic power or reduce loads at longer periods, whereas the cylindrical chamber provides a more consistent reference response over the tested range. Under the wave conditions adopted in this study, further analysis reveals that D₂ exerts a non-monotonic tuning effect varying with wave period. For the selected frustum-shaped configuration, increasing d₂ reduces hydrodynamic loads yet simultaneously weakens pneumatic power output and CWR. Because the air phase is treated as incompressible and the orifice represents an orifice-only damping condition rather than a turbine-controlled PTO system, the reported P_e should be interpreted as a pneumatic/hydrodynamic comparison metric and not as wave-to-wire electrical power. The conclusions are therefore positioned as regular-wave geometry-tuning trends for the present model scale rather than as full coupled jacket-OWC design rules.

Pile–bucket foundation is a newly developed offshore wind turbine supporting structure created by combining a monopile and a bucket foundation. Current research mainly focuses on its bearing characteristics in homogeneous soil layers (e.g., pure sand and pure clay); however, field stratigraphy is normally in a layered distribution, which may alter the bearing mechanism of the pile–bucket foundation and lead to inconsistency in the prediction of the bearing capacity. In this study, three-dimensional finite element simulations are performed, and the static lateral bearing characteristics of pile–bucket foundation in layered sand–clay strata are investigated considering the impact of monopile size, bucket size, and sand/clay thickness ratio (H_s/H_c). Numerical results indicate that an increase in sand layer thickness could help increase the bearing capacity of the pile–bucket foundation, especially as H_s/H_c increases from 0.5 to 1. Then, hyperbolic p-y models are introduced to calculate the lateral load–displacement curves of the pile–bucket foundation, which is divided into three sections: the bucket section and pile section in sand and clay, respectively. The two controlling factors of the p-y model, the initial stiffness and the ultimate lateral resistance, are modified based on numerical results. The modified p-y model provides satisfactory predictions of the pile deflection and the moment distribution of pile–bucket in layered sand–clay strata and could serve as a framework in more complicated soil deposits, such as clay-over-sand or multi-layer.

Long-term time series forecasting is important in energy, transportation, and meteorology, yet existing methods struggle to model multi-scale features and preserve local sequence structure. We propose PatchRNN, a decomposition- and structure-aware framework for multivariate forecasting. Its parallel trend branch uses multi-scale patch embedding to capture and fuse long-term patterns at different temporal granularities. The seasonal forecasting branch separates the residual components via average pooling and employs a segmented recurrent neural network (SegRNN) to efficiently model their periodic fluctuations. Joint optimization combines point-wise error with Patch-wise Structural Loss (PSLoss). Fourier-based adaptive partitioning enables PSLoss to constrain local trend correlation, fluctuation variance, and mean. Extensive experiments across multiple real-world benchmark datasets demonstrate that our model obtains competitive long-term forecasting performance while preserving favorable local-structure consistency. On the real-world ocean-wave dataset School.nc, our model yields an MSE of 0.814 at prediction horizon 96, corresponding to an 8.64% reduction over the best competing baseline.