Norwegian Geotechnical Institute

Predicting the history-dependent cyclic response of granular soils remains a central challenge for data-driven methods, which frequently rely on synthetic training data or exhibit instability during the post-liquefaction regime. This study presents a Long Short-Term Memory (LSTM) framework trained exclusively on raw experimental databases from cyclic Direct Simple Shear (DSS) and Triaxial (TXC) tests on Ottawa and Karlsruhe sands. To overcome the computational and physical constraints of laboratory data, the training pipeline implements systematic data thinning and physics-guided loss weighting, ensuring the precise capture of early-stage pore-pressure generation. Validated through rigorous Leave-One-Test-Out cross-validation, the model robustly predicts unseen stress–strain–pore pressure trajectories, including the broad, highly dissipative hysteresis loops characteristic of post-liquefaction flow without evidence of cumulative drift. Comparative benchmarks demonstrate that the restructured LSTM performs comparably to the PM4Sand constitutive model in reproducing stiffness degradation under different cyclic stress ratios. Furthermore, error diagnostics identify fabric anisotropy as the primary governor of prediction limits in dense sands. This research establishes a stable constitutive surrogate capable of generalising across diverse material states within each loading mode.

Suction caissons have been used successfully in the offshore oil and gas industry since the 1980s and more recently in offshore wind developments. During installation, instrumentation data are collected, including pumping rate and suction inside the caisson. This paper presents a calculation approach for using installation data to back-analyse in situ sand properties. Combining an analytical solution for pore water flow into a caisson and a trust-region optimisation algorithm, it is shown how measured pumping rate, penetration rate, and suction inside the caisson can be used to estimate a profile of hydraulic conductivity, including anisotropy, for sands with relatively high hydraulic conductivity greater than 10^-6 m/s. The extent to which such an approach can be expected to be successful is tested through a set of synthetic demonstration cases, generated with a finite element model, in order to identify capabilities and limitations. It is observed that a reliable estimate of the hydraulic conductivity of layered soil can be obtained, in some cases including anisotropy. Knowledge resulting from such back-calculation as demonstrated herein can be valuable in post-installation assessments, for example for design verification, asset management, and lifetime extension analyses.

Traditional integrated analyses of floating wind turbines (FWTs) commonly neglect the embedment of the mooring line in the seabed, which may increase uncertainties in mooring system design. A macro-model for chain-seabed interaction was previously introduced in the static mooring analyses for FWTs. In this study, dynamic analyses of the VolturnUS-S floater supporting the IEA 15MW FWT are conducted using a macro-model of chain-seabed interaction in SIMA. Five typical cases are considered to investigate the dynamic responses of FWTs under both operational and parked conditions, focusing on the influence of the embedded line and seabed friction. The mooring line responses, floater responses, and turbine responses are analyzed and compared. Finally, mooring line configuration and tension variation are analyzed and discussed. The embedded line notably impacts mooring line responses, primarily by reducing the load acting on the anchor padeye. While seabed friction has minimal effect on the maximum fairlead tension, it has a significant influence on the padeye tension. Additionally, extreme tension drives the embedded lines toward the floaters. When the tension is subsequently reduced, the lying chain moves back toward the padeye, and this reverse displacement induces reverse seabed friction. This study provides an insight into the embedded line influence on the dynamic responses of FWTs.

The environmental health challenges of per- and polyfluoroalkyl substances (PFASs) are well-documented in developed countries, where serious efforts are underway to implement stricter regulations to lower PFAS emissions. However, in developing countries where PFASs have been detected at levels similar to those in developed countries, there is a lack of comparable research or efforts on addressing PFAS pollution. These gaps also apply to many other industrial chemicals and are underpinned by imbalances in chemical regulation between developed and developing countries. These imbalances are likely to create multifaceted global challenges, including the illegal use and trade of PFASs and their products, the relocation of PFAS-based industries, and the global recirculation of PFAS pollution. These challenges can exacerbate pressure on developing countries already grappling with other critical environmental issues. In this Perspective, we explore these challenges arising from global disparities in the regulation of PFASs and other chemicals, along with their repercussions. We propose solutions to bridge the regulatory gaps, including broad, worldwide PFAS bans and regulations, increased funding for PFAS monitoring and emissions reduction, and joint initiatives with developed countries. These efforts would ensure that PFAS management extends beyond the developed world to countries with high economic aspirations and limited resources to address chemical pollution.

Understanding and accurately modelling monopile behaviour is a central challenge in modern offshore wind geotechnics, often requiring, for detailed design, robust finite-element (FE) simulations supported by well-calibrated constitutive models. This study critically evaluates and advances the application of 3D FE modelling for laterally loaded monopiles using the SANISAND-MS model, informed by a comprehensive experimental programme ranging from element-scale testing to centrifuge modelling under both monotonic and cyclic loading, in dry and saturated sand. This work investigates strategies for the reliable calibration of the SANISAND-MS constitutive model. Key calibration challenges are addressed, including limitations in test data availability, variability in material response, and the alignment of model parameters with soil strain levels representative of realistic operational scenarios. The study further highlights practical considerations and limitations associated with the use of SANISAND-MS, particularly when extrapolating features of foundation response observed in physical modelling to full-scale conditions. For the cases considered herein, comparisons between numerical simulations and experimental data show good agreement in dry conditions, whereas reduced accuracy in saturated cases underscores the need for a more detailed treatment of, among other factors, soil-pile interface behaviour and loading-rate effects on excess pore water pressure generation. Overall, the findings provide valuable guidance for improving the fidelity of advanced FE simulations for offshore monopile design.