This paper presents an investigation into the suitability of the SANISAND-MS model for the three-dimensional finite-element (3D FE) simulation of cyclic monopile behaviour in sandy soils. In addition to previous work on the subject, the primary focus of this study is to further assess the model's capability to reproduce the accumulation of permanent deflection/tilt under cyclic lateral load histories. To this end, experimental data from the PISA field campaign are employed, particularly those emerged from the medium-scale cyclic tests conducted at the Dunkirk site in France. The methodology adopted herein involves calibrating the SANISAND-MS model's parameters to align with 3D FE simulation of a selected monotonic pile test reported by the PISA team using a bounding surface plasticity model partly similar to SANISAND-MS. Subsequently, the soil parameters governing SANISAND-MS’ ratcheting response are calibrated using only minimal information from published PISA field data. While representing the first attempt to simulate the reference data set using a fully ‘implicit’ 3D FE approach, this paper offers novel insights into calibrating and using advanced cyclic models for monopile analysis and design – particularly, with regard to the quantitative influence of pile installation effects and sand's microstructural evolution under drained cyclic loading.@en

Many constitutive models are nowadays available to simulate the mechanical behaviour of clays under different loading conditions. For geotechnical engineering applications in a soil dynamics context, the PM4Silt model, as proposed by Boulanger & Ziotopoulou (2018), can simulate the cyclic behaviour of low-plasticity silts and clays. Formulated in a bounding surface plasticity framework, this model adapts the capabilities of a previous model suited for sands and non-plastic silts, i.c. PM4Sand, and enables to simulate more clay-like soil behaviour. In this paper, published calibrations for silts with different plasticity properties are employed to simulate the experimental results of monotonic and cyclic undrained DSS test with the purpose to comment on the effect of some key parameters, including the initial stress ratio 𝐾 and the bounding surface parameter . Numerical simulations are performed with the finite element implementation of PM4Silt in PLAXIS. Besides the two original methods to define undrained shear strength (based on 𝑆􀯨 and 𝑆), a third method is proposed here based on the SHANSEP approach. The application of the model in view of the key parameters is demonstrated by means of a practical example. @en