Generation of a general Cowden clay model (GCCM) using a data-driven method

Kamas, I and Burd, Harvey J. and Byrne, Byron W. and Suryasentana, Stephen (2023) Generation of a general Cowden clay model (GCCM) using a data-driven method. In: Session on 'Pile Design', 2023-09-12 - 2023-09-14, London.

[thumbnail of Kamas-etal-OSIG-2023-Genertion-of-a-general-Cowden-clay-model]
Preview
 Text. Filename: Kamas_etal_OSIG_2023_Genertion_of_a_general_Cowden_clay_model.pdf
Accepted Author Manuscript
License: Strathprints license 1.0
Download (1MB)| Preview

Abstract

Monopiles are the most widespread foundation option for offshore wind turbines. The PISA (Pile Soil Analysis) joint industry research established a one-dimensional (1D) model for calculating the monotonic behaviour of laterally loaded monopiles by using a parametric equation to represent the soil reaction components. During the current PICASO project, a new data-driven method has been developed, as an alternative form of the PISA design model, in which Gaussian process regression (GPR) is employed to formulate the soil reaction curve data that are required to drive the 1D model to predict the monopile’s monotonic behaviour. This paper presents a general Cowden clay model (GCCM) generated by using this data-driven method. The 1D model employing the GCCM can be used to predict the initial stiffness and ultimate capacity of piles embedded in homogenous glacial clay till soils within a range of different stress histories.

ORCID iDs

Kamas, I, Burd, Harvey J., Byrne, Byron W. and Suryasentana, Stephen ORCID logoORCID: https://orcid.org/0000-0001-5460-5089;
CORE (COnnecting REpositories)