Application of Second-Order Cone Programming Finite Element Methods for Laterally Loaded Monopiles in Offshore Wind-Paper

Paper

Pdf Summary

The paper evaluates second-order cone programming finite element (SOCP-FE) methods for modelling offshore wind monopiles under lateral loading. Using the commercial OPTUM G3 software, the authors benchmarked simulations against the Cowden field tests from the PISA research project. The analyses used a calibrated anisotropic undrained clay model and mixed finite elements, with remeshing to improve solution accuracy.<br /><br />The simulations reproduced the main observed response of short, large-diameter piles: rigid-body rotation, separation at the pile–soil interface, displacement discontinuities on the active sides, and toe-kick resistance. Predicted load–displacement behaviour showed generally good agreement with the field data.<br /><br />A 23-case numerical study examined meshes of 1,000–6,000 elements and up to three remeshing stages. Accuracy and computational cost depended on both mesh size and remeshing; cases without remeshing were generally substantially less accurate. A 2,000-element case with two remeshing stages ran about 12 times faster than the 6,000-element reference case, with a comparatively low average error. Differences of roughly 10–15% between solutions were typical, though accuracy did not improve consistently with mesh size alone.<br /><br />The authors conclude that iterative SOCP-FE methods can provide efficient, sufficiently accurate analyses for monopile design and parametric studies. They recommend balancing mesh size and remeshing effort, with finer meshes for highly mobilized soil regions.

Keywords

second-order cone programming

finite element methods

offshore wind monopiles

lateral loading

OPTUM G3

PISA research project

Cowden field tests

anisotropic undrained clay model

remeshing

load-displacement behaviour