A method for identification of an effective Winkler foundation for large-diameter offshore wind turbine support structures based on in-situ measured small-strain soil response and 3D modelling. (1st October 2016)

• Record Type: Journal Article
• Authors: Versteijlen, W.G.; Metrikine, A.V.; van Dalen, K.N.
• Is Part Of: Engineering structures. Volume 124(2016:Oct. 01)
• Page Start: 221
• Page End: 236
• Publication Date: 2016-10-01
• Subjects: Offshore wind foundations -- Fundamental natural frequency -- Small-strain soil stiffness -- In-situ seismic measurement -- 3D to 1D modelling translation; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2016.06.007
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: In-situ seismic measurements are used to derive the shear modulus of the soil. We suggest an inversion method based on Snell's law for wave refraction. The derived shear modulus is used as input in a 3D linear-elastic soil–pile model. We present a method for translating the 3D model to an equivalent 1D model. The full procedure is presented for a design location with mostly stiff sand. Abstract: A procedure is presented for the derivation of an effective small-strain soil stiffness governing the soil–structure interaction of large-diameter monopiles. As a first step, geophysical measurements are used to estimate the depth-dependent shear modulus G of the soil stratum. The second step is to use this modulus and an estimated Poisson's ratio and density in a 3D model, which captures the deformation of both the monopile and the soil. As a final step, a new method is proposed to use the computed 3D response for identification of a depth dependent stiffness of an effective Winkler foundation. This stiffness can be used in a 1D model, which is more fit for design purposes. The presented procedure is deemed more appropriate than the often used " p – y curve" method, which was once calibrated for slender flexible piles and for which the input is based on the large-strain cone penetration test. The three steps are demonstrated for a particular design location. It is also shown that the displacements of the 3D model are smaller and the resulting fundamental natural frequency is higher than calculated with the p – y method.