Determination of stress concentration factors in offshore wind welded structures through a hybrid experimental and numerical approach

Kolios, Athanasios and Wang, Lin and Mehmanparast, Ali and Brennan, Feargal (2019) Determination of stress concentration factors in offshore wind welded structures through a hybrid experimental and numerical approach. Ocean Engineering, 178. pp. 38-47. ISSN 0029-8018

[img] Text (Kolios-etal-OE-2019-Determination-of-stress-concentration-factors-in-offshore-wind-welded)
Kolios_etal_OE_2019_Determination_of_stress_concentration_factors_in_offshore_wind_welded.pdf
Accepted Author Manuscript
Restricted to Repository staff only until 8 March 2020.
License: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 logo

Download (1MB) | Request a copy from the Strathclyde author

    Abstract

    Offshore wind turbine (OWT) monopile support structures generally consist of steel cans connected together through circumferential welding joints. One critical factor to evaluate the localised increase in stresses is the stress concentration factor (SCF) which depends on the welding quality. The complex welding profiles in OWT monopiles makes the accurate calculation of SCF quite challenging. In this work, an innovative approach for the calculation of SCFs in offshore welded structures is proposed based on combined 3D (three-dimensional) laser scanning technology (LST) and 3D finite element analysis (FEA). The precise geometry of the welded specimens is captured using 3D LST, and then imported into a finite element software to perform 3D FEA modelling to accurately calculate SCFs. A 2D (two-dimensional) FEA model of a typical offshore welded structure with ideal geometry is also developed in this work. In addition to numerically calculate SCFs, the 2D FEA model is further combined with non-linear RSM (response surface method) to derive analytical equations, expressing SCFs of offshore welded structures in terms of key welding parameters. Both LST-FEA3D and RSM-FEA2D models are applied to calculate SCFs in large-scale S-N fatigue welded specimens. The results indicate that the LST-FEA3D approach is capable of capturing the variation of SCFs along the width of the welded specimens and identifying the critical points where fatigue crack is most likely to initiate; and the RSM-FEA2D is valuable and efficient in deriving analytical parametric equations for SCFs.