Critical review of turbulence models for CFD for fatigue analysis in large steel structures. Issue 4 (30th January 2018)

• Record Type: Journal Article
• Authors: Lorenzon, A.; Antonello, M.; Berto, F.
• Other Names: Iacoviello Francesco guestEditor.; Susmel Luca guestEditor.; Berto Filippo guestEditor.
• Is Part Of: Fatigue & fracture of engineering materials & structures. Volume 41:Issue 4(2018)
• Page Start: 762
• Page End: 775
• Publication Date: 2018-01-30
• Subjects: computational fluid dynamics -- turbulence models -- wind engineering -- large steel structures -- wind‐induced fatigue; Materials -- Fatigue -- Periodicals; Fracture mechanics -- Periodicals; 620.1123
• DOI: 10.1111/ffe.12780
• Languages: English
• ISSNs: 8756-758X

Abstract

Abstract: The determination of wind‐induced fatigue load in large structures and buildings with complex geometries, which do not fall in typical cases of building codes and standards, represents a consuming task, and it is usually evaluated through the aid of wind tunnel tests, while computational approaches do not yet have a large application. The practical use of computational fluid dynamics (CFD) techniques in civil engineering applications has in fact been rare outside the framework of steady‐state analyses, because of the limitations of RANS approaches and of the higher cost of LES, but with the increase in computing resources and with the introduction of new and more efficient turbulence models, the practical use of CFD to evaluate fluctuating features of wind flow is much more feasible now. With it, it is possible to foresee the possibility for the evaluation of wind‐induced fatigue loads in steel structures by using CFD. This paper has the purpose to discuss the feasibility of a procedure that uses CFD as the first step in a chain of numerical simulations that leads to the fatigue calculation of a large, complex structure. The state of the art of turbulence models for CFD is here shown by performing literature review, with particular attention to their application to case studies related to large steel structures. A special focus in this paper lies in the use of PANS models, which reduce computational cost and grid sensitivity compared with LES, while providing comparable accuracy.