Modern architecture promotes light and efficient structures. With the use of innovative constructions and materials, the realization of wide-spanned and creative buildings is possible. However, increasing lightness and slenderness bring along a higher susceptibility to wind effects, which can become the decisive design factor. An accurate assessment of these wind effects with deterministic tools is complicated, in particular in the case of aeroelastic phenomena. In this regard, numerical multiphysics simulations are a promising complement and enhancement to elaborate experimental approaches.
The long-term aim of this research is to propose a methodology for the analysis and improvement of light, thin-walled structures, such as thin shells and membrane roofs towards wind effects. The focus is on the appropriate combination of different physical and numerical disciplines to account for the relevant factors inherent to the simulation of light, thin-walled structures as well as highly turbulent air flows. To fulfill these requirements the occurring wind-structure interaction is accessed by a surface-coupled fluid-structure interaction (FSI) method. This is realized in a modular and flexible software environment with the use of a partitioned coupling approach: the structural field is solved by the in house finite element program CARAT using several finite element types and advanced solution strategies for form finding, nonlinear and dynamical problems. The fluid field is solved by the CFD software package CFX-5 of ANSYS Inc. Additional care towards the realistic modeling of physical wind is taken. A prerequisite to allow for the assessment of aeroelastic problems, beyond the mere exchange of data between the two physical fields, is the utilization of stable as well as efficient coupling strategies. Moreover, the comprehensiveness of this approach opens the possibility for multiphysics optimization.
The contribution will present theory and realization of an implementation enhanced by illustrative examples. Strategies for the extension of the approach towards multiphysics optimization will be presented.