A nonlinear computational aeroelasticity model for aircraft wings

The design of a large aircraft at high speeds is a challenge in the very active research on aeroelasticity. The computational aeroelasticity analysis in 3D for such a large system of nonlinear equations becomes available due to the recent highly developed computing technology. This thesis deals with the development of a CFD-based coupling code which is based on the equations of the structural motion and the Euler equations of inviscid compressible transonic flows. The strategy of segregating such a complex multidisciplinary system gives the advantages of the software development in modularity and the reuse of the developed solvers of the subsystems. The non-matching of the grids on the fluid-structure interface due to the difference of the element sizes and types of the fluid and the structural models is resolved by adding the matcher module in the coupling algorithm. The information transfers from one solver to another satisfy conservation of energy. The nonlinear aerodynamic model is described by the kinematic ALE description and discretized on the moving mesh which is updated by the mesh solver. The CSD-MAM model in which the modal superposition approach is based on the linear structural theories is used to reduce the computing time and the memory consumption. Another comparable CSD-FEM model based directly on the finite element discrete approach is also built for the extension to general structural dynamics. The nonlinearity is another source of the complexity of the aeroelasticity model although it is assumed only from the aerodynamics of the transonic flow and from the geometric nonlinearities due to the mesh motion. The nonlinear GMRES algorithm with the ILUT preconditioner is implemented in the robust CFD solver where the SUPG numerical stabilization techniques and a shock captor are applied to the transonic flow dominated by convection. The second order Gear-Scheme is used for the time discretization. The components of this nonlinear computational aeroelasticity model are validated one by one with numerical experiments. The complete model is validated by the AGARD 445.6 aeroelastic wing immersed in transonic flows with Mach number 0.96 which corresponds to the lowest point of the transonic dip. The flutter simulations have given satisfying results compared to experimental ones