Abstract:: TH RD ND
Abstract:: TH RD ND
For unstructured meshes, 4th order implicit RK-3rd order backward difference-2nd order
backward difference is analyzed for strongly coupled system. Time integration schemes are
constructed in a way that moving meshes DGCL is satisfied. To compare this schemes wing
flutter is used and efficiency & accuracy is obtained.
Flow
DGCL for RK
Here goal is to study the temporal accuracy of the proposed schemes, and we use a fully
coupled approach to eliminate any other possible sources of error that may contaminate
accuracy.
Acceleration Strategies and Computational Cost
• Agglomeration multigrid methods are used
to accelerate the solution of the nonlinear
problem arising at each time step of the
unsteady-flow equations, as well as for
solving the mesh motion equations.
• The idea of multigrid is to accelerate the
solution on a fine grid by iteratively
computing corrections to the fine grid
problem on coarser grid levels for which
the cost of the iterations is lower and the
global error components are more easily
reduced.
Results For fixed number of sub iterations
• The AGARD 445.6 wing test case , which was tested in the transonic dynamics tunnel at NASA Langley
Research Center, has been widely used as a benchmark for evaluating aeroelastic simulations.
• To find the flutter boundary, the structure responses are analyzed using the modal
identification technique of Bennett and Desmarais . A nonlinear least-squares
curve fit method in the form of Eq. is used to fit the response curves:
Conclusion
• The time histories of the wing deflection computed using the BDF2,
BDF3, and IRK64 schemes have been compared and the IRK64
scheme shows superior overall efficiency.