The python script will drive the optimization process by executing flow solutions, adjoint solutions, gradient projection, geometry evaluations, and mesh deformation in order to drive the design toward an optimum. To best showcase the difference between the windowing function, the original configuration is recommended. Note, that this configuration may produce different designs. %Number of iterations to average the objective %Iteration number to begin the reverse time integration in the direct solver for the unsteady adjoint. Here we set ITER_AVERAGE_OBJ=TIME_ITER-WINDOW_START_ITER=700. The time to average the objective and constraint function is given by the option ITER_AVERAGE_OBJ. We set the start iteration to the final iteration of the direct run, i.e. it starts at iteration given by UNST_ADJOINT_ITER and ends at iteration 0. Note, that the adjoint iterator runs backwards in time, i.e. Make sure that the option INNER_ITER is chosen big enough in your test case to get It may happen that the adjoint inner iterator needs more iterations to reach a steady state. Asymptotically, the convergence speed of the adjoint inner iteration matches the speed of the direct inner iteration. To the one used for the direct simulation. This means, sensitivies are calculated using a dual time-stepping method similar To compute the sensitivity of the optimization objective and constraint, SU2 uses an adjoint iterator. Options (SQUARE, HANN, HANN_SQUARE, BUMP), SQUARE is default. % Window-function to weight the time average. % Iteration to start the windowed time average More information about setting up unsteady simulations can be found here To compute the unsteady shape optimization, we set up the unsteady simulation according to our test case above. The period average is approximated by a windowed time-average over a finite time-span \(M\) A meaningful objective and constraint function is therefore a time average over a period. We want to solve an optimization problem with a time dependent system output, e.g. These subsonic flow conditions will cause a detached flow about the airfoil, that exhibts a vortex street and is therefore periodic for the baseline geometry.ĭepending on the windowing-function used to average the optimization objective, the flow about the optimized geometry will eventually be a steady state flow.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |