Efficient numerical methods to compute unsteady subsonic flows: A new refinement strategy for grid adaptation and a tailored Jacobian-free Newton-Krylov algorithm for unsteady flow problems

Despite the large increase in computer power and the advances in solver technology, unsteady flow problems still may require a vast amount of computing time. In this work two approaches are investigated that should lead to a reduction in computing times for unsteady subsonic flows. In the first part of this work a new refinement strategy for grid adaptation is proposed that enables the user to a priori define the maximum usable computer resources. In the second part of this work a Jacobian-free Newton-Krylov algorithm is tailored for unsteady subsonic flow problems. With this algorithm computing times are reduced with one order of magnitude compared to the traditional used solver.

Peter Lucas as well as Hester Bijl both have been awarded the degree of doctor at the Delft University of Technology. Their expertise lies in the field of computational fluid dynamics. Hester Bijl is currently a full professor at the Delft University of Technology.