Lattice Boltzmann Simulations of the Flow Over a Hump with Flow Control

Abstract

The wall-mounted hump test case was first presented as a benchmark problem for active flow control at the NASA sponsored workshop CFDVAL2004. For this case, both steady suction and oscillatory blowing-suction are used to control the separation and reattachment of the turbulent flow over the hump. In this paper the lattice Boltzmann method, coupled with a very large eddy simulation (VLES) turbulence model, is used to predict three cases: uncontrolled flow, controlled flow using steady suction, and controlled flow using oscillatory blowing-suction. As the lattice Boltzmann method is an inherently unsteady method it is uniquely suitable for predicting separated flows as well as flows with transient boundary conditions. We compare reattachment locations with experiments and previous CFD results. Profiles of velocity and turbulent kinetic energy in the recirculation zone and the recovery zone are also compared with experiments. Comparisons with PIV data at four phases in the oscillatory cycle are made to evaluate the accuracy of the predicted flow structures. Simulations show very good agreement with the experiments for the uncontrolled and oscillatory controlled cases, with the decrease in separation length achieved with oscillatory control accurately predicted by the lattice Boltzmann-VLES method. Simulations were found to be less accurate for the steady suction case; a reduction in recirculation length is predicted, although the reduction is smaller than observed in experiments.