Impact of Turbulence Closures and Numerical Errors for the Optimization of Flow Control Devices

The simulation of turbulent flows including active flow control devices, such as synthetic jets, is still a difficult task. Numerical parameters (grid size, time step, etc.) may have a significant influence on the result, while the choice of the turbulence closure is often critical. In this context, we investigate the use of a Kriging-based global optimization method to determine optimal control parameters. The objective of this study is twofold: firstly, we quantify the impact of some numerical and modeling parameters on the Kriging model constructed using a design of experiment approach. In a second phase, we conduct an optimization process and measure the impact of numerical and modeling errors on the optimal control parameters found. An approach to account for some numerical errors during the optimization is finally presented. The turbulent flow over a backward facing step, including a synthetic jet actuator, is considered as test-case. The time-averaged recirculation length is considered as control criterion, while jet frequency and amplitude are optimized.