Skip to main content
To KTH's start page

Bayesian optimisation and Gaussian-process regression applied to fluid problems

Time: Thu 2022-06-16 10.30 - 11.30

Location: Faxén, Teknikringen 8

Participating: Philipp Schlatter & Saleh Rezaeiravesh (KTH

Export to calendar

Anstract:

Bayesian optimization (BO) based on Gaussian process regression (GPR) is applied to different CFD (computational fluid dynamics) problems of practical relevance. We start from a simple explanatory case, and i.e. a shape optimisation in a lid-driven cavity to maximize the energy dissipation. We will then consider two real turbulence cases: First, the shape optimization of the wall of a diverging/converging channel flow in order to obtain a prescribed pressure gradient distribution along the edge of the turbulent boundary layer formed on the other wall. Finally, we demonstrate the optimisation on a more applied case, i.e. the optimization of the controlling parameters of an ice-spoiler model to attain the aerodynamic characteristics of the airfoil with an actual surface ice.
The diversity of the optimization problems, independence of the optimization approach from any adjoint information, the ease of employing different CFD solvers in the optimization loop, and more importantly, the relatively small number of the required flow simulations reveal the flexibility, efficiency, and versatility of the BO-GPR approach in CFD applications. It is shown that to ensure finding the global optimum of the design parameters of the size up to 8, less than 90 executions of the CFD solvers are needed. Furthermore, it is observed that the number of flow simulations does not significantly increase with the number of design parameters. The associated computational cost of these simulations can be affordable in many optimization cases even with practical relevance.