Hugo Dornier, PhD student, Platon Team (joint with Onera)
Title: Robust and efficient CFD simulation of the ARL-SL19 supersonic cascade through adaptive mesh refinement
Abstract:
The presentation will focus on an efficient adaptive mesh refinement method able to generate automatically high quality grids that allow accurate CFD simulations for supersonic turbomachinery flow configurations. The chosen test case is the well documented ARL-SL19 supersonic cascade. First, a numerical error analysis and a validation against reference results are performed to quantify the reliability of the numerical method. The next part consists in studying two mesh adaptation approaches based on a shock sensor and another one based on wake and recirculation. The results from these methods are compared to those obtained with one relying on the more global Mach number sensor. After defining a relevant adaptation strategy, a comparison is made of the different approaches performance. Finally, the shock and the turbulent features sensors are used simultaneously to find the best combination ratios.
