A numerical demonstration of dynamic stall control

Abstract

This paper presents a numerical demonstration of the real-time application of two dynamic stall onset criteria for identifying and mitigating stall. These criteria - based on the leading-edge suction parameter (LESP) and boundary enstrophy flux (BEF) - are derived from prior research. The present work establishes a proof of concept for the practical use of these indicators in mitigating dynamic stall. Two different unsteady motions that lead to dynamic stall are simulated. For each motion, both baseline cases (where dynamic stall occurs) and controlled cases (where stall is mitigated using the onset criteria) are analyzed using the unsteady Reynolds-averaged Navier-Stokes (uRANS) method. Both parameters were found to be effective in mitigating the adverse effects (large variations in aerodynamic loads) of dynamic stall. The results demonstrate the potential for these physics-based criteria to be integrated with advanced control strategies, such as reinforcement learning agents, for robust stall control across a range of operating conditions.