Machine learning control for experimental shear flows targeting the reduction of a recirculation bubble

International audienceThe goal is to experimentally reduce the recirculation zone of a turbulent flow(Re H = 31500). The flow is manipulated by a row of micro-blowers (pulsed jets) that areable to generate unsteady jets proportional to any variable DC. Already, periodic jet injectionat a forcing frequency of St H = 0.226 can effectively reduce the reattachment length andthus the recirculation zone. A model-free machine learning control (MLC) is used to improveperformance. MLC optimizes a control law with respect to a cost function and applies geneticprogramming as regression technique. The cost function is based on the recirculation lengthand penalizes actuation. MLC is shown to outperform periodic forcing. The current studydemonstrates the efficacy of MLC to reduce the recirculation zone in a turbulent flow regime.Given current and past successes, we anticipate numerous experimental MLC applications