Abstract
In this paper, an accurate model to simulate the dynamics of the flow of synthetic jets (SJ) in quiescent flow is proposed. Computational modeling is an effective approach to understand the physics involved in these devices, commonly used in active flow control for several reasons. For example, SJ actuators are small; hence, it is difficult to experimentally measure pressure changes within the cavity. Although computational modeling is an advantageous approach, experiments are still the main technique employed in the study of SJs due to the lack of accurate computational models. The same aspect that represents an advantage over other techniques also represents a challenge for the computational simulations, such as capturing the unsteady phenomena, localized compressible effects, and boundary layer formation characteristic of this complex flow. One of the main challenges in the simulation of SJs is related to the fact that the spatial and temporal scales of the actuator and the corresponding flow control application differed in several orders of magnitude. Hence, in this study we focus on the use of Computational Fluid Dynamics (CFD) and Reduced Order Models (ROM) to develop an accurate yet low-cost model to capture the complexities of the flow of a SJ in quiescent flow. Numerical results show two possible paths for SJ modeling; (1) to obtain a boundary condition to predict velocity profile and jet formation from experimental data of diaphragm’s deformation; and, (2) to predict peak velocity at the jet’s outlet with a ROM approach and to use the physical details of the actuator to develop an accurate boundary condition for CFD. Both approaches are validated through experimental data available in the literature; good agreement between results from CFD, Lumped Element Model (LEM), and experimental data are achieved. Finally, it was concluded that the coupling between LEM and CFD is a novel and accurate approach, which improves CFD due to the advantages of LEM closing the gap between LEM’s lack of flow detail and CFD’s lack of geometrical/physical information of the actuator.