A Reduced-Order Discrete-Vortex Method for Flows with Leading-Edge Vortex Shedding

Abstract: The formation of the leading-edge vortex (LEV) is a key feature of unsteady flows past aerodynamic surfaces, but is expensive to model in high fidelity computations. Low-order methods based on discrete vortex elements are able to capture the physical behavior of these flows, in particular when enhanced with a criterion that models the ability of the leading edge to sustain suction. These models are significantly faster than high order methods, but their expense still grows as vortex elements are continuously shed and convected into the wake, in effect an $\mathcal{O}(n^2)$ problem. This work proposes accelerating the leading-edge suction parameter discrete vortex method (LDVM) by limiting the number of vortex elements in the LEV coherent structure to N, hence giving the name to the method N-LEV LDVM. The N-LEV LDVM method correctly approximates the flows in comparison with the original LDVM model and computational fluid dynamics (CFD) simulations until the point of LEV detachment, which N-LEV LDVM is unable to model. We propose reintroducing this behavior via two physical detachment criteria studied in LEV literature, a threshold of maximum circulation in the LEV and trailing edge flow reversal. We demonstrate the ability of the N-LEV LDVM method to accurately predict the instant in time this detachment occurs for both mechanisms in comparison with experimental results, laying the ground for their incorporation into the method.