Coherent structures modeling in stenotic transitional flow via resolvent analysis

Abstract

This study investigates the capability of linear modeling to characterize the transitional dynamics in an axisymmetric stenosis and attempts a low-order representation of the turbulent stresses. The transition to turbulence in stenotic flows generates wall shear stress fluctuations that strongly influence the progression of cardiovascular diseases and the risk of plaque rupture. A description of the linear mechanisms driving the forced dynamics at Reynolds number beyond transition is currently missing. Linear modeling of coherent structures is leveraged to identify the flow amplification mechanisms using the mean field from a LES at Re=4000. Global linear stability analysis reveals an unstable and sinuous stationary eigenmode that is known to destabilize the flow at lower Reynolds numbers through a weak Coanda-type wall attachment. At intermediate frequencies, resolvent analysis identifies a second amplification region within the shear-layer where the most amplified fluctuations are axisymmetric, in contrast to findings from studies at lower Reynolds numbers. The linear model is validated against SPOD. At intermediate frequencies, the optimal resolvent response mode demonstrates both high gain separation and strong alignment with the leading SPOD mode. The low-rank nature of the resolvent operator is leveraged to reconstruct the turbulent kinetic energy (TKE) and turbulent wall shear stress (tWSS) from the optimal response mode. In the immediate post-stenotic zone, axisymmetric fluctuations dominate the tWSS and exhibit low-rank dynamics. Our findings highlight that linear mechanisms effectively capture the complex post-stenotic dynamics. The successful reconstruction of turbulent quantities from mean flow data alone opens new predictive possibilities of key turbulent quantities.