Model-based spectral coherence analysis

Recent data-driven efforts have utilized spectral decomposition techniques to uncover the geometric self-similarity of dominant motions in logarithmic layer, and thereby validate attached eddy model. In this paper, we evaluate predictive capability stochastically forced linearized Navier–Stokes equations capturing such structural features turbulent channel flow at $Re_\tau =2003$ . We use linear coherence spectrum quantify wall-normal within velocity field generated by dynamics. addition around mean profile, consider an enhanced variant which molecular viscosity is augmented with eddy-viscosity. judiciously shaped white- coloured-in-time stochastic forcing generate a statistical response energetic attributes that are consistent results direct numerical simulation (DNS). Specifically, white-in-time scaled ensure two-dimensional energy reproduced match normal shear stress profiles. show eddy-viscosity significantly strengthens self-similar resulting layer leads inner-scaled spectrum. extract signature actively contribute momentum transfer responsible for producing Reynolds stress. Our findings support conjunction dynamic damping afforded improving predictions scaling trends associated active accordance DNS-based decomposition.