National and Kapodistrian University of Athens
Dynamics and structures of S3T theory in processes of wall-bounded shear flow turbulence
Abstract
dc:descriptionHydrodynamic flows, which are most commonly in a turbulent state, are ever-present in our environment and affect crucially our life. A fundamental property of turbulent flow is the emergence in these flows of self-organized large scale coherent structures. In wall-bounded turbulent flows these coherent structures assume the form of prominent vortices, which are closed circuitous flows in vertical cross-sections of the flow with a streamwise-oriented rotation axis, accompanied with collocated streaks, which are alternating lanes of low- and high- speed fluid in the streamwise direction. This thesis employs a mean field theory motivated by the statistical state dynamics (SSD) of a second order closure formulation (S3T) in order to investigate the mechanism that maintains an inhomogeneous mean flow (which is obtained as the average across the, assumed as homogeneous, streamwise direction) in turbulence and specifically is responsible for the formation of streamwise vortices. The mechanism forming the vortices is central in the self-sustaining process (SSP) of turbulence, with earlier studies considering modal instabilities or optimal streak transient growth arising in the perturbation field (streamwise-varying deviations of the mean flow) as the source of vorticity regeneration. An alternative explanation was provided in the simplified SSD framework of the restricted nonlinear (RNL) dynamics of the Navier-Stokes (NS) equations, where SSP dynamics are reproduced even in the absence of the perturbation-perturbation nonlinearity acting on the perturbation field dynamics. Consequently, this approximation of the perturbation field dynamics relies explicitly on interactions occurring between the mean flow and the perturbation field. The SSP obtained in RNL revealed that the time-dependent mean flow streaks organize the Reynolds stresses of the perturbation field to induce torques that systematically coincide and accelerate the existing mean flow vortices. The approach adopted in this work aims to advance the statistical similarities of this mechanism in numerical simulations of NS and RNL turbulence. In order to reveal the influence of the mean flow on the perturbation component and to study the mean flow dynamics, we modify appropriately the collection of numerical data to collocate the low speed streaks in the center of the spanwise direction and retain spanwise inhomogeneity in the statistics. Eigenvectors (POD modes) of the perturbation field resulting from eigenanalysis of the conditional time mean covariance posses sinuous or varicose mirror symmetries in the spanwise direction. We show that in a stochastic turbulent model, where the collocated time-averaged mean flow is chosen as the background flow, these disturbances grow unevenly due to streak transient growth. In particular, the presence of the low speed streak induces higher energy gains to the sinuous perturbations compared with their varicose counterparts, which is consistent with the energy weight attributed to the POD modes. This uneven growth is reflected in the amplitude of the Reynolds stress nonlinear forcing that supports the streamwise mean vorticity, with the reinforcing stresses to the low-speed streak induced by the sinuous perturbations overpowering the opposing stresses induced by the varicose perturbations (and vice-versa in high speed streaks). We find that this imbalance in the growth of these perturbations is the salient physical mechanism supporting the underlying roll-streak coherent structure and therefore maintaining the spanwise inhomogeneity of the mean flow. Additionally, we demonstrate a synchronization property occurring between subspaces of the dynamics comprised by low and high wavenumber streamwise Fourier harmonics, which labels the controlling subspace as the active component of turbulent dynamics. We show that a universal cutoff wavenumber can be detected for the passive subspace. This is achieved by computing numerical solutions of the NS and calculating the parametric stability of infinitesimal deviations on passive subspaces. Synchronization experiments identify the subspace that is responsible for the chaotic dynamics of turbulence and associate this subspace with energy extraction from the largest scales of the flow.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Nikolaidis, Marios-Andreas
- Νικολαΐδης, Μάριος-Ανδρέας
Subjects
dc:subject × 26- Διατμητικές ροές
- Τύρβη
- Αναγεννητικός κύκλος
- Συνεκτικές δομές
- Στρόβιλοι
- Θεωρία μέσου πεδίου
- Προσομοιώσες τύρβης
- Ανάλυση POD και CPOD
- Δυναμική μέσης ροής
- Συγχρονισμός
- Shear flow
- Turbulence
- Self-sustaining process
- Coherent structures
- Vortices
- Mean field theory
- Turbulence simulation
- POD and CPOD analysis
- Mean flow dynamics
- Synchronization
- Φυσική
- Φυσικές Επιστήμες
- Φυσική ρευστών και πλάσματος
- Physical Sciences
- Natural Sciences
- Physics, Fluids and Plasmas
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/53709