Session: 20-17-01: Rising Stars of Mechanical Engineering

Paper Number: 172725

High-Fidelity Numerical Experiments of Pulsating Turbulent Flows 

Turbulent wall-bounded flows in nature and engineering applications are often subjected to complex multiphysics effects that deviate the flow from the canonical boundary layer behavior. Examples include pressure gradients (both steady and unsteady), surface roughness or microstructures, and external forces such as wall oscillations in non-inertial systems and the Coriolis effect in rotating systems. These influences introduce non-equilibrium modulations to the turbulence, making the flow behavior difficult to predict. Current prediction tools often rely on idealized assumptions - such as smooth surfaces and equilibrium boundary layers - which can result in significant errors or designs that perform well in controlled environments but fail under realistic conditions.

Our research group aims to advance the understanding of non-equilibrium turbulence through high-fidelity simulations, including direct numerical simulation (DNS) and wall-resolved large-eddy simulation (LES). These simulations resolve the full range of flow scales using several hundred million to over ten billion grid cells, providing unprecedented datasets. These data not only enable detailed statistical analysis to elucidate the underlying flow physics, but also serve as benchmark databases for model development and the training of data-driven predictive tools.

In this presentation, we focus on the combined effects of flow pulsation and surface roughness. The unsteady nature of the driving force fundamentally alters how surface roughness influences drag. Additionally, the presentation will include highlights from our ongoing investigations into roughness–pressure gradient interactions and flow separation phenomena. 

A limited understanding of unsteady flows over rough surfaces has led to significant economic and societal consequences across a wide range of sectors. For instance, modern aircraft are typically optimized for steady cruise conditions, yet in reality, airflow around the airframe and through engines constantly fluctuates due to atmospheric variability and flight maneuvers. Sudden wind gusts over crop canopies can strip leaves or break stalks, severely impacting yields. Pulsed currents over oyster reefs can disrupt larval settlement and bury oysters under sediment, threatening reef sustainability. Similarly, gusty winds can accelerate wildfire spread across complex terrain, complicating firefighting efforts. Our study seeks to close the fundamental knowledge gap by investigating the underlying physics of these unsteady roughness-affected flows.

DNS is conducted using an immersed boundary method based on the volume-of-fluid approach to represent complex surface geometries. The Cartesian grid is carefully constructed to ensure that even the smallest surface curvatures are resolved with sufficient grid points. Flow pulsation is imposed via sinusoidal pressure fluctuations at various frequencies, selected based to achieve various the Stokes layer (the near-wall region where the interaction between the pulsating freestream and the wall-damping leads to magnitude and phase interferences) and the potential overlap with the roughness sublayer, buffer layer, and logarithmic region.

Our results reveal that pulsation significantly enhances the unsteady wake (i.e., local recirculation and vortex shedding) around individual roughness elements. These spatial and temporal fluctuations - strongly influenced by surface geometry and local flow conditions - play a critical role in transferring energy from the mean flow to the turbulent fluctuations. We find that this turbulence production mechanism is fundamentally altered under pulsation, leading to both a thickening and upward shift of the Stokes layer. These findings underscore the need for near-wall model corrections even under modest deviations in flow forcing. The detailed characteristics of the unsteady wake captured in our simulations inform the development of physics-based predictive tools that incorporate surface topology effects for improved accuracy.

Presenting Author: Wen Wu University of Mississippi

Presenting Author Biography: Dr. Wen Wu has been an Assistant Professor at the University of Mississippi (UM) since 2020. Before joining UM, he completed his Ph.D. at Queen’s University, Canada, and conducted postdoctoral research at Johns Hopkins University. He also served as a Visiting Assistant Professor at the Center for Turbulence Research at Stanford during the summer of 2022 and 2023. Dr. Wu combines high-fidelity simulations and theoretical analysis for his research, with an emphasis on the multiscale and non-equilibrium aspects of turbulence. His recent research has focused on vortex-dominated flows, flow separation, and surface roughness.

Authors:

Wen Wu University of Mississippi
John Cooper Univesity of Mississippi
Benjamin Savino University of Mississippi