CTR Seminars Archive 2016
Fractional PDE Modeling of Anomalous Transport with Applications in Fluids
Date and Time: Friday, November 11, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Mohsen Zayernouri, Assistant Professor of Computational Mathematics, Science, & Engineering (CMSE) and Mechanical Engineering (ME) at Michigan State University (MSU)
Fractional PDE models generalize the standard (integer-order) calculus and PDEs to any differential form of fractional orders. Fractional PDEs open up new possibilities for robust mathematical modeling of physical processes that exhibit anomalous (sub- or super-) diffusion, nonlocal interactions, self-similar structures, long memory dependence, and power-law effects. Fractional PDEs are emerging as the right tool for exploring fractal operators and for modeling sharp interfaces in multi-phase problems, wave propagation in disordered media, and multi-scale materials. Such phenomena occur in many applications, including non-Gaussian (Levy) processes in turbulent flows, non-Newtonian fluids and rheology, non-Brownian transport phenomena in porous and disordered materials, and non-Markovian processes in multi-scale complex fluids.
In such applications, fractional PDEs naturally appear as the right governing equations leading to multi-fidelity modeling and predictive simulations, which otherwise cannot be achieved by employing the standard PDEs. However, the extension of existing numerical methods to fractional PDEs is not trivial because of their non-local and history-dependent nature. To this end, we first present a new theory on Fractional Sturm-Liouville Problems, which serves as a fundamental spectral theory providing explicit (non-polynomial) eigenfunctions, namely as "Jacobi Poly-fractonomials." These eigenfunctions extend the well-known family of Jacobi polynomials to their fractional counterparts. Based upon this base fractional spectral theory, we develop a series of high-order numerical methods that efficiently treat fixed-order, variable-order, and distributed-order fractional PDEs in low-to-high dimensions. Finally, introduced is the wealth of fractional-order modeling in the context of fractional conservation laws, chaotic quasi-geostrophic flows, zonal flows & incomplete mixing, in addition to the fractional turbulence modeling.
Bio:
Mohsen Zayernouri is an assistant professor of Computational Mathematics, Science, & Engineering (CMSE) and Mechanical Engineering (ME) at Michigan State University (MSU). Although he started as a passionate engineer, his mathematical appetite bent him to somehow a deeper academic direction. He obtained his first PhD in ME from the University of Utah in 2010, and subsequently he attended Brown University, where he obtained his second PhD in Applied Mathematics under the advice of Prof. George Em Karniadakis in 2014. Prior to joining MSU in 2015, he was a postdoctoral associate in the division of applied mathematics at Brown University. He is now the director of Fractional Mathematics for Anomalous Transport and Hydromechanics (FMATH) group at MSU, and recently, Mohsen Zayernouri received the 2017 AFOSR Young Investigator Program award.
------------------------------------
Bumblebees in turbulence: massively parallel numerical simulations
Date and Time: Friday, November 4, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Professor Kai Schneider, Centre de Mathematiques et d'Informatique, Aix-Marseille Universite
Insects fly even under heavy turbulent air flow conditions. To understand the impact of turbulent fluctuations on the aerodynamics of flapping wings, we model a bumblebee with fixed body and prescribed wing motion, flying in a numerical wind tunnel. The inflow condition of the tunnel varies from unperturbed laminar to strongly turbulent. Massively parallel simulations show that turbulence does not significantly alter the wing's leading edge vortex that is required for elevated lift production. Mean flight forces, moments and aerodynamic power expenditures are thus unaffected, suggesting little significance of turbulence on overall flight performance in insects. The increase in variance of the aerodynamic measures with increasing turbulence intensity, however, leads to flight instabilities in freely flying animals. This is joint work with Thomas Engels, Dmitry Kolomeskiy, Fritz-Olaf Lehmann and Jorn Sesterhenn.
Ref.:
T. Engels, D. Kolomenskiy, K. Schneider, F.O. Lehmann and J. Sesterhenn. Bumblebee flight in heavy turbulence. Phys. Rev. Lett., 116, 028103, 2016.
T. Engels, D. Kolomenskiy, K. Schneider and J. Sesterhenn. FluSI: A novel parallel simulation tool for flapping insect flight using a Fourier method with volume penalization. SIAM J. Sci. Comput., 2016, in press.
Bio:
Kai Schneider is a Professor of Mechanics and Applied Mathematics at I2M (Institute of Mathematics of Marseille), Aix-Marseille University, France. He obtained his Master degree in 1993 and his Ph.D. degree in 1996 both from Universitaet Kaiserslautern, Germany. He obtained his habilitation in 2001 from the Universit\'e Louis Pasteur, Strasbourg, France. His current research activities are focused on the development of multiscale and wavelet techniques for scientific computing and their application for modeling turbulent fluid and plasma flows. More recently he also got interested in bio-inspired fluid-structure interaction.
-----------------------------------------
Discontinuous Galerkin method for Computational Fluid Dynamics
Date and Time: Friday, October 28, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Dr. Yu Lv, Post-doctoral Fellow, Department of Mechanical Engineering, Stanford University
In recent years, variational-formulation based high-order schemes, such as discontinuous Galerkin (DG) scheme, have demonstrated promising capabilities for CFD applications. Compared to commonly used numerical methods, these high-order schemes are particularly attractive for (i) providing high-order accurate solutions with less sensitivity to mesh topology; (ii) enforcing physical realizability on solutions to guarantee numerical nonlinear stability; and (iii) better balancing computational robustness and resolution requirement for multi-physics simulations. This talk will start by illustrating several challenges associated with modern CFD applications, followed by a short tutorial introducing DG formulation and its numerical properties. DG’s performance will be demonstrated by considering canonical flow configurations, and compared against the commonly used numerical methods.
After the tutorial, the talk will focus on the further development of DG method for predicting discontinuity-containing flows (i.e. hypersonic flows up to Mach 18). Entropy-bounded DG scheme that has provable numerical stability will be briefly discussed, which serves as the foundation for simulations with explicit time integration. The talk will then move on to elaborate the recently developed artificial-viscosity method for predicting convective heat transfer on the surfaces of hypersonic re-entry vehicles. Emphasis will be given on how to obtain zero-residual implicit calculations while preserving high-order benefits. The developed DG-based numerical capability will be tested and validated in a wide range of flow applications, from LES of canonical turbulent flows and reacting flows, to detonation and hypersonic flows.
Bio:
Dr. Yu Lv is Postdoc Fellow working in the ME Department, Stanford University, under Professor Matthias Ihme. He joined Stanford in 2013 after finishing his Master’s degree in Aerospace Engineering (University of Michigan-Ann Arbor, 2011) and his Bachelor’s degree in Thermal Physics (Zhejiang University, China).
------------------------------------
Atomistic Scale Investigations of Supercritical Phase-Transitions
Date and Time: Friday, October 14, 2016 - 16:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Dr. Muralikrishna Raju, Post-doctoral Fellow, Department of Mechanical Engineering, Stanford University
Atomistic scale simulations are a powerful tool to study supercritical fluids and can provide information that is not accessible experimentally. The first part of this talk is an introduction to molecular dynamics (MD) simulations using interatomic potentials, commonly referred to as force fields. The second part of the talk presents results from MD simulations of supercritical solid-liquid phase transitions in nanoconfined ice and supercritical liquid-vapor transitions in binary mixtures.
Phase transitions in nanoconfined ice: We investigate the phase transitions of nanoconfined water in 2D graphene nanocapillaries and 1D carbon nanotubes using MD simulations in search for a solid-liquid critical point. We observe three regimes along the solid-liquid phase transition line:
(i) a sub-critical regime at low densities in which a first-order phase transition separates the solid and liquid phases
(ii) a continuous-transition regime at intermediate densities in which a solid-liquid continuous transition line that is characterized by a crossover in diffusivity separates the solid and liquid-like regions
(iii) a single phase regime at high densities where thermodynamic, structural and transport properties cannot distinguish the solid and liquid-like phases.
Widom line in binary mixtures: The second part of the talk is on supercritical liquid-vapor transitions in binary mixtures. Recent experiments have identified distinct liquid-like and vapor-like regimes in pure fluids under supercritical conditions. The supercritical liquid-vapor transition in a pure fluid occurs across an extension to the coexistence line, marked by almost discontinuously changing fluid properties. Nishikawa and Tanaka first identified this line, called the Widom Line, experimentally. Here, we perform MD simulations on binary Ar-Kr and Ne-Kr mixtures and present evidence for the existence of the Widom Line in binary mixtures. Interestingly, the Ar-Kr and Ne-Kr mixtures transition from a liquid-like to gas-like regime differently via distinct structural and dynamic pathways. The presence of distinct phase-transitions even in these ‘simple’ monoatomic binary mixtures highlights the complexity to be expected in higher order mixtures in real-life applications.
Bio:
Dr. Muralikrishna Raju received his PhD in Physics from Pennsylvania State University in 2015. During his PhD, he used molecular dynamics, Monte Carlo and accelerated molecular dynamics simulations using ReaxFF Reactive Force Fields to study Fluid/Solid Interfaces for applications in catalysis, nanocrystal growth, capacitive-mixing, desalination and Li-ion batteries.
---------------------------------------
Axially homogeneous buoyancy driven turbulence – free convection analog of fully developed turbulent pipe flow
Date and Time: Friday, August 12, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Professor Jaywant H. Arakeri, Indian Institute of Science, Bangalore
Turbulent convection driven by an unstable density difference created across the ends of long vertical tube has many interesting features. Away from the two tube ends, the flow, driven by a linear density gradient, is axially homogeneous. The mean velocities, and the mean Reynolds shear stresses are zero; thus the turbulent production due to shear is zero and is solely due to buoyancy. Near the tube walls, we have shear free turbulence, but which does not decay with time. This flow may be considered to be a free convection analog of pressure driven pipe flow. Both are axially homogeneous, one is driven by a linear density gradient and the other by a linear pressure gradient. Tube convection (TC) has several features, which make it different from the extensively studied Rayleigh-Benard convection (RBC). The scaling for RBC predicted by Kraichnan (1962) for the Nusselt number, Nu~Ra1/2 for very high Rayleigh numbers (Ra) and constant Prandtl number (also known as the ‘ultimate regime’), which has not been observed so far in experiments, is easily achieved in tube convection (TC) at relatively lower Ra. This scaling implies that the buoyancy flux is independent of viscosity and thermal diffusivity. Also, compared to RBC, orders of magnitude higher fluxes and turbulence Reynolds numbers are obtained in TC. Results from two types of experiments will be presented. The density difference in one set of experiments is created by brine and fresh water, and in another set by using heat. Besides the Nu~Ra1/2 scaling, we show that below a critical value of Grashof number a different scaling, Nu~Ra0.29, is observed at which is similar to the scaling observed in turbulent RBC. The scalar spectra are found to follow the Bolgiano-Obukhov (BO) scaling while the energy spectra of lateral and longitudinal velocity show Kolmogorov-Obukhov (KO) scaling. Also presented are some results from experiments of light propagation through the convective turbulence. Light propagation through convective turbulence is encountered in many situations, one example being stellar scintillation.
Bio:
Professor Jaywant H. Arakeri has been a faculty member of the India Institute of Science since 1988. He is currently a Professor in Mechanical Engineering, and Centre for Product Design and Manufacture. His research interests are in Fluid Mechanics and Heat Transfer, in particular stability, transition and turbulence, unsteady flows, flows with curvature, turbulent natural convection, and fluid mechanical phenomena associated with plants. His group has extensively studied the near-wall dynamics in turbulent Rayleigh-Benard convection and the axially homogeneous turbulent convection in a vertical tube. Studies of unsteady flows include instability, transition to turbulence and boundary-layer separation in decelerating flows, and pulsatile flows in highly curved tubes. The work related to plants is concerned with heat and moisture loss from leaves and from soil like porous media. Professor Arakeri received his Btech from IIT, Madras; ME from IISc, Bangalore; and PhD from Caltech, Pasadena, all in Aeronautical engineering.
-------------------------------------
The Conditional Source-term Estimation (CSE) approach to modeling MILD combustion
Date and Time: Friday, June 17, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Mr. Jeffrey W. Labahn, Mechanical Engineering, University of Waterloo
As governments realize the negative impact of greenhouse gasses on the planet, a drive to reduce our dependence on fossil fuels and switch to renewable clean energy sources has begun. However, while reducing our dependence on fossil fuels is required, this cannot be accomplished instantaneously without a significant negative impact on the world economy. Thus, to reduce the amount of greenhouse gasses in the near future, more efficient combustion processes are required to reduce both greenhouse gas emissions and pollutants such as nitrogen oxides (NOx). In conjunction with experimental investigations, numerical simulations of turbulent combustion are required to further understand the physical phenomena involved.
The present talk focuses on Moderate and Intense Low Oxygen Dilution (MILD) combustion as it increases efficiency, while reducing emissions. In this talk, I will present two turbulent combustion model formulations based on the principles of Conditional Source-term Estimation (CSE) which have been applied to MILD combustion. First, a non-adiabatic RANS-CSE formulation is applied to a semi-industrial furnace. Detailed predictions for temperature and emissions are obtained and compared to experimental data. Next, a multi-stream CSE formulation is applied to the Delft-jet-in-hot-coflow burners in the RANS and LES framework to determine if CSE can reproduce the physical characteristics seen in these flames. The present simulations demonstrate that these CSE formulations are able to predict the main characteristics seen in these semi-industrial and laboratory scale burners, including the ignition stabilization mechanism and emissions.
Bio:
Jeffrey W. Labahn is currently a Ph.D. candidate in Mechanical Engineering from the University of Waterloo. His research has focused on the development of turbulent combustion models to predict the behavior of laboratory burners and semi-industrial furnaces operating in the MILD combustion regime. His research interests include turbulence, MILD combustion, and combustion modeling.
---------------------------------------------
Dynamics of stratified flow past a sphere: simulations using body-inclusive numerical model.
Date and Time: Friday, June 3, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Mr. Anikesh Pal, Department of Mechanical and Aerospace Engineering, University of California San Diego
Wakes of bluff bodies in a stratified environment are common in oceanic and atmospheric flows. Some examples are marine swimmers, underwater submersibles and flow over mountains and islands. Direct numerical simulations of flow past a sphere in a stratified fluid at a sub-critical Reynolds number (Re) of 3,700 and for a range of Froude numbers, Fr = U/ND ∈ [0.025,∞] are performed. The conservation equations are solved in a cylindrical coordinate system and an immersed boundary method is employed to represent the sphere. The prime objective of this investigation is to understand the statistical response of the near, intermediate and far wake of a sphere at sub-critical Re under the influence of buoyancy. It is observed that buoyancy leads to the inhibition of vertical motion resulting in faster decay of r.m.s. velocity in the vertical direction as compared to the horizontal r.m.s. velocity, collapse of the wake, propagation of internal gravity waves and the organization of the primarily horizontal flow into coherent vortical structures. Unprecedented with respect to previous studies, the time averaged turbulent kinetic energy budget is closed for the unstratified and stratified cases. A novel finding of this research is the regeneration of turbulent fluctuations in the near wake when the stratification increases beyond a critical level (Fr decreases beyond a critical value) which is in contrast to the previous results at lower Re that suggest monotone suppression of turbulence with increasing stratification. Vorticity evolution, energy spectra and the turbulence energy equation explain turbulence regeneration. Another objective of this study is to quantify the distinction between the body and turbulence generated internal waves, in terms of the amplitude, frequency, potential energy distribution and propagation angles. With a decrease in Fr, the body generation mechanism become stronger and waves exhibit upstream propagation.
Bio:
Anikesh Pal is a Ph.D. candidate in the Department of Mechanical and Aerospace Engineering at University of California San Diego. He received his M.Tech from Indian Institute of Technology Kanpur, INDIA in Mechanical Engineering (Fluid and Thermal Sciences). His current research focuses on the dynamics of the flow past bluff bodies in stratified fluid. He is also interested in studying particle laden flows in stratified environment and their applications in geophysical flows.
----------------------------------
Large-eddy simulation study of combined configuration and stability effects in wind farms and development of geometry-based models for layout evaluation
Date and Time: Friday, May 20, 2016 - 04:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Niranjan S. Ghaisas, CTR Postdoctoral Fellow
The diurnal variation of temperature leads to significant deviations from neutral atmospheric stability conditions in the atmospheric boundary layer. Previous field observations have revealed ambiguities in the effect of atmospheric stability on wake losses and power production of wind farms. Wake losses are also known to be affected significantly by the configuration (layout and wind direction) of the wind farm. Large-eddy simulations (LES) of the Lillgrund wind farm (48 closely-spaced turbines) are conducted in order to study the combined effects of layout, wind direction and stability in large but finite wind farms. In keeping with previous theoretical estimates, wake losses are found to be larger under stable than under neutral conditions for the perfectly aligned configuration. In contrast, in three of the four configurations studied, wake losses actually decrease under stable stratification compared to under neutral conditions. Two competing effects, based on the rate of wake recovery and the lateral spread of the wakes are identified, which explain this unexpected influence of stability on the power production. This study reveals that atmospheric stability and configuration interact in a complicated manner to determine the total power produced by large, finite wind farms. The LES results are also used to develop simple statistical models that predict the power based only on geometric parameters derived from the wind farm layout. These geometry-based models are proposed as alternatives to the industry-standard PARK model, and can be used for quick evaluation and screening of wind farm layouts. Finally, current efforts towards high-order simulations of deformations of solids, and their coupling to fluids, in a fully Eulerian framework, will be briefly discussed.
Bio:
Niranjan S. Ghaisas received his Ph.D. from Purdue University in December 2013 and completed a postdoctoral research stint at the University of Delaware before joining CTR as a postdoctoral fellow in July 2015. His research interests include subgrid scale modeling for stratified flows and simulations for wind energy applications, and current work at CTR focuses on Eulerian multi-material simulations using high-order methods.
--------------------------------------------
Particle-laden Droplet Removal from Superhydrophobic Surfaces: A Computational Study
Date and Time: Friday, April 29, 2016 - 16:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Samaneh Farokhirad, Postdoctoral Research Associate and Adjunct Lecturer, CCNY
Interface-driven fluid dynamics is one of the recent interesting and challenging phenomena in fluid dynamics problems. Self-propulsion of droplets on superhydrophobic surfaces is one example of such phenomena, in which the fluid has a tendency to flow towards regions of higher surface energy, and subsequently a net flow results in the droplet motion. Rapid removal of self-propelled droplets from the surface is an essential factor in terms of expense and efficiency for many applications, including self-cleaning and enhanced heat and mass transfer to save energy and natural resources. Self-cleaning surfaces found in nature show great potential for application in many fields, ranging from industry to medicine. A potential mechanism for self-cleaning of natural surfaces in places with no precipitation is coalescence-induced self-propelled jumping of droplets, which was first reported in Phys. Rev. Lett. 103, 184501 (2009). The micro- and nanostructures of natural surfaces help the dew droplets to coalesce, become bigger, and finally jump several millimeters into the air to carry unwanted particles, such as microbial and virus particles off the surface. This process requires neither gravity nor wind and provides a fundamentally different self-cleaning mechanism than the conventional lotus effect. This talk focuses on the influences of the ambient environment and particle presence on the development of a rapid self-cleaning mechanism using a three-dimensional Lattice Boltzmann Method (LBM). As a diffuse interface method, LBM links microscopic phenomena with the continuum macroscopic equations, and is suitable for simulations of complex fluid flow problems, including single-phase and multiphase flows, particle-laden flows, turbulent flows, and free surface problems. The fact that the nonlinear flow physics is fully contained in the local collision process makes LBM method readily parallelizable.
Bio:
Samaneh Farokhirad graduated with a Bachelor of Science in Mechanical Engineering from Sharif University of Technology in 2006, and received her Master of Science in Mechanical Engineering from Iran University of Science and Technology in 2009. Then, she joined the Computational Multiphase Fluid Dynamics Group at the City College of New York, where she has collaborated with researchers from different fields including engineering, physics, and applied mathematics. After concluding her PhD in 2015, she has been working as a postdoctoral research associate and adjunct lecturer at CCNY. Samaneh Farokhirad’s main research interests are particle-laden multiphase flow and high performance computing.
-------------------------------------
Forced wavepacket models of turbulent jet noise
Date and Time: Friday, April 1, 2016 - 16:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Aaron Towne, CTR Postdoctoral Fellow
The high noise levels generated by the jet exhaust from commercial and military aircraft make mixing-noise reduction an important objective. Crucial to this effort will be the availability of robust, rapidly computable noise models that can be used to guide and optimize noise control strategies. Numerous studies have identified large-scale coherent structures in the form of wavepackets within the turbulent flow field as an important noise source. The typical approach to modeling wavepackets is to approximate them as linear modal solutions of the Navier-Stokes equations linearized about the long-time mean of the turbulent flow field. The near-field wavepackets obtained from these linear models show compelling agreement with those educed from experimental and simulation data for both subsonic and supersonic jets, but the associated far-field acoustic radiation is severely under-predicted in the subsonic case. This suggests that nonlinear effects are important in increasing the radiative efficiency of the wavepackets.
In this talk, I will summarize recent efforts to educe and characterize the nonlinear forcing experienced by wavepackets in subsonic jets. The main conclusion is that random turbulent fluctuations, rather than direct nonlinear interactions amongst wavepackets, are primarily responsible for producing acoustically efficient wavepackets. This suggests that the essential ingredients of sound generation in high Reynolds number jets are contained within the linearized Navier-Stokes operator, a conclusion that has important implications for jet noise modeling.
Bio:
Aaron Towne recently joined CTR as a postdoctoral fellow. He received his B.S. in Engineering Mechanics from the University of Wisconsin-Madison and his M.S. and Ph.D. in Mechanical engineering from the California Institute of Technology. His research interests include aeroacoustics, reduced-order modeling, and flow instability.
-----------------------------
Optimal Design by Morphing
Date and Time: Friday, March 11, 2016 - 04:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Professor Philip Marcus, University of California, Berkeley
We present a new method, which we call design-by-morphing, for the optimal design of the shape of an object. Traditional morphing methods, which require covering the surface of an object with a large number (typically millions) of triangular meshed points, cannot be used in searches for optimal designs because traditional morphing methods break down without human intervention. With our new methodology, the surfaces of one or more objects (or the sub-objects from which they are composed) are represented as truncated series of exponentially-convergent spectral basis functions multiplied by spectral coefficients. A morphed object (or sub-object) is obtained from a new set of spectral coefficients, which are a weighted average of the spectral coefficients of the original objects (or sub-objects) from which it is morphed. Optimized designs, say, for reducing aerodynamic drag, are created by choosing the weights such that a cost function of the new morphed shape is minimized. Re-purposing the applied mathematics that were developed for spectral methods in computational fluid dynamics, the boundaries of an object and the interfaces between sub-objects can be forced to satisfy constraints on their shapes, slopes, curvature, etc. With these constraints, sub-objects can be seamlessly attached to each other to create a complex object. Our design-by-morphing method can be automated and is computationally efficient, so it requires much less human input than traditional design methods and is therefore not only inexpensive but also free from human bias in finding optimal designs that are radical and non-intuitive. Examples are presented of optimal designs of trains, airplanes, and turbine draft tubes. The efficiencies of the designs are improved by more than 10%.
Bio:
Philip Marcus is a Professor of Fluid Dynamics, Mechanical Engineering, at the University of California, Berkeley where he heads the Computational Fluid Dynamics Laboratory. His research group is focused on the fluid dynamics of vortices, waves, turbulence, and hydrodynamic stability.
------------------------------------------
Tracking eddies in wall-bounded turbulence
Date and Time: Friday, March 4, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Adrian Lozano-Duran, CTR Postdoctoral Fellow
Eddies, understood as regions of the flow, which maintain spatial and temporal coherence, are widely used by the turbulence community as a conceptual model to organize and understand the flow. However, are they really there? Can they be identified and tracked in time? The present talk deals with the temporal evolution of vortices and eddies responsible for the momentum transfer in turbulent channels studied via time-resolved direct numerical simulation at high Reynolds numbers in a five hundred Terabytes database.
Eddies are identified as connected regions of the flow above a prescribed threshold, and tracked in time with a novel method specifically designed for the task. Once the evolutions are properly organized, they provide all the necessary information to test the coherence of the eddies and to characterize their lives from birth to death with a level of detail never achieved before. Finally, all the new information is compiled to build a structural model for the logarithmic layer of wall-bounded turbulence based on self-similar sweep-ejection pairs embedded within streamwise rolls. It is shown that these eddies act as the fundamental dynamical units of the flow and provide a great insight into the Turbulence physics.
Bio:
Dr. Adrian Lozano-Duran received his PhD from the Technical University of Madrid in 2015 at the Computational Fluid Mechanics Laboratory headed by Professor J. Jiménez. His main research has focused on Computational Fluid Mechanics and the fundamental physics of Turbulence. Currently, he is a post-doc at the Center for Turbulence Research at Stanford University working on Large Eddy Simulation and wall-modeling.
-----------------------------------------------
High Reynolds Number Smooth/Rough-wall Turbulent Boundary Layers
Date and Time: Friday, February 26, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Xiang I. A. Yang, Mechanical Engineering Department, Johns Hopkins University
We use tools including Large-eddy-simulations, wind tunnel experiments and the framework provided by the Townsend attached eddy hypothesis to study the flow physics in high Reynolds number turbulent boundary layers. We developed a hierarchical random additive process model (HRAP) for the cascading process in wall bounded flows. With this HRAP, a new family of two-point logarithmic scalings in the inertial region is discovered and confirmed using the experiment data from the Melbourne HRNBLWT. The scalings of single-point, two-point moment-generating-functions in high Reynolds number wall bounded flows are also investigated. The MGFs provide us with new insights into the near wall flow physics that are un-available in conventional moments. While the work mentioned above focuses on smooth-wall turbulent boundary layers, in this talk, we discuss the flow behaviors in the presence of ground roughness. LES are used for this purpose. By conducting LES of flow over various ground roughness, we find the mean flow behavior beneath the height of ground roughness follows a generic exponential profile. This exponential behavior, combined with the commonly accepted logarithmic behavior in the inertial layer, as well as a geometric sheltering model that accounts for the wake interactions among roughness elements, leads to a rough wall model that enables us to make rapid predictions on rough wall hydrodynamic properties solely based on roughness morphology. LES data are used to evaluate the model performance and we find reasonably good agreement between the LES measurements and the model predictions. Last, possibilities of using the recently gained knowledge of generalized log laws and the HRAP model to construct LES wall models are discussed.
Bio:
Xiang I. A. Yang is currently a Ph.D. candidate in Mechanical Engineering Department of the Johns Hopkins University. He received M.S. in Mechanical Engineering from the Johns Hopkins University. His graduate research topic is mainly on high Reynolds number turbulent boundary layers over smooth and rough surfaces. Some of his work also includes the development of numerical schemes.
--------------------------------------------
Modeling and understanding of supercritical injection
Date and Time: Friday, February 19, 2016 - 16:15
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Daniel Banuti, CTR Postdoctoral Fellow
Despite being ubiquitous in technical applications (gas turbines, diesel engines, rocket engines), supercritical injection is generally considered not well understood. Far from idealized gaseous or liquid fluid behavior, there is to this date no real understanding of the underlying physical processes. Nonetheless, accurate modeling and understanding are key factors for effective CFD modeling and optimization of the technical systems.
The present talk discusses a new thermodynamic view of supercritical state transitions akin to vaporization - pseudo-boiling. The talk introduces 'pseudo-boiling' and highlights some results of the approach relevant for injection and turbulent combustion, e.g. interpretation and explanation of a reference injection experiment which could not be explained with the state-of-the-art theory of a mechanical atomization process; the structure of a supercritical jet; simplified and more accurate real gas mixture modeling.
It seems that research so far has been hindered by the lack of a quantitative theory of supercritical fluids behavior. The pseudo-boiling approach provides a framework that may show where to look and what questions to ask when devising numerical or physical experiments.
Bio:
Dr. Daniel Banuti's focus is CFD model development and thermodynamics of high pressure injection. Before joining the Center for Turbulence Research at Stanford University, Daniel Banuti was a Research Scientist at the German Aerospace Center (DLR) in Göttingen, mainly working on numerical modeling of injection and combustion in rocket engines, other work ranged from reactive nozzle flow to hypersonics. Daniel Banuti received his PhD from Stuttgart University while working at DLR, and his MSc from RWTH Aachen University, Germany. He spent a year as Graduate Research Assistant at the University of Tennessee Space Institute.
-----------------------------------------
Development and Applications of PDE Solvers on Octree Adaptive Grids
Date and Time: Friday, January 29, 2016 - 04:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Frederic Gibou, Professor and MechE Graduate Program Director, University of California, Santa Barbara
It is well recognized that computational science is the third pillar of discovery along with theory and experiments. The challenges to modern scientific computing are (1) multiscale nature of most important physical phenomena, with a dynamic coupling between smaller and larger scales, (2) the need to impose non-trivial boundary conditions on irregular domain or on moving boundaries and (3) the need to perform large 3 dimensional simulations.
Professor Gibou’s research group’s strategy is to develop computational methods on Cartesian grids. The advantage of this approach is that they do not need to impose the mesh to conform to the boundary of irregular domains or a moving free boundary and thus avoids the difficulties associated with the meshing procedures of body-fitted-like approaches. In this talk, Professor Gibou will present computational methods they have developed on adaptive Octree grids, which enable to capture small length scales at the continuum level, without having to refine the entire domain.
Bio:
Professor Gibou is a faculty member in the Department of Mechanical Engineering, in the Department of Computer Science and in the Department of Mathematics at the University of California, Santa Barbara. He is also a core faculty member in the Computational Science and Engineering program. He received his PhD from the Applied Mathematics Department at UCLA, and did his post-doctoral research in the Departments of Mathematics and Computer Science at Stanford University. The research of Professor Gibou is at the interface between Applied Mathematics, Computer Science and Engineering Sciences. It is focused on the design of a novel paradigm for high resolution computational methods for large scale computations and their use for a variety of applications including Computational Materials Science, Computational Fluid Dynamics and Computational Image Analysis. Professor Gibou leads a multidisciplinary research group, named Computational Applied Science Laboratory (CASL).
---------------------------------------------
Multiscale challenges in direct numerical simulation of multiphase flows
Date and Time: Friday, January 15, 2016 - 04:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Bahman Aboulhasanzadeh, Postdoctoral Research Fellow, University of Michigan
In the past couple of decades, computational fluid dynamics of multiphase flows has evolved tremendously. With the fast growth in computational power, researchers who once had been bounded to study the problems in small domains or for very limited number of bubbles or droplets, now are able to do the simulations of more realistic problems or to look at additional physics. However, computational simulation of multiphase flows can become exponentially expensive when thin films appears in the physical domain, either as a results of interactions between interfaces or the existence of different physics which usually leads to diverse range of length and time scales in the systems. Resolving thin films using conventional methods like adaptive mesh refinement can adversely increase the computational cost and also make the parallel scaling on massive clusters hard to achieve. Fortunately, it is possible for many cases to develop analytical subscale models which bridge the gap between length and time scales in the problem. We developed a subscale model for mass transfer in bubbly flows and demonstrated its accuracy and efficiency. This model can be essential in simulations of bubble columns, one of the most important processing units in chemical and petrochemical industries. Additionally, we developed two other subscale models, the first is for the correction of viscous forces for simulation of colliding non-coalescing droplets and the second is a thermal model for simulations of a cavitation driven heat transfer problem, critical to naval engineering applications.
Bio:
Bahman Aboulhasanzadeh received his Ph.D. in Aerospace and Mechanical Engineering from the University of Notre Dame in 2014 and he has been a postdoctoral research fellow at the University of Michigan during 2015. His research interests include multiphase flow, high performance computing using CPUs and GPUs, grid generation, and turbulent flows.
---------------------------------------------
A LES study of interactions between wind turbine wakes and the atmospheric boundary layer
Date and Time: Friday, January 8, 2016 - 16:00
Location: CTR Conference Room 103
Event Sponsor: Parviz Moin, Director of Center for Turbulence Research
Speaker(s): Shengbai Xie, Graduate Research Assistant, University of Delaware
It is critical to understand how wind turbine wakes interact with the atmospheric boundary layer (ABL) in order to better estimate the wake loss in a wind farm. The ABL is a complex system affected by various forces, e.g., pressure gradient force (PGF), Coriolis force, buoyancy force due to thermal stability, and shear force imposed by the ground, etc. The wind shear and turbulence level in the ABL is strongly influenced by the thermal stability conditions, which in turn affects the properties of wind turbine wakes [1]. Meanwhile, the Coriolis force interacts with the wind shear and the PGF to induce changes of wind directions at different heights, known as wind veering. The wind veering is potentially able to alter some properties of wind turbine wakes, which has not been fully understood before. Here, a large-eddy-simulation based study will be presented to show how the wind turbine wakes evolve according to various thermal stability conditions with consideration of the Coriolis’ effect.
Second, a new approach of the vertically staggered wind farm is proposed, which involves both large-scale horizontal axis wind turbines (HAWT) and small-scale vertical axis wind turbines (VAWT). From our LES study, this vertically staggered wind farm is shown to have improved performance compared to the traditional wind farm, as a result of the enhanced wake recovery due to stronger ambient turbulence. A theoretical top-down model [2] is also developed to further analyze the potential of this new type of wind farm.
References
[1] Abkar, M., Porté-Agel, F., 2010. Influence of atmospheric stability on wind-turbine wakes: A large-eddy simulation study. Phys. Fluids 27, 035104.
[2] Calaf, M., Meneveau, C., Meyers, J., 2010. Large eddy simulation study of fully developed wind turbine array boundary layers. Phys. Fluids 22, 015110.
Bio:
Shengbai Xie received his PhD from University of Delaware in 2015. His PhD work was mainly focused on numerical simulations of wind turbine wakes and the atmospheric turbulence. He also has a master degree from Johns Hopkins University, where he used LES to study wind-wave-structure interactions.