# Keynotes

## William K. George – A Dummy’s Guide to Turbulence

Keynote Title: "A Dummy's Guide to Turbulence"

Presentation Schedule: soon

#### William K. George – A Dummy’s Guide to Turbulence

*Professor Emeritus (SRI)*

Imperial College of London

**London, UK**

#### A Dummy's Guide to Turbulence

Lecture 1: A Dummy’s Guide to Turbulence

This lecture will focus on things we think we know about turbulence that are probably true. It will begin with a discussion of what turbulence is, why it is so difficult (both to learn about and to study), and why we usually end up working with averaged equations. Then it will use the single point Reynolds-averaged equations to show how transition to turbulence occurs, why some simple linearization methods sometimes work and sometimes do not work, and finally how one builds a set of equations useful for engineering purposes. These single point RANS equations will pulled apart to show how simple physical ideas have led to the various levels of turbulence models from eddy viscosity to RANS to LES. The discussion should be particularly useful to anyone trying to decide which model to choose, what level of complexity is justified, or most importantly, why and when some will always fail.

**Lecture 2: Letting the Genie Out of the Bottle?**

This lecture will focus on things we thought we knew about turbulence that are probably false, or at least not as general as we might have been led to believe from books. It is well-known that in science that the more slowly a field advances and the more difficult the problem, the more likely it is that things creep into our knowledge base that are really not true. Original hypotheses that seemed to be good ideas when proposed become accepted based on the lack of contradictory evidence. As time progresses, they become impossible to overturn no matter how much evidence is assembled. Turbulence is the prototype for this kind of problem, so much so that aspects of it are more like a religion with its canons and creeds than ideas to be tested with experiment and reason.

Some of the examples addressed in this lecture will include the following: Are coherent structures really all that important? Is similarity theory really all that bad? Is the Kolmogorov/Batchelor view of the turbulence really correct? Specifically, is there really a universality of the small scales? Is the Kolmogorov microscale really the smallest dynamically significant scale? Is ‘local equilibrium’ really that good an idea? Is the dissipation really like u-cubed over L and independent of viscosity in the limit of zero viscosity (the zeroth law of turbulence)? The goal will be to liberate researchers entering the field to not only be open to new ideas, but to actively challenge old ones. Or in other words, when stumbling through the forest of facts and ideas, be sensitive and alert enough to recognize the diamond in the rough when you see it. Only if this happens can the field of turbulence (or any field) advance.

## Anurag Agarwal – University of Cambridge

Keynote Title: "Instability and transition"

Presentation Schedule: soon

#### Anurag Agarwal – University of Cambridge

**Department of Engineering, University of Cambridge**

United Kingdom

#### Instability and transition

In this talk we will explore various examples of hydrodynamic instability that lead to transition to interesting unsteady phenomena with biological, mechanical and aerospace applications. These include:

• Whistling of a steam kettle

• Delta wing rock and rustling of leaves in the wind

• Precession of vortex core in cyclone separators

• Whistling of rocket balloons

• Production of music from free reeds found in musical instruments

• Fluttering in heart valves

• Wheezing in lungs

Prerequisite for this talk is basic understanding of fluid dynamics and structural vibration. This includes classical aerofoil theory, boundary layers, Kelvin-Helmholtz instability, acoustics (sound waves, Helmholtz resonators, duct resonance) and vibration of thin beams and plates.

## Tim Colonius – California Institute of Technology

Keynote Title: "Turbulent jets, wavepackets, and jet noise"

Presentation Schedule: soon

#### Tim Colonius – California Institute of Technology

**California Institute of Technology**

#### Turbulent jets, wavepackets, and jet noise

After a brief review of the status and relevance of jet noise, we discuss wavepacket structures that are observed in the near pressure field of turbulent jets and their relation with the radiated acoustic field. These wavepackets have been observed for many years, but their importance to the far-field noise is still debated, particularly in subsonic jets. We discuss a theoretical framework for modeling the wave packets based on local and global modes of the equations of motion linearized about the mean jet flow field. These linear models generally provide good quantitative agreement with the spatial distribution of wave packet amplitude, wavelength and phase speed inferred from near-field microphone measurements and time-resolved particle image velocimetry. However, linear models generally under predict the associated acoustic radiation. Intermittency of the wavepacket activity and forcing of the wavepackets by uncorrelated turbulence are two hypothesized mechanisms by which sound radiation is amplified beyond what the linear models predict. We discuss recent attempts to use high-fidelity simulation data to probe the turbulence and acoustic field to characterize wavepacket intermittency and forcing, and to amend linear models to account for these features. Next, we discuss a novel parabolization of the Euler and Navier-Stokes equations that can be used to develop computationally fast, linear and nonlinear reduced-order models for the wavepacket structures and their sound radiation. The new approach remedies some flaws in the Parabolized Stability Equation (PSE) approach, particular with respect to accurately representing acoustic radiation. We close by surveying recent attempts to reduce jet noise through geometric changes to the nozzle geometry (e.g. chevrons) and through active control, and attempt to connect the observed noise reduction with alterations in the amplification and propagation characteristics of the wavepackets.

## James Gregory – Ohio State University

Keynote Title: "Fast-Response Pressure-Sensitive Paint for Measurement of Unsteady Surface Pressures"

Presentation Schedule: soon

#### James Gregory – Ohio State University

*Associate Professor
*

**The Ohio State University**

#### Fast-Response Pressure-Sensitive Paint for Measurement of Unsteady Surface Pressures

This lecture will detail the development and application of Fast Pressure-Sensitive Paint (PSP) for the measurement of unsteady surface pressures. PSP is an optical technique that has been used extensively over the past 25 years for the measurement of steady pressures on a model of interest. The technique offers the advantages of very high spatial resolution, relatively low cost, and easy application. These features make PSP ideal for investigating complex flow phenomena, particularly in locations that are difficult to instrument with conventional techniques. It is only recently, however, that the capabilities of Fast PSP have been developed to extend the frequency response routinely beyond 5 kHz, and in limited cases up to 1 MHz.

This lecture will begin with the fundamental principles of PSP, including a brief tutorial of common strategies for making PSP measurements and processing data. Beyond that introductory overview, this lecture will provide an assessment of the state of the art in fast-responding PSP. There are a number of challenges associated with expanding the utility of PSP for studying a wide range of unsteady flow phenomena. In particular, strategies for dramatic improvements in frequency bandwidth will be detailed. The effects of surface roughness, paint thickness, and luminophore selection will be elucidated through modeling and experimental results. Both the magnitude attenuation and phase delay will be considered (see Figure 1), and calibrations via various techniques will be presented. A second significant challenge to be addressed is the sensitivity limit of PSP, and its implications for measuring very small pressure changes. Third, the inherent temperature-sensitivity of PSP will be treated, including a discussion of strategies for overcoming this deficiency.

The implementation of these developments will be expressed through several example applications of Fast PSP, to include acoustic pressure measurements and unsteady surface pressure measurements on rotating helicopter blades. Finally, the outlook for PSP and its application to turbulence and acoustic studies will be highlighted.

## Siegfried Raasch – Hannover University

Keynote Title: "Fundamentals of Atmospheric Turbulence Modeling"

Presentation Schedule: soon

#### Siegfried Raasch – Hannover University

Department of Meteorology and Climatology (IMUK)

**Leibniz Universität Hannover**, Germany

#### Fundamentals of Atmospheric Turbulence Modeling

The lecture will give a basic introduction to atmospheric boundary layer turbulence. It will start with key properties of turbulent flows, the underlying equations and how they are modified in order to study turbulence. Typical examples for different turbulence regimes (stable, neutral, unstable/convective) will be presented. The main part of the lecture will be on fundamentals of turbulence modeling with a special focus on large eddy simulation (LES). Computational methods will be explained and current practical applications to fundamental as well as applied problems will begiven. The lecture will close with future perspectives on atmospheric turbulence modeling.

## Julio Soria – Monash University

Keynote Title: "Advanced Optical Image-based Flow Measurement Methods with Application to Turbulence Research"

Presentation Schedule: soon

#### Julio Soria – Monash University

Laboratory for Turbulence Research in Aerospace and Combustion

Department of Mechanical and Aerospace Engineering

**Monash University** (Clayton Campus)

Melbourne, VIC Australia

Department of Aeronautical Engineering

**King Abdulaziz University**

Jeddah, Saudi Arabia

#### Advanced Optical Image-based Flow Measurement Methods with Application to Turbulence Research

This lecture will present the underlying theory upon which Particle Image Velocimetry (PIV) is based in its most general form. The number of different approaches ranging from two-component two-dimensional (2C-2D) velocity field measurements to three-component three-dimensional (3C-3D) velocity field measurements will be presented and discussed. The discussion will include highlighting the advantaged and disadvantages of different approaches, such as 2C-2D high-speed PIV, 3C-2S stereo-PIV, 3C-3D tomographic and holographic PIV, as well as their respective experimental implementations.

Examples of PIV applications pertaining to wall-bounded turbulent shear flows including zero-pressure gradient turbulent boundary layer flow and adverse pressure gradient turbulent boundary layer flow will be presented, as well as turbulent free-shear flows including compressible gas flows such as under-expanded supersonic impinging jet flows will be presented and discussed.

## Jean Philippe Laval – Laboratoire de Mécanique de Lille

Keynote Title: "Coherent structures in near wall turbulence"

Presentation Schedule: soon

#### Jean Philippe Laval – Laboratoire de Mécanique de Lille

Laboratoire de Mécanique de Lille

#### Coherent structures in near wall turbulence

The recent developments of optical metrology such as PIV (Particle Image Velocimetry) or Tomo-PIV as well as the increasing capabilities of direct numerical simulations (DNS) allow us to have a deeper knowledge of coherent structures of wall-bounded turbulent flows at fairly large Reynolds numbers. The boundary layer wind tunnel of the Lille Mechanics Laboratory (LML) was used to study the organization of coherent structures in the inner layer of a flate plate turbulent boundary layer. A detail investigation of the near wall streaks and vortices was conducted and a model of self organization was proposed [1]. This organization is in agreement with most of previous studies. However, recent measurement at large Reynolds numbers have shown the emergence of a second peak of normal Reynolds stresses in the lower part of the log layer. This peak contribute to the production of turbulent kinetic energy and is known to have a footprint down to the buffer layer and wall shear stress [2]. New experimental results at very large Reynolds number were used to show that the Towsend-Perry [3] model needs to be adapted in order to take into account the development of this second peak which is the consequence of very long streamwise coherent structures [4].

One of the remaining challenge is to improve our knowledge of wall turbulence subjected to pressure gradient. The presence of even a moderate normal adverse pressure gradient is known to modify significantly the physics of wall bounded flows. The effect of APG is to generate a stronger peak of tubulent kenetic energy that moves away from the wall. The mechanism of this new physics is not well understood even if the consequence on the turbulent statistics have been shown in various configurations with a wide range of pressure gradients and Reynolds numbers. This new physics has dramatic consequences on the performance of statistical turbulence models which have been designed and tuned for zero pressure gradient wall turbulence. Understanding the impact of pressure gradient is a first requirement in order to propose new turbulence models with improved performances for real flow configurations.

In order to address this problem, a DNS of a turbulent channel flow with a lower curved wall have been performed at LML [5,6]. Low-speed streak structures were extracted from the turbulent flow field using methods known as skeletonization in image processing. Individual streaks in the wall-normal plane averaged in time and superimposed to the mean streamwise velocity profile are used as basic states for a linear stability analysis. Instability modes are computed at positions along the lower and upper wall and the instability onset is shown to coincide with the strong production peaks of turbulent kinetic energy near the maximum of pressure gradient on both the curved and the flat walls. The instability modes are spanwise-symmetric for the adverse pressure gradient streak base flows with wall- normal inflection points, when the total average of the detected streaks is considered. The size and shape of the counter-rotating streamwise vortices associated with the instability modes are shown to be reminiscent of the coherent vortices emerging from the streak skeletons in the direct numerical simulation. The distance of the streak’s centre from the wall is used as a criterion for the conditional averages and the corresponding streak base flows are characterised by more or less pronounced contours of inflection points in the averaging windows normal to the wall. It is shown that the strength of instability of the streak base flows can be inferred from a simplified 1D stability analysis, using local inflectional profiles at different spanwise locations. It remains to be assessed whether this type of simplified analysis is useful to adapt the turbulence models by anticipating a strong production of turbulent kinetic energy.

1. J. Lin, J.-P. Laval, J.-M. Foucaut, M. Stanislas, 2008, Quantitative Characterization of coherent structures in the buffer layer of near-wall turbulence : Part 1 : streaks, Exp. Fluids. 45, 999-1013.

2. R. Mathis, N. Hutchins, and I. Marusic, 2009 , Large-scale amplitude modulation of the small-scale structures in turbulent boundary layers, J. Fluid Mech. 628, 311.

3. A.E. Perry, S.M. Henbest, M.S. Chong, 1986, A theretical and experimental study of wall turbulence. J. Fluid. Mech. 165, 163-199.

4. J. C. Vassilicos, J.-P Laval, J.-M. Foucaut, M. Stanislas, 2015, The streamwise turbulence intensity in the intermediate layer of turbulent pipe flow. J. Fluid Mech. 774, 324-341.

5. M. Marquillie, U. Ehrenstein and J.-P Laval, 2011, Instability of streaks in wall turbulence with adverse pressure gradient, J. Fluid Mech. 681 205-240 .

6. J.-P. Laval, M. Marquillie, U. Ehrenstein, 2012, On the relation between kinetic energy production in adverse-pressure gradient wall turbulence and streak instability, J. Turbulence 13 (21) 1-19 .

## Juan Pablo de Lima Costa Salazar – Universidade Federal de Santa Catarina -UFSC

Keynote Title: "PANS - Partially Averaged Navier-Stokes"

Presentation Schedule: soon

#### Juan Pablo de Lima Costa Salazar – Universidade Federal de Santa Catarina -UFSC

#### PANS - Partially Averaged Navier-Stokes

In this talk the Partially Averaged Navier-Stokes (PANS) is presented, which is a method for simulation that allows the ajustment of the spatial filter resolution to any value. In this model the model filter width, that is the spatial extension of the average operator is controled through two parameters: the unresolved fraction of the turbulent kinetic energy and the dissipation rate. A PANS closure model will be presented that may be formally derived from the Reynolds averaged (RANS) model. The method allows a gradual variation between a RANS simulation and a DNS through the PANS parameters. That way the simulation level of details may be taylored to the given computational resorces available. An simulation example will be presented using a comercial code largely used in universities and industries.

## Marcello Augusto Faraco de Medeiros – Universidade de São Paulo – EESC

Keynote Title: "Turbulence and noise on a slat cove: experiments, simulations, analysis and control"

Presentation Schedule: soon

#### Marcello Augusto Faraco de Medeiros – Universidade de São Paulo – EESC

#### Turbulence and noise on a slat cove: experiments, simulations, analysis and control

The seminar describes the use of cross-correlation methods in trying to elucidate the appearance of narrow band peaks in the slat noise. It is an example of a turbulent flow with large coherent structures that could be associated with flow instabilities. Slat noise is an important component of aircraft noise, in particular at approach and landing. The work involved experiments and simulations. In experiments, cross correlation methods of the beamforming type were used to measured and locate the source of slat noise. The simulations produced results that agreed remarkably well with the experiments. The simulation results were analyzed with a cross correlation method of the POD type. It was possible to extract the structures correlated to the sound. They were Kelvin-Hemholtz vortices. Further analysis led to the establishment of a feed back loop responsible for the frequency selection of the narrow band peaks. The seminar also presents results of the influence of slat settings on the slat noise and a method of passive control of slat noise.

## Guenther Carlos Krieger Filho – Universidade de São Paulo -USP – Escola Politécnica

Keynote Title: "Turbulent combustion"

Presentation Schedule: soon

#### Guenther Carlos Krieger Filho – Universidade de São Paulo -USP – Escola Politécnica

#### Turbulent combustion

The role of the turbulent combustion processes in the energy supply scenario will be highlighted. Some examples of premixed and difusion flames will be presented. The basic equations for the turbulent reactive flows shall be formulated, with special attention given to the problematic of the averaged chemical source term in the scalar transport equations. Some laminar combustion models like equilibrium and flamelet will be presented as basis for the turbulent combustion models. Results of a methane difusion flame using RANS and LES turbulence models shall be discussed.

## Gervázio Annes Degrazia – Universidade Federal de Santa Maria – UFSM

Keynote Title: "Simulation of the Convective Boundary layer"

Presentation Schedule: soon

#### Gervázio Annes Degrazia – Universidade Federal de Santa Maria – UFSM

#### Simulation of the Convective Boundary layer

Using the Wiener-Khinchin theorem the Taylor statistic diffusion model will be written in a spectral form and the different parameters that quantify the scalar and vector turbulent quantities dispersion will be derived. The relation between Eulerian and Lagrangian statistic parameters will be discussed and the formulation, in terms of turbulent parametes, will be described in a Lagrangeam perspective. Using the formulas for the spectrum and turbulent structure functions, in the inertial subinterval, the Eulerian and Lagrangian constantes will be related. Examples of turbulent parameters, expressed in terms of convective boundary layer similarity and the one dominated by the wind shear, will be presented and discussed. Finally the horizontal wind phenomenon will be described and the its characteristic scales will be presented.

## André Valdetaro Gomes Cavalieri – Instituto Tecnológico de Aeronáutica – ITA

Keynote Title: "Experimental diagnostic for turbulence studies"

Presentation Schedule: soon

#### André Valdetaro Gomes Cavalieri – Instituto Tecnológico de Aeronáutica – ITA

#### Experimental diagnostic for turbulence studies

## Leandro Franco de Souza – Universidade de São Paulo – ICMC/USP

Keynote Title: "Stability of Non-Newtonian Fluid Flows"

Presentation Schedule: soon

#### Leandro Franco de Souza – Universidade de São Paulo – ICMC/USP

#### Stability of Non-Newtonian Fluid Flows

Fluid flow may be either in a laminar ou turbulent regime. There are many investigations on the transition process between laminar and turbulent regimes, which may take place in different routes. One of this routes is through the amplification of Tollmien-Shclichting waves, bouth linear and non linear interaction. Assuming a parallel base flow and that the disturbances take on a normal mode solution one may derive the Orr-Sommerfeld equations, which shows the wave frequencies that correspond to stable, neutral and unstable solutions. This type of analysis is the classic Linear Stability Analysis. Another method to verify how a given disturbance propagates in a given base flow is the direct numerical simulation.

Fluids may be classified in different ways, depending on its response when subject to a given force. When the stress in the fluid is proportional to the strain the fluid is classified as newtonian. There are many fluids that behave in a non-newtonian way, when the stress is not linearly related to the strain. One of the non-newtonian fluids is the viscoelastic fluid, which may be model in different ways, such as the Oldroyd-B model. This model may be understood as an extension of the Maxwell model and is equivalent to the mechanical mass-spring model. The name Oldroyd-B derives from James G. Oldroyd, who first proposed the model.

The present text presents the stability characteristics of a Oldroyd-B viscoelastic fluid in a flow between parallel plates to disturbances that propagate as Tollmien-Schlichting waves. The analysis is performed both through the linear stability theory using a modified version of the Orr-Sommerfeld equations and through direct numerical simulation. The results show how the Oldroyd-B model parameters change the behaviour of the propagating waves and their stability characteristics.

## Elmer Mateus Gennaro – Universidade Estadual Paulista Júlio de Mesquita Filho -UNESP

Keynote Title: "Advanced tools for the analysis of global instability of complex flows"

Presentation Schedule: soon

#### Elmer Mateus Gennaro – Universidade Estadual Paulista Júlio de Mesquita Filho -UNESP

#### Advanced tools for the analysis of global instability of complex flows

Laminar-turbulent transition is usually associated with instabilities of a laminar flow of reference. Thus, small disturbances grow in amplitude while extracting energy from the mean flow and give rise to finite-amplitude structures that trigger, through non-linear interactions, the transition process. Similarly, coherent structures often observed in turbulent flows have their origin in instabilities of the mean turbulent flow. The study of the flow instabilities provides, in addition to a deeper insight in the physical mechanisms involved, a base for the devise of theoretical and reduced-order models for the prediction of the effects associated with the flow structures generated, as well as a guide for the development of flow control strategies.

The study of hydrodynamic instability is based on the behavior after the introduction of very small amplitude disturbances, enabling the use of the linearized Navier-Stokes equations. The homogeneity of the linearized equations with respect to time allows for the assumption of a exponential behavior for the disturbances, resulting into an eigenvalue problem for modal instabilities. However, owing to the non-normal behavior of the linearized Navier-Stokes equations, transient energy growth can take place as a consequence of the superposition of multiple stable linear modes, resulting into energy amplification of several orders of magnitude. The adequate approach in this case addresses the determination of the initial condition for which, after a given time interval, a maximum transient amplification is obtained. Despite the apparent complexity of this approach, results can be calculated easily once the set of linear eigenmodes have been computed for the same base flow.

In essentially one-directional flows (Poiseuille, boundary layer, mixing layer), disturbances are non-homogeneous in only the cross-flow direction and the instability mechanisms are described via the solution of the Orr-Sommerfeld equation. For more complex flows, the parallel-flow hypothesis that assumes that the base flow does not depend on more than one direction is no longer valid, and a new theory is required in order to introduce the true nature of the flow into the analysis. Classic methodologies for the hydrodynamic instability analysis were conditioned by the limited calculation power at the times of their introduction, and were restricted to local analysis of parallel flows. The advances in the state of art of computers and existing algorithms allow us to extend the approach to global instability problems, considering two- and three-dimensional base flows.

## William Roberto Wolf – Universidade Estadual de Campinas -UNICAMP

Keynote Title: "Introduction to aeroacoustics including applications in turbulent flows"

Presentation Schedule: soon

#### William Roberto Wolf – Universidade Estadual de Campinas -UNICAMP

#### Introduction to aeroacoustics including applications in turbulent flows

This lecture will provide an introduction to the field of aeroacoustics including applications in turbulent flows and airframe noise. Some fundamental tools will be presented including the acoustic analogies of Lighthill, Curle and Ffowcs Williams and Hawkings. A particular focus will be given to the source terms appearing in the analogies and their physical mechanisms. This lecture will also discuss about the application of non-dispersive, non-dissipative, numerical methods for aeroacoustics including the computation of noise sources and their subsequent noise radiation to the far-field. Results obtained by hybrid approaches consisting of high-fidelity CFD simulations and boundary integral equations will be shown. We will close presenting results of airframe noise predictions from reduced-order models.

## Henrique Fanini Leite – Instituto Tecnológico de Aeronáutica

Keynote Title: "Introduction to Pressure Sensitive Paint (PSP) and Temperature Sensitive Paint (TSP)"

Presentation Schedule: soon

#### Henrique Fanini Leite – Instituto Tecnológico de Aeronáutica

**Instituto Tecnológico de Aeronáutica**

#### Introduction to Pressure Sensitive Paint (PSP) and Temperature Sensitive Paint (TSP)

This presentation aims to give an introduction toflow diagnostics based on pressure sensitive paint (PSP) and temperature sensitive paint (TSP). It will cover the underlying physics in which the techniques are based, basic hardware setup, radiometric and lifetime measurement approaches, static and dynamic calibration procedures, the limitations of conventional PSPs and the most common types of fast PSP. In addition, examples of the usage of such technologies in the study of transition and turbulence will be presented, as well as potential developments and the current fields of development.