Assaad Masri
University of Sydney

3.30pm Friday 10 November 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Nicole Jones
University of Western Australia

3.30pm Friday 22 September 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Monte Carlo methods for oscillatory nanoscale gas flows

Daniel Rowan Ladiges
University of Melbourne

3.30pm Friday 1 September 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)
Gas flows generated by resonating nanoscale devices inherently occur in the non-continuum, low Mach number regime. As the continuum assumption underpinning the Navier-Stokes equations is violated, these flows must be modelled using an approach that accounts for the non-continuum nature of the gas, i.e. the Boltzmann equation. Solution of the Boltzmann equation is difficult using standard numerical approaches - this has motivated the development of a range of Monte Carlo methods for simulating gas flows, the most well known of which is direct simulation Monte Carlo (DSMC)

We begin this talk with an introduction to the Boltzmann equation applied to nanoscale flows, followed by a discussion of methods which solve this equation using a Monte Carlo approach. We then present a frequency-domain Monte Carlo method which may be applied to oscillatory low Mach number gas flows. This circumvents the requirement of current Monte Carlo methods to operate in the time-domain, providing direct access to amplitude and phase information using a pseudo-steady algorithm. This frequency-domain method is shown to provide a significant improvement in computational speed compared to existing time-domain Monte Carlo methods.

Daniel received his PhD in applied mathematics from the University of Melbourne in 2016, where he is currently a postdoctoral researcher. His areas of research include analytic and numerical methods for solving the Boltzmann equation, particularly in the context of nanoscale gas flows.

Coherent structures and particles in turbulent boundary layers

Ellen Longmire
University of Minnesota, USA

3.30pm Friday 18 August 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

This talk will focus on our experience employing planar and volumetric velocimetry techniques to investigate the nature of turbulent boundary layers and the motions of finite particles within them. Fluid velocity fields are analyzed to quantify and understand the organization of coherent eddies as well as their relation to regions of long streamwise coherence. In the logarithmic region, extended packets of eddies can be tracked over significant distances. Our efforts to manipulate the packet organization and to understand packet recovery from perturbations will be discussed. In addition, our first results characterizing translation and rotation of spherical particles within the boundary layer will be presented.

Ellen Longmire received an A.B. in physics (1982) from Princeton University and M.S. (1985) and Ph.D. (1991) degrees in mechanical engineering from Stanford University. Prior to receiving her Ph.D., she worked as an engineer at Hauni-Werke Koerber & Co in Germany and at Honeywell and SAIC in the U.S. Since 1990, she has taught and directed research in the Department of Aerospace Engineering and Mechanics at the University of Minnesota where she holds the rank of Professor. She uses experimentation and analysis to answer fundamental questions in fluid dynamics that affect industrial, environmental, and biological applications. She is a Fellow of the American Physical Society and received the UM Distinguished Women Scholars Award (2007), the McKnight Land-Grant Professorship (1994), and the NSF National Young Investigator Award (1994). She serves as Editor-in-Chief of Experiments in Fluids. She previously served as Chair of the American Physical Society Division of Fluid Dynamics (2016), member of the U.S. National Committee on Theoretical and Applied Mechanics (2011-2015) and Associate Editor of Physics of Fluids (2008-2013).

Completion: Sub-miniature hot-wire anemometry for high Reynolds number turbulent flows

Milad Samie
University of Melbourne

4pm Friday 11 August 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Despite several decades of research in wall-bounded turbulence there is still controversy over the behaviour of streamwise turbulence intensities near the wall, especially at high Reynolds numbers. Much of it stems from the uncertainty in measurement due to finite spatial resolution. Conventional hot-wire anemometry is limited for high Reynolds number measurements due to limited spatial and temporal resolution issues that cause attenuation in the streamwise turbulence intensity profile near the wall. To address this issue we use the NSTAP (nano-scale thermal anemometry probe) developed at Princeton University to conduct velocity measurements in the high Reynolds number boundary layer facility at the University of Melbourne. NSTAP is almost one order of magnitude shorter than conventional hot-wires. This enables us to acquire fully-resolved velocity measurements of turbulent boundary layers up to a friction Reynolds number of 20,000.

Wall-resolved large-eddy simulation of flow about smooth-walled and span-wise grooved cylinders

Dale Pullin
Graduate Aerospace Laboratories California Institute of Technology, USA

3.30pm Friday 4 August 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Wall-resolved large-eddy simulation (LES) of flow over both smooth- and grooved-walled cylinders will be discussed. The stretched-vortex sub-grid scale model is embedded in a general fourth-order finite-difference code discretization on a curvilinear, body-fitting mesh. The grooved cylinder consists of 32 equal, sinusoidal, spanwise grooves around the cylinder circumference, each of equal height ε. The LES is first used to investigate the effect of groove height in 0 ≤ ε/D ≤ 1/32 at Re_D = 3.9 x 10³. A second set of LES uses fixed ε/D = 1/32 with 3.9 x 10³ ≤ Re_D ≤ 6 x 10⁴, the latter value reaching the transcritical range. The presence of grooves substantially reduces the Re_D range of the drag crisis. Both smooth-walled and grooved-wall cylinder flows exhibit mean-flow secondary separation bubbles embedded within the large-scale, separated flow region at subcritical Re_D. These transform to prior separation bubbles as Re_D is increased through and beyond the respective drag crises. For the smooth-wall flow this is associated with local transition to turbulence and a strong drag crisis. For the grooved-cylinder case the flow remains laminar but unsteady through its drag crisis and into the early transcritical flow range.

This research is performed in collaboration with W. Cheng and R. Samtaney (King Abdullah University of Science and Technology).

Dale Pullin is the Robert H. Goddard Professor at the Graduate Aerospace Laboratories California Institute of Technology. His research interests include computational and theoretical fluid mechanics, vortex dynamics, compressible flow and shock dynamics, turbulence, and large-eddy simulation of turbulent flows.

Confirmation: Development of algebraic Reynolds stress models for low pressure turbines using machine learning techniques

Harshal Akolekar
University of Melbourne

2pm Wednesday 2 August 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Enhancing the efficiency of the low pressure turbine (LPT) of an aircraft engine leads to the reduction in specific fuel consumption and carbon emissions. This enhancement can be achieved by accurately capturing two important phenomena in the designing and modelling process, namely: separation-induced transition and turbulent wake mixing. Direct numerical simulations (DNS) offer the most accurate solution to this problem but are still computationally expensive. On the other hand, some RANS-based turbulence and transition models, six of which are used in this work to analyse the T106A LPT profile, estimate the separation to some degree of accuracy, but predict inaccurate wake-loss profiles. This is partly due to the use of the Boussinesq approximation to derive stress-strain relationships. An approach is therefore sought to improve the Boussinesq relationship by the development of algebraic Reynolds stress models. Instead of using a conventional analytical approach, this work uses an evolutionary Gene Expression Programming algorithm (a machine learning approach) to develop algebraic models based on reference DNS data sets. Preliminary analysis has shown that the developed algebraic Reynolds stress models improve the turbulent wake-mixing and consequently wake-loss profiles.

Confirmation: Turbulent flow over surfaces with spanwise heterogeneity

Dea Wangsawijaya
University of Melbourne

3.30pm Friday 28 July 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Examples of wall-bounded turbulent flows over surfaces with heterogeneous roughness are abundant in both engineering applications and nature. A specific case of heterogeneity is a surface with roughness variation in the spanwise direction. Such surface generates secondary flow in the form of roll modes along the streamwise direction. Studies regarding the behaviour of wall-turbulent flows over spanwise heterogeneous roughness found contradictory results in terms of the sign of the roll modes and the mean streamwise velocity component U. The latter, as shown by a recent study by Chung, Monty & Hutchins (2017), might be affected by the ratio of spanwise roughness wavelength to δ. In this initial study, hot-wire measurements were performed over surfaces with spanwise heterogeneous roughness with a certain half-wavelength Λ, yielding a range of Λ/δ from 0.32 to 6.50. It is found that as Λ/δ becomes very large, secondary flows are restricted to the interface between high and low shear stress region, leaving the high and low stress region away from the interface to behave as if the surface becomes homogeneous based on the local stress. Similarly, measurement results of the very small Λ/δ test case show that secondary flows are confined within the near-wall region. Away from the wall, flow over the small Λ/δ case becomes spanwise homogeneous. As Λ/δ ≈ 1, U above the high and low shear stress region switches sign: U is higher above high shear stress region and lower above low stress region. Further analysis of 1-D energy spectra profile shows the occurrence of a second peak in the wake region, which might be related to an instability of secondary flows (Kevin et al. 2017).

Completion: Boundary layer and bulk behaviour in vertical natural convection

Chong Shen Ng
University of Melbourne

3.30pm Friday 21 July 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Thermal convection is ubiquitous in nature and many engineering applications, for example, from the scale of convection in stars and in Earth's atmosphere, down to the scale of convection-driven heating or cooling of the built environment and in the computer that you are using to read this text. In this research, we consider the buoyancy driven flow that is sustained between two vertical walls, where the left wall is heated and the right wall cooled. We refer to this setup as vertical natural convection. Results from direct numerical simulations for Rayleigh numbers 10⁵–10⁹ and Prandtl number 0.709 are found to support a generalised applicability of the Grossmann–Lohse theory for thermal convection, originally developed for Rayleigh–Bénard convection. Similar to Rayleigh-Bénard convection, a pure power-law relationship between the Nusselt, Rayleigh and Prandtl numbers is not the best description for vertical natural convection and existing empirical relationships should be recalibrated to better reflect the underlying physics.

In this talk, we show that for the present Rayleigh number range, the mean laminar-like boundary layers coexist with local near-wall patches that exhibit turbulent behaviour. Thus, the Nusselt number versus Rayleigh number scaling relation is contaminated by dynamics of the two regions. When the walls are removed and boundary layers omitted, the new setup mimics turbulent bulk-dominated thermal convection and we show that the Nusselt number versus Rayleigh number scaling relation instead follows the 1/2 power-law asymptotic scaling relation, in agreement with the theoretical predictions of Kraichnan (Phys. Fluids, 1962, vol. 5, pp. 1374–1389) and Grossmann and Lohse (J. Fluid Mech., 2000, vol. 407, pp. 27–56). An extension of this result to the wall-bounded setup is discussed.

Completion: Characterisation of a microfluidic hydrotrap to study the effect of straining flow on waterborne microorganisms

Farzan Akbaridoust
University of Melbourne

3.30pm Friday 30 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

We present a systematic study on the effect of straining flow on the harmful filamentous cyanobacterium Anabaena circinalis using a cross-slot type microfluidic device equipped with an advanced image-based real-time control system and simultaneously quantified by micron-resolution particle image velocimetry (micro-PIV). This enables us to monitor potential morphological damage to, and simultaneously compute the forces on, filaments in real-time while the filaments of the cyanobacterium are exposed to high strain rates.

Interaction of synthetic jets with turbulent boundary layers

Tim Berk
University of Southampton

3.30pm Friday 23 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Synthetic jets are zero-net-mass-flux actuators that can be used in a range of flow control applications. For several pulsed/synthetic jet in cross-flow applications, two physical phenomena are important: 1) The variation of the jet trajectory in the mean flow with jet and boundary layer parameters, which will provide an indication of the penetration depth of the pulsed/synthetic jet in to a boundary layer and 2) the losses associated with this penetration and the various physical mechanisms that are responsible for it. In this talk, I will present results that discuss both the variation of trajectories with different jet and boundary layer parameters and the mechanisms that are responsible for the losses.

The first part will focus on the mechanisms that are responsible for momentum losses created by the interaction of a synthetic jet with a turbulent boundary layer. Previous studies have ascribed the momentum deficit caused by the jet to a combination of viscous blockage of the crossflow and the upwash of low-momentum fluid. However, these contributions have not been quantified systematically. In order to resolve these contributions, detailed Particle Image Velocimetry measurements were carried out to examine the interaction between a synthetic jet and a low-Reynolds-number turbulent boundary layer at the University of Southampton. Based on an analysis of the vortical structures identified in the PIV data, the momentum deficit is attributed to the streamwise velocity induced by these vortical structures. Using Biot-Savart calculations, this induced velocity can be determined from the vorticity field obtained from PIV. The resulting momentum deficit is shown to account for 90% of the total momentum deficit, indicating that the advection of vortices is the main source of momentum deficit in these flows.

In the second part of the talk I will present results that compare the mean flow trajectories of the jet at low Reynolds number to the data obtained in the High Reynolds Number Boundary Layer Wind Tunnel at Melbourne. Trajectories of the synthetic jet in the turbulent boundary layer are measured for a range of actuation parameters in both low and high Reynolds numbers. The important parameters influencing the trajectory are determined from these comparisons. The Reynolds number of the boundary layer is shown to only have a small effect on the trajectory. In fact, the critical parameters are found to be the Strouhal number of the jet based on jet dimensions as well as the velocity ratio of the jet (defined as a ratio between mean jet blowing velocity and the freestream velocity). An expression for the trajectory of the synthetic (or pulsed) jet is derived from the data, which (in the limit) is consistent with known expressions for the trajectory of a steady jet in a cross-flow.

Tim received his BSc in Advanced Technology (2011), MSc in Mechanical Engineering (2014) and MSc in Sustainable Energy Technology (2014) from the University of Twente in the Netherlands. He is currently a PhD student at the University of Southampton under supervision of Bharath Ganapathisubramani, hoping to finish his PhD at the end of this year. For the past two months Tim has visited the University of Melbourne, performing experiments in the High Reynolds Number Boundary Layer Wind Tunnel.

Turbulent mixing and entrainment in density stratified riverine flows

Michael Kirkpatrick
University of Sydney

3.30pm Friday 16 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Maintaining the health of our river systems requires management of a limited resource, and involves balancing the water needs of agriculture, fisheries, urban water supplies, power generators, industry, tourism, recreational users and the environment. As a result, many rivers are now highly regulated and have a large proportion of their water withdrawn for human uses, leading to flow rates significantly below their natural levels. Lowering of flow rates can lead to serious water quality issues due to the combined effects of longer residence times, decreased turbulent mixing and increased levels of stabilizing density stratification due to solar heating and/or salinity. The resulting stagnant conditions are characterized by low levels of dissolved oxygen and the accumulation of contaminants and nutrients, which can lead to long-term, systemic damage to riverine ecosystems. Thermal stratification combined with low turbulence, and high radiation and nutrient loads are also conducive to the proliferation and accumulation of toxic cyanobacteria (blue-green algae) and other micro-organisms. Persistence of these conditions for more than a few days has been identified as a key risk factor for cyanobacterial outbreaks.

In this talk I will give an overview of our research on turbulent mixing and entrainment in density stratified riverine flows. The first half of the talk will focus on experimental and numerical studies of entrainment and mixing processes in saline pools that form commonly in river bed depressions. The second half of the talk will focus on recent Direct Numerical Simulation studies of thermal stratification in turbulent open channel flow due to radiative solar heating. I will present our recently published findings (J. Fluid Mech., 2015) regarding the equilibrium states reached by turbulent open channel flow with solar heating. I will then discuss our current work in which we are investigating the primary physical processes that break down these stratified equilibrium states, namely nocturnal surface cooling, wind shear and river bends.

The aim of our research is to develop scaling relations that can be used in combination with other data, such as weather forecasts, to predict the likely occurrence of persistent stratification events. These improved modelling capabilities will enable river managers to initiate short time-scale tactical responses such as an environmental flow release from upstream reservoirs in response to drought or heat-wave conditions in order to maintain water quality and river health. It will also enable them to better optimize long-term water management strategies such as the distribution of water allocations amongst stakeholders, taking into account predicted changes in land-use patterns and climate.

Michael Kirkpatrick graduated with a degree in Mechanical Engineering from the University of Sydney in 1996. He then worked with the international building engineering consultancy Ove Arup & Partners in Sydney and their London office. Michael returned to Australia in 1998 to study for a PhD at the University of Sydney. On completion of his PhD in 2002, he was awarded a two year Postdoctoral Fellowship with the Center for Turbulence Research at Stanford University and NASA Ames. In 2004 he accepted a lecturing position in the School of Engineering at the University of Tasmania. In 2006 Michael returned to the University of Sydney where he is currently an Associate Professor in Fluids, Energy and Environment in the School of Aerospace, Mechanical and Mechatronic Engineering.

Generation of a classical vortex ring and its axial interaction with a cylinder

Debopam Das
Department of Aerospace Engineering, Indian Institute of Technology Kanpur

3.30pm Wednesday 14 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

In this talk I will discuss a novel method for generating vortex rings that circumvents some of the drawbacks associated with existing methods. The predominant effects that occur in previously used methods are due to the presence of some of the other vortices such as, stopping vortex, piston vortex, image vortex and orifice lip generated vortices in the early stage of development. These disturbances influence the geometric, kinematic and dynamic characteristics of a vortex ring and lead to mismatches with classical theoretical predictions. It is shown in the present study that the disturbance free vortex rings produced follow the classical theory. Flow visualization and PIV experiments are carried out in the Reynolds number (defined as ratio of circulation Γ and kinematic viscosity ν) range, 2270 < Re_Γ < 6790, to find the translational velocity, total and core circulation, core diameter, ring diameter and bubble diameter. In reference to the earlier studies, significant differences are noted in the variations of the vortex ring diameter and core diameter. A model for core diameter during formation stage is proposed. The translational velocity variation with time shows that the second-order accurate formula derived using Hamilton's equation by Fraenkel (J. Fluid Mech., 1972, vol. 51, pp. 119–135) predicts it best.

Subsequently the axial interaction of this classical vortex ring with a thin circular cylinder has been studied. It is observed that due to the presence of the cylinder, there is an increase in the velocity of the vortex ring. Also, noticeable changes in the characteristic properties of vortex ring such as core circulation, core diameter and ring diameter have been observed. To justify these experimental observations quantitatively, an analytical study of the interaction under inviscid assumption is performed. The inviscid analysis does predict the increase in velocity during the interaction, but fails to predict the values observed in experiments. However, when the theory is used to correct the velocity change through incorporation of the effects of axisymmetric induced boundary layer region over the cylinder, modelled as an annular vortex sheet of varying strength, the changes in the translational velocities of vortex rings match closely with the experimental values.

Dr. Debopam Das is a professor of the Department of Aerospace Engineering, IIT Kanpur, India. He obtained PhD degree from Indian Institute of Science Bangalore. He has worked as Senior Research Scientist, Temasek Laboratory at National University of Singapore, Postdoctoral Research fellow in Florida State University and as an Assistant Professor, in the Department of Mechanical Engineering, IIT, Guwahati, India. His research interest is in flapping flight, incompressible and compressible vortex rings, instabilityo and transition of unsteady internal flows and instability of buoyant and non-buoyant free shear flows. He received Excellence in Aerospace Education award for 2015 by Aeronautical Society of India.

Flow generated by surface waves

Horst Punzmann
Research School of Physics and Engineering, Australian National University

3.30pm Friday 9 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Trajectories of fluid parcels on the surface have been described analytically for progressing irrotational waves, where particles move in the direction of wave propagation. Waves in the laboratory and in nature are more complex due to the development of instabilities that render ideal planar 2D propagating waves into complex 3D waves. In this talk I will present experiments in the short wavelength gravity-capillary range that demonstrate different surface flow phenomena. Propagating waves, driven by a vertically oscillating plunger, can lead to a flow reversal when wave instabilities render planer wave fronts into wave packets at high amplitude. At small-amplitude, stable flow pattern can be achieved in stationary (rotating) two-dimensional wave fields that are produced by phase-synchronised wave paddles. The role of surface vorticity generation by waves will be discussed.

Dr Horst Punzmann is research engineer at the Research School of Physics and Engineering at the Australian National University (ANU), Canberra. He obtained his undergraduate degree in Electrical Engineering at the University of Applied Science in Regensburg, Germany and worked in analog integrated circuit design in the automotive industry for 5 years. After this, he did his PhD in high-temperature plasma physics at the ANU and studied plasma turbulence and particle transport while being facilities manager of the Australian Plasma Fusion Research Facility for 10 years. He switched to hydrodynamic turbulence a while ago and his current research interest in the Physics of Fluids group is related to the interaction of surface waves and flows.

Confirmation: Air-sea interaction: laboratory study on the air-side turbulence of wind-wave

Tunggul Bhirawa
University of Melbourne

4pm Friday 2 June 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Wind is the main driving force in ocean wave generation, however the process of energy transfer from wind to wave is poorly understood. In order to explain the process comprehensively, we need to study how the atmosphere (air) and sea interacts. Air-sea interaction is relevant to understanding the coupled atmosphere-ocean system. The interactions involve small- and large-scale transfers of momentum, heat and mass, which affect our weather and climate. The knowledge of air-sea interaction is essential for industries related to the ocean system (ports and harbors, shipping, oil and gas offshore platforms) and for the ocean research community as well.

The common problem in studying this subject is to measure important quantities (sea drag, roughness length) close to the ocean surface which brings engineering and logistical challenges. In extreme conditions, such as in tropical cyclones, the uncertainty is higher since predictions are often extrapolated from lower wind speeds. This is where laboratory study is needed to make measurements possible.

PIV measurements were conducted to investigate the air-side turbulence above wind generated waves at the Extreme Air-Sea Interaction (EASI) facility in the Michell Hydrodynamics Laboratory. The objective is to study the behaviour of sea drag coefficient and roughness length under different wind speed and sea state conditions, which have been the most sought-after measurements in the study of air-sea interaction (Donelan et al. 1993).

Dynamic behaviours of composite panel subjected to underwater blast: fluid-structure interaction (FSI) experiment and numerical simulation

Jonathan Tran
Department of Infrastructure Engineering, University of Melbourne

4pm Friday 26 May 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Designing lightweight high-performance materials that can sustain underwater impulsive loadings is of great interest for marine applications. In this study, underwater experiments are conducted and validated with a novel finite element fluid-structure interaction to understand the deformation and failure mechanisms of both monolithic and E-glass/Vinylester/H250 PVC sandwich composite panels. The experimental setup simulates fluid-structure interactions (FSI) encountered in various applications of interest. To generate impulsive loading similar to blast, a specially designed flyer plate impact experiment was designed and implemented. The design is based on scaling analysis to achieve a laboratory scale apparatus that can capture essential features in the deformation and failure of large scale naval structures. In the FSI setup, a water chamber made of a steel tube is incorporated into a gas gun apparatus. A scaled structure is fixed at one end of the steel tube and a water piston seals the other end. A flyer plate impacts the water piston and produces an exponentially decaying pressure history in lieu of explosive detonation. The pressure induced by the flyer plate propagates and imposes an impulse to the structure (panel specimen), which response elicits bubble formation and water cavitation. Failure modes, damage mechanisms and their distributions are identified and quantified for composite monolithic and sandwich panels subjected to typical blast loadings. The temporal evolutions of panel deflection and centre deflection histories were obtained from shadow Moiré fringes acquired in real time by means of high speed photography.

Dr Jonathan (Phuong) Tran received his PhD degree in Theoretical and Applied Mechanics at the University of Illinois, Urbana Champaign (UIUC, USA, 2010) on the development of Laser-induced compressive shockwave technique to investigate the dynamic delamination of multilayer thin film structures. Following his PhD, Dr Tran has been working as a postdoctoral researcher in Mechanical Engineering department at the Northwestern University (Evanston, IL, USA) on an ONR-funded project to investigate the performance of composite material subjected to underwater blast loading. His current position is a lecturer in the department of Infrastructure Engineering, the University of Melbourne. He has been involved and contributed to number of successful ARC Linkage and Discovery projects, CRC-Project and other defence-related grants from Defence Materials Technology Centre (DMTC) and CSIRO Manufacturing Flagship. His research interests include dynamic behaviours of composite structure under extreme loadings (air/underwater blast, ballistics impacts, wind and fire) and design optimisation of bio-mimicry and lightweight composite materials. He has published 50 journal papers and a book chapter on related topics.

Natural convection in the near shore regions of lakes and reservoirs induced by the absorption of solar radiation

John Patterson
School of Civil Engineering, University of Sydney

4pm Friday 19 May 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

In the near shore region of lakes and reservoirs where the depth decreases towards the shore, the incidence of solar radiation results in a complex motion which provides a mechanism for natural convection heat and mass transfer between the shore and the main water body. This has potentially significant consequences for water quality issues and therefore management of the resource. The flow also has wider implications for any contained fluid heated by radiation. The mechanism operates as follows. Incident radiation is absorbed by the water column in an exponentially decaying manner. The absorption of the radiation in this fashion gives rise to a stable temperature stratification in the upper regions of the water column. However, in the shallower parts, some radiation reaches the bed, and is re-emitted as a heat flux. There are two consequences of this: first, the volumetric rate of heating is greater in the shallow part than in the deep, which gives rise to a horizontal temperature gradient and therefore to a circulation up the slope and outward at the surface from the shore line; and second, the emission of heat from the bed gives rise to a potentially unstable temperature gradient at the bed which may cause intermittent rising plumes. In this presentation I discuss the development of the understanding of the key characteristics of this complex flow, including recent numerical, scaling analysis and experimental investigations. In particular I report some recent laboratory experiments using concurrent PIV and shadowgraph for measurement and visualisation which demonstrate both the underlying circulation and the presence of the rising plumes. The characteristics of these flows are compared with earlier scaling analysis results, and the underlying mechanism descriptions verified. Finally, linear stability analysis provides some of the characteristics of the rising plumes.

John received his PhD in Applied Mechanics from the University of Queensland. He moved to University of Western Australia as postdoctoral fellow, took up an academic post in the Department Civil Engineering, and eventually became Head of the School of Environmental Engineering. He moved to James Cook University in Townsville in 1997, and became the founding head of the merged School of Engineering. In 2009 he joined the School of Civil Engineering at the University of Sydney as Professor of Fluid Mechanics and Director of the Wind Waves and Water Centre. From 2011–2015 he was Associate Dean Research for the Faculty of Engineering and IT at the University of Sydney. John has served on a number of journal editorial boards, and has been a member of ARC assessment panels, including what is now called the College of Experts, Centre of Excellence selection panels, and the ERA Evaluation Committee. He has held visiting research appointments in Canada, Germany, UK, and the USA. He has published over 200 articles in the international literature. He has generated in excess of $7M in ARC Discovery or equivalent research grants. John's research interests are historically in mixing and transport in lakes and reservoirs and the implications for water quality, and more recently, in natural convection flows, with environmental and industrial applications. His research encompasses theoretical, experimental and numerical investigations.

Adjoint-based optimal flow control and invariant solutions in compressible 2D cavity flows

Javier Otero
University of Melbourne

4pm Friday 12 May 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The advances in the understanding of compressible flows challenge researchers to tackle more ambitious problems that surpass human intuition. Hence, it becomes necessary to rely on methods that give an insight into the underlying physical mechanisms that govern these complex flows to perform non-trivial tasks such as flow control. To this end, an adjoint-based optimal flow control framework for compressible flows has been appended to an existing in-house DNS code (HiPSTAR). In particular, we focus our efforts on a 2D cavity flow at Re=5000, where we aim to reduce noise levels at the sensor location, by either reducing the overall sound radiation or altering the sound directivity. In addition, with minor coding effort, this framework is extended to permit the computation of both (stable and unstable) exact steady and periodic flow solutions in compressible flows over complex geometries. Thus, we present the families of exact periodic and steady solutions across Mach number, using the same 2D cavity flow setup with a lower Reynolds number (Re=2000). These two families of flow solutions are found to meet at the quasi-incompressible flow regime, where a stability analysis on both periodic and equilibrium solutions shows that they form a subcritical Hopf bifurcation

Javier did his undergraduate in Aerospace Engineering at the Technical University of Madrid (UPM), Spain (2008–2012); an MSc in Computational Fluid Dynamics at Cranfield University, UK (2012–2013); and PhD in development and application of an adjoint-based optimal flow control framework for compressible DNS at Southampton University, UK (2013–2017). He is now a postdoc working with Prof Richard Sandberg.

Shock driven instabilities in two-fluid plasmas

Vincent Wheatley
University of Queensland

4pm Friday 28 April 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

In inertial confinement fusion (ICF) experiments, a capsule filled with fuel is imploded by shock waves generated by laser-driven ablation. The goal is to ignite a fusion burn at the centre of the implosion, and have that burn propagate through and consume the inertially confined fuel. Hydrodynamic instabilities cause mixing between the capsule material and the fuel, which is highly detrimental to the propagation of a fusion burn. One of the key hydrodynamic instabilities in ICF is the Richtmyer-Meshkov instability (RMI). This occurs when a shock interacts with a perturbed density interface, such as the interface between the capsule and the fuel. In ICF and astrophysical applications, the RMI typically occurs in plasmas. Here, we study the RMI in the context of the ideal two-fluid, ion-electron, continuum equations. These couple a separate set of conservation equations for each species to the full Maxwell equations for the evolution of the electromagnetic fields. We focus on cases with and without an imposed magnetic fields and with Debye lengths ranging from a thousandth to a tenth of the interface perturbation wavelength. For all cases investigated, the behaviour of the flow is substantially different from that predicted by the Euler or ideal magnetohydrodynamics equations. Electric fields generated by charge separation cause interface oscillations, particularly in the electrons, that drive a secondary high-wavenumber instability. Consequently, the density interface is substantially more unstable than predicted by the Euler equations for all cases investigated. Self-generated magnetic fields are predicted within our simulations, but their orientation is such that they do not dampen the RMI. In ideal magnetohydrodynamics (MHD), it has been shown that in the presence of such a seed magnetic field, the growth of the RMI is suppressed by the transport of vorticity from the interface by MHD shocks. Our two-fluid plasma simulations reveal that while the RMI is suppressed in the presence of the seed field, the suppression mechanism varies depending on the plasma length-scales. Two-fluid plasma RMI simulations also reveal that the secondary, high-wavenumber, electron-driven interface instability is not suppressed by the presence of the seed field.

Dr Vincent Wheatley is a senior lecturer in the Centre for Hypersonics within the School of Mechanical and Mining Engineering at the University of Queensland. He obtained his PhD in Aeronautics from the California Institute of Technology in 2005. He also earned an MEngSc (Mechanical) and a BE (Mechanical and Space) from the University of Queensland (UQ). After completing his PhD in the US, Dr Wheatley spent two years as post-doctoral fellow at ETH Zurich. He was then a Lecturer in Aerospace Engineering at the University of Adelaide before taking up his position at UQ in 2009.

Non-laminar solutions for grooved Couette flow

Sabarish Vadarevu
University of Melbourne

4pm Friday 21 April 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The dynamical systems approach to turbulence has gained a lot of attention since the turn of the century. A large set of exact, symmetrical solutions of the Navier-Stokes equations have been found for canonical wall-bounded flows such as Couette, channel, and pipe flows at low Reynolds numbers (~ 400). The defining characteristic of these solutions is the vortex-streak structure that is known to play a crucial role in sustaining near-wall turbulence. In state-space, some of these solutions lie in the region that is densely visited by turbulence, while some others resemble the laminar solution and regulate laminar-turbulent transition. These solutions, along with their connections in state-space, form a skeleton for trajectories of turbulent flow. In addition to providing a description of turbulence dynamics, this approach also produces an equation-based, low-rank basis for use in flow control.

This vision of turbulence has not been extended to rough-walled flows despite their practical significance. We have taken the first steps in this direction by computing exact invariant solutions for plane Couette flow with (longitudinal) grooved walls, through continuation of known solutions for smooth-walled flows using a simple domain transformation method. We find that the principal role of such grooves is to localize the vortex-streak structure near the wall, as well as to reduce the bulk energy density while increasing the dissipation rate. Similar localization of the structure is considered to produce drag reduction (~ 10%) in riblet-mounted boundary layers. While these results are fairly primitive, especially due to the low Reynolds numbers involved, we believe that further efforts in this direction can help in optimizing riblet geometries for drag reduction, and provide insight on the influence of surface-roughness on the near-wall flow.

Sabarish Vadarevu is a research fellow in the fluids group at the University of Melbourne. He worked on his PhD (awaiting completion) in Aerospace Engineering at the University of Southampton, on computing exact invariant solutions in grooved plane shear flows. Prior to that, he obtained his Bachelors and Masters degrees from the Indian Institute of Technology-Madras, with his Masters thesis involving a stability analysis of partially premixed flames in laminar mixing layers. His current research is focused on linear modelling and control of wall-bounded turbulence.

LES for loss prediction in an axial compressor cascade at off-design incidences

Jake Leggett
University of Southampton

4pm Tuesday 11 April 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The work focuses on the prediction of loss in axial compressors. The majority of modern design methods use cheaper more accessible models, such as RANS, to predict flow performance. While these methods are reliable and easier to use they do not always provide the accuracy needed. The aim of the work presented is to improve our understanding of how models like RANS behave under more challenging conditions. Using higher fidelity LES simulations it is possible to assess the performance of RANS under such conditions and aid designers by highlighting short comings and improving the application of such tools.

Jake Leggett is a PhD candidate at the University of Southampton, studying loss prediction in axial compressors. Having previously done a bachelors in mechanical engineering followed by an MSc in CFD at Imperial College London. His interests are in compressible fluid dynamics and numerical methods, focussing on modelling behavior and the performance of numerical models in capturing physically realistic flows.

Confirmation: Numerical investigations of hemodynamic changes in stented coronary arteries

Bo Jiang
University of Melbourne

9am Wednesday 5 April 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Incomplete stent apposition (ISA) is sometimes found in stent deployment at complex lesions, and it is considered to be one of the causes of post-stenting complications, such as late stent thrombosis and restenosis. The presence of ISA leads to large recirculation bubbles behind the stent struts, which can reduce shear stress at the arterial wall that retards neointimal formation process and thus lead to complications. Computational fluid dynamics (CFD) simulations are performed on simplified two-dimensional axisymmetric arterial models with stents struts of square and circular cross-sectional shapes at a malapposition distance of 120 μm from the arterial wall. Under the condition of the same flow rate, both square and circular strut cases show that shorter period provides greater flow deceleration, leading to the formation of a larger recirculation bubble. With the same thickness, circular strut has a significant improvement over the square strut in terms of the size of the recirculation bubble, and therefore less likely to lead to complications. We also carried out three-dimensional computational fluid dynamics studies on a partially embedded coronary stent. Stent struts are partially embedded in portion of the circular and elliptical artery's circumference and malaposed elsewhere. Maximum malapposition distances (MD) used in this study are 0.255 mm (moderate) and 0.555 mm (severe). Time-averaged wall shear stress (TAWSS) is used as the haemodynamic metric to evaluate the clinical significance of ISA. TAWSS decreases along the circumferential direction as MD decreases from the maximum ISA side to the embedded side. The case with severe ISA has a larger area of high TAWSS on the arterial wall at the proximal end and low TAWSS at the distal end of the stent.

Theoretical framework for the upscaling of physical interactions in aquatic mobile-boundary flows

Konstantinos Papadopoulos
University of Melbourne, University of Aberdeen

4pm Tuesday 28 March 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Most environmental flows often exhibit high levels of spatial and temporal heterogeneity in hydrodynamic fields due to the effects of multi-scale roughness elements and their mobility. These effects are especially profound in the near-bed region. The development of a unifying framework for the integration and upscaling of the fluid mechanical, ecological and biomechanical processes occurring in such conditions is needed. Particular focus in this study is on the interactions of the fluid motion with the motions of aquatic plants and sediments in aquatic systems.

To cope with the temporal and spatial heterogeneity of the flow field near the flow interface with a second non-fluid phase, the governing equations of motion are averaged over time and space. To deal with the possible discontinuity of the averaged fields within the averaging domains, appropriate definitions and theorems for time and space averaging are used.

Appropriately formulated coupled double-averaged conservation equations are developed for fluid, sediment, and plant motions. Complementary equations for the second-order velocity moments introduced by the averaging process are also proposed. Due to the double-averaging methodology (i) the governing equations are up-scaled to the scales relevant to applications, (ii) the fluid motion is rigorously coupled with the non-fluid (plants or sediments) motions, and (iii) the effect of the moving interfacial boundary is introduced explicitly in the governing averaged equations.

This presentation introduces to the main concepts and key steps involved in the derivation of the double-averaged equations outlined above, discusses their application to the analysis of high-resolution data sets and identifies potential applications in mobile-boundary flow studies.

Konstantinos performed his PhD research on flow-biota and flow-sediment interactions in rivers and open-channel flows at the University of Aberdeen, where he was on a Marie-Curie Fellowship. Prior to that, he obtained an MSc in coastal engineering at the National Technical University of Athens (2013) and MEng in environmental engineering at the Technical University of Crete (2009).

High resolution simulation of dissolving ice-shelves in sea water

Bishakhdatta Gayen
Research School of Earth Sciences, Australian National University

4pm Friday 24 March 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Precise knowledge of ocean dynamics and interactions with the grounded ice at high latitudes is very crucial for predicting the sea-level rise and further development of adaptation strategies in a warming global climate. The physics of these ocean–ice interactions particularly related to small scale processes, is poorly understood which, along with limited observation constraints, leads to uncertainties in the predictions of future melt rate. We perform high resolution numerical simulation to investigate dissolving of ice into cold and salty sea water. The three coupled interface equations are used, along with the Boussinesq and non-hydrostatic governing equations of motion and equation of state for seawater, to solve for interface temperature, salinity, and melt rate. The main focus is on the rate of dissolving of ice at ambient water temperatures between −1°C and 2°C and salinity around 35 psu and the dependence on stratification (as characterizes many sites around Antarctica). Our simulation also shows boundary layer next to the ice face is dominated by turbulent motions. It is also important to quantify the difference between the melting of a vertical ice wall and the melting of a sloping ice shelf. The basal slope is observed to vary significantly, due to the formation of crevasses, channels and terraces. Our high-resolution simulations are designed for direct comparison with laboratory measurements and theory. The temperature and density structures found under Pine Island Glacier show several layers having a vertical scale that can also be explained by this study.

Bishakhdatta Gayen received a degree in Bachelor of Mechanical Engineering from Jadavpur University in Kolkata, India in 2006, and a M.S. degree in Engineering Science, majoring in Fluid Mechanics, from the Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore, in 2007. He was then awarded a University of California Graduate Fellowship and received an M.S. (2010) and Ph.D. (2012) from University of California, San Diego. He pursued research work on "Turbulence and Internal Waves in Tidal Flow over Topography" under the guidance of Prof. Sutanu Sarkar and was awarded the Andreas Acrivos Award for Outstanding Dissertation in Fluid Dynamics from the American Physical Society. He moved to Australia to pursue his postdoctoral research with Prof. Ross W. Griffiths at the Australian National University in Canberra. Bishakh is currently an Australian Research Council Discovery Early Career Fellow. His current research interests are nonlinear internal waves in the ocean, turbulent convection, modeling of Antarctic ice melting and Southern ocean dynamics.

The vortices of V. Strouhal

David Lo Jacono
Institut de Mécanique des Fluides de Toulouse

4pm Friday 17 March 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

During this seminar, I will talk about the link between Strouhal's findings and vortices. We learn that the Strouhal number characterises the shedding frequency of vortices behind a bluff-body, yet Strouhal never encountered a vortex. I will try to show the role played by various well-known researchers (Rayleigh, Bénard, Kármán, etc.), and others that are less well known yet have a key role towards understanding wake dynamics. Starting from the motivation of Strouhal and ending with the modern analysis of wakes in 1930, I will try to build an incomplete and brief history of 19th-20th century (wake) fluid mechanics focussing on motivation and experimental insights. This work was originally part of my ScD dissertation and further completed for the recently held colloquium "A century of Fluid Mechanics 1870-1970" celebrating the IMFT century anniversary.

Skin-friction and vorticity fields in wall-bounded flows and the attached eddy hypothesis

Min Chong
University of Melbourne

4pm Friday 10 March 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The invariants of the velocity gradient tensor have been used to study turbulent flow structures in order to extract information regarding the scales, kinematics and dynamics of these structures. These invariants cannot be used to study structures at a no-slip wall since they are all zero at the wall. However, the flow structures at the wall can be studied in terms of the invariants of the "no-slip tensor". Employing surface flow patterns generated using local solutions of the Navier-Stokes equations, the relationship between the surface skin-friction field and vorticity field will be explored. These local solutions, together with data from the Direct Numerical Simulations of channel flows, pipe flows and boundary layer flows may perhaps lead to a better model for the structure of attached eddies in wall bounded flows.

Linear estimation of large-scale structures in channel flow at Re_τ = 1000

Simon Illingworth
University of Melbourne

4pm Friday 3 March 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Given the time-resolved velocity field in a plane at a single wall-normal height, how well can one estimate, using a linear model alone, the time-resolved velocity field at other wall-normal heights? This question will be explored for channel flow at Re_τ = 1000 using data from the John Hopkins Turbulence database. Two different linear models will be explored. Each linear model has its origins in the Navier–Stokes equations.

Subfilter-scale stress modelling for large-eddy simulations

Amirreza Rouhi
University of Melbourne

4pm Friday 24 February 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

A subfilter-scale (SFS) stress model is developed for large-eddy simulations (LES) and is tested on various benchmark problems in both wall-resolved and wall-modelled LES. The basic ingredients of the proposed model are the model length-scale, and the model parameter. The model length-scale is defined as a fraction of the integral scale of the flow, decoupled from the grid. The portion of the resolved scales (LES resolution) appears as a user-defined model parameter, an advantage that the user decides the LES resolution.

The model parameter is determined based on a measure of LES resolution, the SFS activity. The user decides a value for the SFS activity (based on the affordable computational budget and expected accuracy), and the model parameter is calculated dynamically. Depending on how the SFS activity is enforced, two SFS models are proposed. In one approach the user assigns the global (volume averaged) contribution of SFS to the transport (global model), while in the second model (local model), SFS activity is decided locally (locally averaged). The models are tested on isotropic turbulence, channel flow, backward-facing step and separating boundary layer.

In wall-resolved LES, both global and local models perform quite accurately. Due to their near-wall behaviour, they result in accurate prediction of the flow on coarse grids. The backward-facing step also highlights the advantage of decoupling the model length-scale from the mesh. Despite the sharply refined grid near the step, the proposed SFS models yield a smooth, while physically consistent filter-width distribution, which minimizes errors when grid discontinuity is present.

Finally the model application is extended to wall-modelled LES and is tested on channel flow and separating boundary layer. Given the coarse resolution used in wall-modelled LES, near the wall most of the eddies become SFS and SFS activity is required to be locally increased. The results are in very good agreement with the data for the channel. Errors in the prediction of separation and reattachment are observed in the separated flow, that are somewhat improved with some modifications to the wall-layer model.

Amirreza Rouhi is a postdoctoral fellow in the fluids group at the university of Melbourne. Amirreza received his PhD in Mechanical Engineering from Queen's University of Canada under supervision of Prof. Piomelli, doing subfilter-scale (SFS) stress modelling as his PhD thesis. His other research interests include wall-modelled LES, spectral methods and rotating turbulence.

Completion: The minimal-span channel for rough-wall turbulent flows

Michael MacDonald
University of Melbourne

4pm Friday 10 February 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Turbulent flows over roughness are ubiquitous in engineering and geophysical applications, however their effects are primarily given through semi-empirical models and approximations. The accuracy of these methods is sensitive to the model and roughness topology in question, so that laboratory experiments and conventional direct numerical simulations (DNS) remain the desired standard in rough-wall studies. However, these techniques are expensive for both industry and researchers, making design predictions and the examination of rough-wall flows challenging. In this talk, we outline a framework termed the minimal-span channel in which fully resolved numerical simulations of rough-wall flows can be conducted at a reduced cost compared to conventional DNS. The minimal-span channel is used to simulate turbulent flow over a variety of roughness geometries that would otherwise be prohibitively expensive to study. Special attention is given to recent simulations of rectangular bars aligned in the spanwise direction, commonly called d-type roughness.

Completion: Reorganising turbulence using directional surface patterns

Kevin
University of Melbourne

4pm Wednesday 1 February 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

We attempt to passively reorganise wall turbulence using transitionally-rough surface patterns in the form of herring-bone riblets. The flow fields are investigated experimentally using large field-of-view particle image velocimetry in all orthogonal planes. The pronounced modification of the boundary layer suggests that a preferential arrangement of the naturally-occurring turbulence events may have been introduced. The spatial information captured in these multiple orientations enable us to clearly observe distinct turbulence events, such as the unstable outer layer, non-symmetrical vortical motions and strong streamwise-periodic events. Interestingly, our recent analysis indicates that the aforementioned events we thought was induced by the surface pattern, though weaker, are actually present in the smooth-wall (canonical) flows. In the average picture however, these structural attributes are masked by their random occurrences in space. This further suggests that we can passively reposition and perhaps manipulate large turbulence structures.

DNS study on relation between vorticity and vortex

Chaoqun Liu
University of Texas at Arlington, Texas, USA

2pm Monday 23 January 2017
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Vorticity, vorticity line, vorticity tube have rigorous mathematical definition, but the rigorous definition of vortex is still an open question. However, for long time until now, in many research papers and text books, vortex is defined as vorticity tube confined by vorticity lines without vorticity line leakage. Vortex is also considered as congregation of vorticity lines with larger vorticity. Our DNS study shows that vortex is not a vortex tube but open for vorticity line penetration, is not a congregation of vorticity lines but a dispersion of vorticity lines, is not a concentration of vorticity with larger vorticity but with smaller vorticity in most 3-D cases. In general, vortex is an intuitive concept of rotation core with weak dissipation, e.g. zero dissipation when it becomes rigid rotation. In addition, vortex is not a vorticity tube, being different from what suggested by many research papers and textbooks. At the laminar boundary layer, vorticity is large but we have no vortex. A core with 2 nearly pure rotations per second, where vorticity is small, is a vortex, but a core with 10000 strongly sheared rotation, where vorticity is very large but deformation is very large as well, may not be a vortex. Therefore, vorticity magnitude and vortex are irrelevant. Unlike solid body, vortex is always a mixture of vorticity and deformation and vortex is really defined as an open area (not tube) where deformation is weak and vorticity is dominant. A function so-called "Omega" is defined to identify the vortex and Omega=0.52 well represents the vortex area boundary. Vortex is mathematically defined as Omega >0.5 which means a place where vorticity overtakes deformation. The minimum gradient of "Omega" is well representing the vortex axis. Several examples including DNS for late flow transition and LES for shock vortex interaction have been tested and the outcome is promising.

Chaoqun Liu is a Distinguished Professor and CNSM Center Director in the Department of Mathematics, University of Texas at Arlington, Arlington, Texas, USA.

Completion: The structure and scaling of rough-wall turbulent boundary layers

Dougal Squire
University of Melbourne

2pm Wednesday 21 December 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Turbulent wall layers are a pervasive and influential feature in nature and engineering; common examples include the atmospheric and benthic layer (relevant to weather prediction and pollutant dispersion, for example), boundary layers developing on aerial, marine and terrestrial vehicles (such as aeroplanes, naval vessels, cars and trains), and flows in piping networks. These flows are characterised by high Reynolds numbers and, more often than not, surface roughness that exerts a dynamical effect on the flow. The latter may result from manufacturing defects, erosion and/or deposition, including that of living organisms. In this talk, we will present results from recent measurements of rough-wall turbulent boundary layers spanning a very wide range of friction and roughness Reynolds numbers. The results comprise 38 datasets and four experimental techniques, including hot-wire anemometry and particle image velocimetry. Our analysis will focus broadly on the relationship between the outer region flow and near-wall structures which are directly influenced by the roughness scale(s). Features of this relationship will be discussed using single-point statistics, measures of spatial structure, and the inner-outer interaction model of Marusic et al. (Science, 2010, vol. 329, pp. 193–196).

Confirmation: Entrainment and interface dynamics of turbulent plumes

Himanshu Mishra
University of Melbourne

10am Wednesday 9 November 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Turbulent plumes form when a fluid of one density is injected into another quiescent fluid with a different density. From violent volcanic eruptions to the smoke rising from a cigarette, turbulent plumes are omnipresent in nature at wide range of scales. One of the fundamental aspects in the understanding of turbulent plumes is the process of 'entrainment', the mixing of surrounding fluid into the plume. Unlike non-buoyant flows, plumes pose a challenge in using common optical measurement techniques like particle image velocimetry (PIV) and planar laser induced fluorescence (PLIF), because of the local changes in refractive index, when two fluids mix. This has led to most of the previous research being focused on global measurements of entrainment, whereas the local measurements, which are required for clearer understanding the entrainment phenomenon are practically non-existent. One of the ways to circumvent this problem is to match the refractive index of two solutions while maintaining the density difference, by adding certain chemicals to them. Alternatively, a measurement technique named Background Oriented Schlieren (BOS), which uses the local refractive index changes to quantify the local density variations, can be used.

With the final aim of understanding the process of entrainment in turbulent plumes, we present preliminary results for two experimental studies. (i) Velocity measurements in a vertical round axisymmetric turbulent jet in a newly constructed experimental facility, and (ii) free settling sphere in a sharp density interface using BOS.

Hydrodynamic forces on geometrically porous structures

Scott Draper
University of Western Australia

4pm Friday 28 October 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Prediction of hydrodynamic forces on geometrically porous structures is important for many applied problems in offshore engineering. However, the accurate prediction of these forces is complicated by the fact that the bulk velocity of the flow passing through a porous structure is a function of its porosity. In this talk I will present a theoretical model built on control volume arguments which may be used to estimate the bulk velocity of the flow passing through an isolated porous structure and, in turn, the resulting hydrodynamic force on the structure. Extensions of the model will be introduced to consider multiple porous structures in close proximity, inviscid shear flow and non-uniform porosity. It will be shown how these extensions have been used in recent work to optimise the layout of wind and tidal turbine arrays, estimate forces on offshore space frame structures and predict the flow through patches of aquatic vegetation.

Dr Scott Draper is a senior lecturer at the University of Western Australia. His research focuses on offshore fluid mechanics applied mostly to the oil and gas and renewable energy industries. This has included research on the stability of subsea infrastructure, the optimum arrangement of marine renewable energy devices and the hydrodynamics of floating bodies. He is fortunate to work with numerous industry partners including Shell, Woodside and Carnegie Wave Energy.

Pulse-burst PIV in high-speed flows

Steven Beresh
Aerosciences Department Sandia National Laboratories

4pm Wednesday 26 October 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Time-resolved particle image velocimetry (TR-PIV) has been achieved in a high-speed wind tunnel and a shock tube, providing velocity field movies of compressible turbulence events. The requirements of high-speed flows demand greater energy at faster pulse rates than possible with the TR-PIV systems developed for low-speed flows. This has been realized using a pulse-burst laser to obtain movies at up to 50 kHz with higher speeds possible at the cost of spatial resolution. The constraints imposed by use of a pulse-burst laser are a limited burst duration of 10.2 ms and a low duty cycle for data acquisition. Pulse-burst PIV has been demonstrated in a supersonic jet exhausting into a transonic crossflow and in transonic flow over a rectangular cavity. The velocity field sequences reveal the passage of turbulent structures and can be used to find velocity power spectra at every point in the field, providing spatial distributions of acoustic modes and revealing turbulence scaling laws. Additional applications in a shock tube show the transient onset of a von Kármán vortex street shed from a cylinder and particle drag in a shocked dense gas-solid flow. The present work represents the first use of TR-PIV in a high-speed ground test facility.

Steven J. Beresh is a Distinguished Member of the Technical Staff at Sandia National Laboratories in Albuquerque, New Mexico, U.S.A., where he has worked since 1999 and currently leads the Experimental Aerosciences Facility. He received his B.S. in Mechanical Engineering from Michigan State University in 1994 and his Ph.D. in Aerospace Engineering from The University of Texas at Austin in 1999. His research interests emphasize the use of optical diagnostics for compressible aerodynamics, particularly particle image velocimetry, but utilize a variety of laser-based instrumentation techniques and high-frequency surface sensors. He also is responsible for a wide range of wind tunnel testing and facility operation. He is an Associate Fellow of the American Institute for Aeronautics and Astronautics, the Chair for the AIAA Aerodynamic Measurement Technology Technical Committee, and is a past President of the Supersonic Tunnel Association International.

Development effects in turbulent boundary layers: zero and adverse-pressure-gradients

Ramis Örlü
KTH Royal Institute of Technology Sweden | Website

4pm Friday 14 October 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

In this talk, we discuss a number of different aspects pertinent to zero pressure gradient (ZPG) turbulent boundary layers (TBLs), and extend these to moderate and strong adverse pressure gradients (APGs). We start out with inflow and tripping effects in ZPG TBLs and show their persistence in the outer layer beyond Reynolds numbers (Re) that could be reached just few years ago by means of direct numerical simulations (DNS). When it comes to APG TBLs, the situation has usually been more blurry when considering literature data, due to their dependency on the pressure-gradient strength and streamwise history, besides Reynolds number effects. Based on different APG TBLs developing on flat plates and the suction side of a wing, we aim at distinguishing the importance of these parameters based on well-resolved in-house large-eddy and direct numerical simulations as well as wind tunnel experiments.

A recurring tool in the investigation of these topics is the diagnostic-plot concept, which indicates a linear dependence between the turbulence intensity and its mean velocity and turns out to be useful when establishing a well-behaved state in the outer layer of wall-bounded turbulent flows and determining the boundary layer edge in strong PG cases or TBLs on curved surfaces. A possible analogy between high-Re ZPG TBL flows and strong APG TBLs is also considered in light of negative wall-shear stress events.

Dr. Ramis Örlü received his M.Sc. (Dipl-.Ing.) in 2003 from the Ruhr University of Bochum, Germany in Mechanical Engineering and holds a Ph.D. in Fluid Mechanics (2009) KTH Royal Institute of Technology, Stockholm, Sweden. His research is focused on experimental methods and wall-bounded turbulent flows. Since 2009 and 2015 he works as a researcher and docent (in Experimental Fluid Physics), respectively, at the Linné FLOW Centre located at KTH. His research portfolio covers the experimental investigation of wall-bounded turbulent flows with strong numerical collaborations. Other active areas cover measurement technique development/correction for high-Reynolds number wall-bounded flows as well as applications in internal combustion engine related flows and various flow control strategies for separation delay and skin-friction drag reduction. Dr. Örlü has about 50 journal publications, and a wide network with other researchers in wall-bounded turbulence research.

Experimental investigation of two-phase flows in naval hydrodynamics

Paul Brandner
Australian Maritime College University of Tasmania

4pm Thursday 29 September 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Flow about ships and submarines vary from pure liquid-gas to pure liquid-vapour two-phase flows. They involve a large range of spatial and temporal scales and physical phenomena such that gaining insight remains a challenge via either experiment or computation. Naturally occurring microbubble populations in the ocean are of particular interest due to their potential to both affect and be affected by various phenomena and flow processes. They provide nuclei for cavitation inception on lifting surfaces of ships and submarines. Surface ships are prolific sources of polydisperse bubble populations that add to bubbly disperse flows about lifting surfaces. Cavitation and turbulence ultimately create complex long-lived large-scale microbubble laden wakes with properties significantly altered compared with those of the outer fluid. Techniques and results on various two-phase flow experiments in naval hydrodynamics are presented including microbubble generation for artificial seeding of nuclei and for PIV, microbubble disperse flows, cavitation inception, macroscopic cavitation phenomena, supercavitation, millimetre-bubble break and coalescence in turbulent shear flows and fluid-structure interaction.

Paul Brandner is research leader of the Cavitation Research Laboratory at the Australian Maritime College. His research interests include cavitation inception and dynamics, bubbly flows, supercavitation, fluid-structure interaction and hydroacoustics.

Completion: Numerical simulations of two-phase flow: Eulerian and Lagrangian predictions

Shuang Zhu
University of Melbourne

4.15pm Friday 16 September 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The term two-phase flow refers to any fluids flow of two coexisting media in motion. The difference between the media can be its thermodynamic state, called phase (e.g. gas, liquid or solid) and/or its multiple chemical components. Two-phase flow is common in many environmental systems including rain, snow, sandstorms, avalanches, sediment transport debris flow and countless other natural phenomena. It is also of extreme importance in many industrial and engineering applications, such as fluidised bed, droplet spray process, drug aerosol delivery, dense gas dispersion, particle deposition and pollution control.

A persistent theme throughout the study of two-phase flow is the need to model the detailed behaviour of these flow phenomena as the ability to predict the behaviour of these flow systems is central to safety, efficiency and effectiveness of those events and processes. Similar to single-phase flow research, the development of predictive models for the two-phase flow follows along three parallel paths, namely theoretical models, laboratory experiments and numerical simulations. There are some cases where full-scale experimental models and/or accurate theoretical models are possible. However, in many other cases, the use of those models can be impossible for many of reasons. Consequently, the predictive capability and physical understanding of two-phase flow problems heavily rely on the employment of numerical simulations. The study of dynamics of two-phase flow using high resolution numerical simulations has gained considerable momentum with the recent development in computational fluid dynamics (CFD) methodologies. In particular, the use of direct numerical simulations (DNS) resolves the entire range of spatial and temporal scales of the fluid motion, which provides new insights into the structure and dynamics of the two-phase flow.

Current two-phase flow simulations modelling can be broadly classified based on the treatment of the dispersed phase. Two types of approaches are prevalent, Eulerian approach and Lagrangian approach, which are based on the concept of the continuous phase described in the Eulerian reference frame in a flow domain with the dispersed phase described either in the Eulerian reference frame as the continuous phase, leading to the Eulerian approach (commonly known as the "two-fluid" model), or in the Lagrangain frame, leading to the Lagrangain approach (also known as the "trajectory" model). The primary motivation behind the present work is to perform high fidelity numerical simulations to study two-phase flow using both Eulerian and Lagrangian approaches on three different two-phase flow problems, namely gravity currents, particle-laden currents and enhanced targeted drug delivery in a vascular tree.This PhD work provides a much needed knowledge on different two-phase flow numerical modelling approaches by solving the aforementioned real life scenarios. Specifically, the success of this PhD work would lead to better two-phase computational fluid dynamics strategy.

Statistical characteristics of fully nonlinear potential deep water wave fields

Elena Sanina
Department of Infrastructure Engineering, University of Melbourne

4pm Friday 12 August 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

We present an analysis of long-term wave simulations performed using a fully nonlinear potential deep water wave model. The results of the simulations are compared with the spectra obtained using a variety of directional methods and are discussed in the context of their applications. The short-crestedness of a wave field is investigated in terms of the three-dimensional steepness defined as the vector whose magnitude is equal to the average steepness calculated along the vector direction in a horizontal plane. Several statistical characteristics of the surface elevation field and the wave spectrum development such as a non-uniformity of the wave spectrum and a migration of its peaks are discussed. The appearance of coherent structures on the ocean surface closely related to the tendency of high waves to occur in groups is analysed. Various features of the identified groups such as velocity of the groups, their lengths, lifetime and steepness are studied. A general analysis of the number of detected groups is also performed for the computed wave fields.

Dr. Elena Sanina is a Research Fellow in Ocean Engineering at the Department of Infrastructure Engineering, University of Melbourne. With Bachelor and Master degrees in Pure Mathematics from the Voronezh State University (Russia), she continued her studies performing research in the area of applied mathematics. In 2015 she completed her PhD in the Centre for Ocean Engineering, Science and Technology (COEST) at Swinburne University of Technology, then continued her work at the University of Melbourne in May 2016. Her research interests include non-linear waves, extreme waves and wave statistics.

Completion: Evolution of zero pressure gradient turbulent boundary layers

Will Lee
University of Melbourne

4pm Monday 8 August 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

An experimental investigation of evolving turbulent boundary layers in a tow tank facility is presented. The main aim is to study the dynamics and the development of coherent features in a turbulent boundary layer from the trip to a high-Reynolds number state. With this goal in mind, time-resolved particle image velocimetry measurements are performed with a towed plate to provide a unique view of the streamwise evolution of turbulent boundary layers. The merit of this temporally-resolved dataset is it enables us to investigate the evolving dynamic properties of coherent features specifically in the outer region of turbulent boundary layers. The results reveal the formation mechanism and temporal evolution of shear layers and their associated large-scale coherent motions. Based on these findings, a conceptual model which describes dynamic interactions of coherent features is proposed and discussed.

Self-sustaining motions and periodic orbits in statistically stationary homogeneous shear turbulence

Atsushi Sekimoto
Monash University

4pm Friday 5 August 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Homogeneous shear turbulence (HST) is the most canonical flow to investigate the shear-induced turbulence, where it is known that there are coherent structures, like velocity streaks and streamwise elongated vortices, similar to wall-bounded flows. The coherent structures and their dynamics are considered as incomplete realisations of nonlinear invariant solutions in the incompressible Navier–Stokes equation, i.e. equilibrium solutions or periodic orbits, which have been reported in the plane Couette, Poiseuille, pipe flow, isotropic turbulence, and so on. In this study, unstable periodic orbits (UPOs) in HST are investigated. The ideal HST grows indefinitely, and in simulations, the growing integral scale reaches the size of the computational domain. The largest-scale motion is restricted by the computational domain, and it grows, break-downs, and regenerates quasi-periodically, reminiscent to the self-sustaining motion and bursts in wall-bounded flow. The long term simulation of HST reaches the statistically stationary state. Direct numerical simulations (DNS) of statistically stationary homogeneous shear turbulence (SS-HST) are performed by a newly developed code. The box dependency is investigated to establish 'healthy' turbulence in the sense that the turbulence statistics are comparable to wall-bounded flow. SS-HST is essentially a minimal flow, and constrained by the spanwise box dimension as in minimal channel flow (Sekimoto, Dong & Jiménez, 2016 Phys. Fluids 28:035101; Flores & Jiménez 2010 Phys. Fluids 22:071704). In these good boxes, UPOs are numerically obtained. It represents a regeneration cycle of streamwise vortices and a streak, similar to the self-sustaining process (SSP) in wall-bounded flow.

The speaker received a PhD in thermo-fluid mechanics from Osaka University in 2011. He joined the Fluid Mechanics group in the Technical University of Madrid (UPM) as postdoctoral fellow. He is now a research fellow in the Laboratory for Turbulence Research in Aerospace & Combustion (LTRAC) in Monash University.

Fluid mechanic challenges at DST Group: from submarines to hypersonic vehicles

Malcolm Jones
Defence Science and Technology Group

4pm Friday 29 July 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The physics of fluid flow plays a dominant role in modelling vehicles traveling through the atmosphere or oceans. Whether it is predicting the drag experienced on a submarine, the lift generated by a flapping wing micro air vehicle or the heating loads acting a hypersonic vehicle. The Navier–Stokes equations governing these flows are known, but except for a few trivial cases cannot be solved directly. For this reason a range of alternative methods must be used. These include: approximations to the equations, numerical solutions, laboratory scale experiments and field trials. This talk will discuss how these methodologies are applied Defence Science and Technology Group to three example research areas: submarine hydrodynamics, flapping wing aerodynamics and hypersonic aerodynamics. While these vehicles appear diverse there is a common theme in the technical challenges and hence in the scientific approach taken.

Malcolm Jones received a PhD in fluid mechanics from The University of Melbourne in 1998. He received a degree in mechanical engineering from the same university in 1994. He joined the Defence Science and Technology Group in 2007 where he is currently employed as a Research Scientist in the Aerospace Division. Malcolm currently works on a range of diverse research projects which include: aerodynamics of flapping wings, aerodynamics of hypersonic vehicles (HIFiRE program), cavity aeroacoustics, boundary layer transition, and submarine hydrodynamics. Prior to joining DSTO he was employed as Research Fellow and lecturer in the Department of Mechanical Engineering at The University of Melbourne and also at the School of Mathematical Sciences, Queensland University of Technology.

Confirmation: DNS of a turbulent line plume in a confined region

Nitheesh George
University of Melbourne

4pm Thursday 28 July 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

We present results from a direct numerical simulation (DNS) of a turbulent line plume in a confined region with adiabatic side, top and bottom walls. The plume originates from a local line heat source of length, L, located at the centre of the bottom wall (z/H = 0) and it rises until it hits the top wall (z/H = 1) and spreads laterally to produce a buoyant fluid layer. Since the region is confined, the continuous supply of buoyant fluid forces the layer downwards, until it reaches the bottom wall, where the flow is said to be the asymptotic state (Baines and Turner 1969). In the present case, two Reynolds numbers, 3840 and 7680, are selected for plume lengths, L/H = 1, 2 and 4, where the Reynolds number of the plume is based on H and the buoyant velocity scale, F₀^1/3, where F₀ is buoyancy flux per unit length. The current simulations are validated against the analytical model presented by Baines and Turner (1969). The simulations exhibit a slow flapping motion of the confined line plume in the asymptotic state, which precludes a straightforward comparison with Baines and Turner's analytical model. For the purpose of comparing the analytical model, we have adapted a shifting method introduced by Hubner (2004), which improves agreement with the analytical model.

Confirmation: Optimal actuator and sensor placement for feedback flow control using the complex Ginzburg–Landau equation

Stephan Oehler
University of Melbourne

4pm Friday 8 July 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The process of selecting the optimal actuator and sensor positions for a single-input single-output flow control problem is investigated and its limitations are discussed. Multiple unstable modes and time delays between actuating and sensing often result in a challenging flow control problem. Previous approaches have focused on the use of modern control schemes, but these do not always provide the full picture and difficult control problems are often predicated on fundamental limitations, which are not made clear in a modern control framework. In this paper, a previously developed control scheme for the complex Ginzburg–Landau equation is presented, the challenges for different set-ups are analysed and it is shown how fundamental limitations can lead to an intractable control problem. The research explores the control schemes developed for the complex Ginzburg–Landau equation from the fundamental perspective, while further connecting the two fields of fluid dynamics and control theory.

The mitigation of pulsation in ventilated supercavities

Grant Skidmore
University of Melbourne

4pm Friday 1 July 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

It is possible for an underwater body to greatly reduce the drag caused by skin friction, using ventilated supercavitation. While the idea of ventilated supercavitation works well in theory, the process of generating a ventilated supercavity in practice is often plagued by pulsation. When a supercavity pulsates, the walls of the supercavity begin to periodically expand and contract. This can lead to the supercavity walls clipping the body, which can become problematic for stability of the supercavitating body. This seminar will explore the internal cavity pressure and near-field noise generated by experimental and computational supercavities. The results of the acoustic study revealed that the radiated acoustic pressure of pulsating supercavities is at least 40 dB greater than comparable twin vortex and re-entrant jet supercavity closure regimes. For pulsating supercavities it was also found that, at the pulsation frequency, the cavity interior pressure spectrum level was related to the near-field and far-field noise spectrum level through spherical spreading of the sound waves from the supercavity interface. As a result, the cavity interior pressure can be used as a measure of the radiated noise. The oscillatory nature of the internal cavity pressure time history was used to develop a method to mitigate supercavity pulsation. The method was explored with a numerical model, experiments, and CFD. The method is based on modulating the ventilation rate injected into a ventilated supercavity with the addition of a sinusoidal component. The effect of this modulation is the ventilated supercavity being effectively driven away from the resonance frequency. A wide range of ventilation rate modulation frequencies can cause the pulsating supercavity to transition into twin vortex closure. A reduction in the radiated noise accompanies the transition from pulsation to twin vortex closure, oftentimes 35 dB or more. Other modulation frequencies do not suppress pulsation but change the supercavity pulsation frequency.

Grant Skidmore is a postdoctoral researcher in Mechanical Engineering at the University of Melbourne. Grant obtained his PhD in aerospace engineering at the Pennsylvania State University. His research interests include supercavitation, drag reduction, hydrodynamic stability, free surface flows, multiphase flows, and turbulent wakes.

Confirmation: Geometric aspects of wall-flow dynamics

Rina Perven
University of Melbourne

4pm Tuesday 28 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Complex but coherent motions form and rapidly evolve within wall-bounded turbulent flows. Research over the past two decades broadly indicates that the momentum transported across the flow derives from the dynamics underlying these coherent motions. This spatial organization, and its inherent connection to the dynamics, motivates the present dissertation research. The local field line geometry pertaining to curvature (κ) and torsion (τ) has apparent connection to the dynamics of the flow, and preliminary results indicate that these geometrical properties change significantly with wall-normal position. One part of this research is thus to clarify the observed changes in the field line geometry with the known structure and scaling behaviours of the mean momentum equation. Towards this aim, the planar curvature of the streamlines at each point in streamwise–spanwise slices of existing boundary layer DNS has been computed for different wall-locations. The computation of κ and τ arise from the local construction of the Frenet–Serret coordinate frame. The present methods for estimating κ and τ are briefly described, as is the need to improve the efficiency of this computation. The present results are briefly described, as are the broader aims to better understand the relationships between the geometry and dynamics of wall-flows. A schedule of the future work to be conducted is outlined.

Laser light sheet profile and alignment effects on PIV performance

Kristian Grayson
University of Melbourne

11am Monday 27 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The sensitivity and impact of laser profile misalignment and shape mismatch on Particle Image Velocimetry (PIV) measurements are investigated in this study. While the effects of laser profile misalignment can be equivalent to an out-of-plane velocity component, light sheet mismatch can be identified and corrected prior to an experiment, decreasing PIV uncertainties. Synthetic particle image simulations are used to isolate and systematically vary laser profile mismatch parameters between successive PIV laser pulses. Two simulation cases are discussed, analysing the effects of a misalignment between two otherwise identical laser pulses, as well as a mismatch in the width of two laser profiles. Our results reveal a steady degradation in mean correlation coefficient as the laser profiles are increasingly mismatched in shape and alignment, coupled with a rapid rise in the detection of spurious vectors. These findings reinforce the need to consider laser sheet alignment and intensity distribution when seeking to capture high quality PIV measurements. The design of a modular and inexpensive laser profiling camera is outlined to enable robust and repeatable quantification of laser sheet overlap and beam characteristics. The profiling system is also found to be a valuable tool for laser diagnostics and aiding the setup of experiments. Various potential applications of this device are presented for PIV and other laser-based measurement techniques. Finally, preliminary results from PIV experiments which involve the deliberate misalignment of laser profiles are discussed. These data reiterate the trends observed in simulations, but also emphasise the coupled complexity of laser profile mismatch behaviour in experimental scenarios, placing the idealised simulation results in some context. Collectively, our findings highlight the importance of well-matched laser profiles. A more rigorous experimental quantification of these behaviours has the potential to enhance the quality of PIV results.

Towards a Moody-like diagram for turbulent boundary layers

Dale Pullin
Graduate Aerospace Laboratories, California Institute of Technology

4pm Friday 24 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

An empirical model is presented that describes a fully developed turbulent boundary layer in the presence of surface roughness with a nominal roughness length-scale that varies parametrically with stream-wise distance. For Reynolds numbers based on the outer velocity and stream-wise distance that are large, use is made of a simple model of the local turbulent mean-velocity profile that contains the Hama roughness correction for the asymptotic, fully rough regime. It is then shown that the skin friction coefficient is constant in the streamwise direction only for a linear streamwise roughness variation. This then gives a two parameter family of solutions for which the principal mean-flow parameters can be readily calculated. Results from this model are discussed for the zero-pressure-gradient turbulent boundary layer, and some comparisons with the experimental measurements of Kameda et al. (2008) are made. Trends obtained from the model are supported by wall-modeled, large-eddy simulation (LES) of the zero-pressure-gradient turbulent boundary layer at very large Reynolds numbers. Both model and LES results are consistent with the self-preservation arguments of Talluru et al. (2016). It is argued that the present model/LES can be interpreted as providing the asymptotically rough-wall equivalent of a Moody-like diagram for turbulent boundary layers in the presence of small-scale wall roughness.

The speaker is the von Kármán Professor of Aeronautics at the California Institute of Technology. His research interests include computational and theoretical fluid mechanics, vortex dynamics, compressible flow and shock dynamics, turbulence, and large-eddy simulation of turbulent flows.

The universal nature of freestream coherent structures

Philip Hall
Monash University

4pm Friday 17 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Exact coherent structures are nonlinear solutions of the Navier Stokes equations not necessarily arising as bifurcations from a base state. The solutions are thought to be relevant to bypass transition and fully turbulent flows. There are two fundamental types of solutions: vortex-wave interaction states and freestream coherent structures. A brief survey of the field is given and then freestream coherent structures are discussed in detail. These are shown to be driven in a nonlinear layer located where the base flow adjusts to its freestream value through exponentially small terms. It is shown that they are implicated in the production of passive near wall streaks in a layer distant log R from the wall where R is the Reynolds number. They are shown to be canonical states relevant to any 2D boundary layer, vortex sheet, jet or wake and relevant to weakly 3D forms of the latter flows. Remarkably it turns out that they also exist in any fully developed flow having a local maximum of the unperturbed velocity field. The relevance to fully turbulent flows is discussed.

Professor Philip Hall is the Head of School of Mathematics, Monash University. Professor Hall is presently on leave from Imperial College, where he became the first Director of the Institute of Mathematical Sciences and where he has been the Director of LFC (Laminar Flow Control)-UK. His research interests are in applied mathematics, particularly in nonlinear hydrodynamic stability theory, computational fluid dynamics, boundary layer control, convection, lubrication theory, chaotic fluid motion, geomorphology of rivers and coherent structures in high Reynolds number flows.

A voyage from the shifting grounds of existing data on zero-pressure-gradient turbulent boundary layers to infinite Reynolds number

Hassan M. Nagib
Illinois Institute of Technology, Chicago, USA

4pm Friday 10 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Aided by the requirement of consistency with the Reynolds-averaged momentum equation, the 'shifting grounds' are sufficiently consolidated to allow some firm conclusions on the asymptotic expansion of the streamwise normal stress normalized with the friction velocity; i.e. <u+u+>. A detailed analysis of direct numerical simulation data very close to the wall reveals that its inner near-wall asymptotic expansion must be of the form f0(y+) − f1(y+) / U∞+ + O(U∞+)−2, where f0 and f1 are O(1) functions fitted to data. This means, in particular, that the inner peak of the normal stress does not increase indefinitely as the logarithm of the Reynolds number but reaches a finite limit. The outer expansion of the normal stress <u+u+>, on the other hand, is constructed by fitting a large number of data from various sources. This exercise, aided by estimates of turbulence production and dissipation, reveals that the overlap region between inner and outer expansions of <u+u+> is its plateau or second maximum, extending to ybreak+ = O(U∞+), where the outher logarithmic decrease towards the boundary layer edge starts. The common part of the two expansions of <u+u+>, i.e. the height of the plateau or second maximum, is of the form A∞ − B∞ / U∞+ + ... with A∞ and B∞ constant. As a consequence, the logarithmic slope of the outer <u+u+> cannot be independent of the Reynolds number as suggested by 'attached eddy' models but must slowly decrease as 1 / U∞+. A speculative explanation is proposed for the puzzling finding that the overlap region of <u+u+> is centered near the lower edge of the mean velocity overlap, itself centered at y+ = O(1 / Reδ*0.5) with Reδ* the Reynolds number based on free stream velocity and displacement thickness. Similarities and differences between <u+u+> in ZPG TBLs and in pipe flow will be briefly discussed. Finally, the composite profile of <u+u+> is used with our mean velocity composite profile to demonstrate the detailed behavior of the so-called "diagnostic plot" with Reynolds number.

Professor Nagib is the John T. Rettaliata Distinguished Professor of Mechanical and Aerospace Engineering at the Illinois Institute of Technology, Chicago, Illinois, and the Founding Director of the Institute's Fluid Dynamics Research Center. His field of specialty is in fluid mechanics, turbulent flow and flow management and control. Professor Nagib is the recipient of a number of prestigious honors including being a Fellow of the American Physical Society, the American Association of Advancement of Science, the American Institute of Aeronautics and Astronautics, and the American Society of Mechanical Engineers.

Large-amplitude flapping of an inverted-flag in a uniform steady flow

John E. Sader
School of Mathematics and Statistics, University of Melbourne | Website

4pm Friday 3 June 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The dynamics of a cantilevered elastic sheet, with a uniform steady flow impinging on its clamped-end, have been studied widely and provide insight into the stability of flags and biological phenomena. Recent measurements show that reversing the sheet's orientation, with the flow impinging on its free-edge, dramatically alters its dynamics. In contrast to the conventional flag, which exhibits (small-amplitude) flutter above a critical flow speed, the inverted-flag displays large-amplitude flapping over a finite band of flow speeds. In this talk, a combination of mathematical theory, scaling analysis and measurement is used to investigate the origin of this large-amplitude flapping motion. Flapping is found to be periodic predominantly, with a transition to chaos as flow speed increases. These findings have implications to leaf motion and other biological processes, such as the dynamics of hair follicles, because they also can present an inverted-flag configuration. This collaborative work is with the Gharib group at the California Institute of Technology.

John E. Sader is a Professor in the School of Mathematics and Statistics, The University of Melbourne. He leads a theoretical group studying a range of topics including the dynamic response of nanoparticles under femtosecond laser excitation, mechanics of nanoelectromechanical devices, high Reynolds number flow of thin films and rarefied gas dynamics in nanoscale systems.

Turbulent horizontal convection and heat transfer

Bishakhdatta Gayen
Research School of Earth Sciences, Australian National University | Website

4pm Friday 27 May 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Horizontal convection (HC) is driven by a horizontal difference in temperature or heat flux at a single horizontal boundary of a fluid. In a thermally equilibrated state, net heat flux over the boundary is zero and circulation cell involves a horizontal boundary flow, turbulent plume motion at the end wall, and interior return flow which covers the entire the flow domain. HC may be a simple model of the meridonial overturning circulation (also known as global thermohaline circulation) based on such convective flow.

I will present three-dimensional convective circulation under differential heating on a single horizontal boundary of a rectangular channel, using direct and large eddy simulations over a wide range of Rayleigh numbers, Ra ∼ 10⁸–10¹⁵. A sequence of several stability transitions lead to a change from laminar to fully developed turbulent flow. At the smallest Ra, convection is maintained by a balance of viscous and buoyancy forces inside the thermal boundary layer, whereas at the largest Ra, inertia dominates over viscous stresses. This results in an enhancement of the overall heat transfer at Ra ≥ 10¹⁰, while both dynamical balances give Nu ∼ Ra^1/5. We have recently extended our study on circulation by applying thermal forcing on a lengthscale smaller than the domain, and with variation in both horizontal directions instead of traditional unidirectional gradient over the domain scale. Simulations show turbulence throughout the domain, a regime transition to a dominant domain-scale circulation, and a region of logarithmic velocity in the boundary layer. Scaling theory shows a new regime dominated by inertia of the symmetric interior large scale circulation, coupled to thermal dissipation in the boundary layer which explains the Nu ∼ Ra^1/4 behaviour.

Finally, I will show circulation in a rotating rectangular basin forced by a surface temperature difference but no wind stress. Here, our focus is on the geostrophic regime for the horizontal circulation with a strong buoyancy forcing (large Ra).

Buoyancy effects on turbulent entrainment

Dominik Krug
University of Melbourne

4pm Friday 6 May 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

In this talk, we will address the question which role buoyancy plays in the entrainment process in unstable configurations such as turbulent plumes. Based on data from direct numerical simulations of a temporal plume we show that the entrainment coefficient can be determined consistently using a global entrainment analysis in an integral framework as well as via a local approach. The latter is based on a study of the local propagation of the turbulent/non-turbulent interface (TNTI) relative to the fluid. We find that locally this process is dominated by small-scale diffusion which is amplified by interface-convolutions such that the total entrained flux is independent of viscosity. Further, we identify a direct buoyancy contribution to entrainment by the baroclinic torque which accounts for 8–12% percent of the entrained flux locally, comparable to the buoyancy contribution at an integral level (15%). It is concluded that the effect of the baroclinic torque is a mechanism which might lead to higher values of the entrainment coefficient in spatial plumes compared to jets. Finally, some results will be presented from applying the local analysis to a stable configuration. Here, buoyancy is observed to reduce the entrainment rate by reducing the surface area of the TNTI and we will consider how this affects the fractal scaling.

Conversion: Experimental factors influencing the quality of PIV results

Kristian Grayson
University of Melbourne

2pm Wednesday 27 April 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

Particle Image Velocimetry (PIV) is a useful tool for investigating the behaviour of turbulent flows, enabling the capture of undisturbed instantaneous velocity fields. The shape and alignment of the pulsed laser sheets used in PIV can have a significant impact on the quality of results, where laser profile mismatches can degrade correlation quality. This study develops two key experimental and analysis tools to aid the refinement of PIV techniques and systematically quantify the impact of unoptimised laser characteristics. Improvements to PIV simulation software enable much more experimentally realistic scenarios to be modelled and a laser profiling camera allows laser characteristics and alignment to be quantified in experimental setups for troubleshooting and analysis. The capabilities of these tools can ultimately be used to assess the practicality and possible benefits of more advanced and complex experimental PIV configurations.

Symmetries and symmetry breaking in bluff-body wakes

Justin Leontini
Swinburne University of Technology

4pm Friday 22 April 2016
Mechanical Engineering Seminar Room Level 3 (Room 311, Bldg 170)

The periodic vortex shedding in the wake of a circular cylinder is one of the most well-studied flows in fluid mechanics. It often serves as a canonical flow for understanding all bluff-body flows, as well as being important in its own right due to the common use of cylindrical cross sections in engineering structures. The periodic forcing on the structure caused by this vortex shedding can lead to large correlated forces and subsequent vibration. Often, it is desirable to control this motion to avoid failure; at other times, this motion may be deliberately amplified to be used as a source of energy harvesting. Here, I will present results looking at simple control strategies of this wake, such as periodic forcing and geometrical modifications such as streamlining. I will show that even these simple modifications to the base cylinder flow can result in significant changes in the flow, and that many of these changes can be understood by considering the interaction of the symmetries of the base flow and the imposed control.

Dr. Justin Leontini joined Swinburne in December 2013. Previous to this, he was an Australian Postdoctoral Fellow at Monash University, spent time working at the Centre for Maths and Information Science at CSIRO, and at l'Institut de la Recherche sur les Phenomenes Hors Equilibre (IRPHE) (trans. Dynamic Systems Research Institute) in Marseille, France. His research work has spanned fundamental fluid-structure interactions, planetary core flows, ship hydrodynamics, and flow stability. Recently, he has been building a project investigating the gas transport mechanisms in specialty ventilation machines in neonatal ICU wards. All of this research is focussed around understanding the fundamental physics of flow phenomena that have relevance to engineering or natural problems.