**Date:** Friday Dec 14, 2012**Time: **4:00pm**Location:** CTR Conference Room**Speaker:** Prof. Jan Nordström Linkoping University**Title:** New Developments for Initial Boundary Value Problems: Time Integration and Dual Consistency for High Order Finite Difference Approximations

**Abstract:** During the last decade, stable high order finite difference methods as well as finite volume methods applied to initial-boundary-value-problems have been developed. The stability is due to the use of so-called summation-by-parts operators(SBP), penalty techniques for implementing boundary and interface conditions, and the energy method for proving stability. In this talk we discuss new aspects of this technique including the relation to the initial-boundary-value-problem. By reusing the main ideas behind the development, new time-integration procedures and increased accuracy as well as new boundary conditions have been derived.

**Date:** Friday Nov 30, 2012**Time: **4:00pm**Location:** CTR Conference Room**Speaker:** Dr. Ricardo Rossi, Visiting Scholar, Laboratorio di Termofluidodinamica Computazionale, Universita di Bologna, Italy**Title:** Scalar Dispersion in Complex Flows: from applied research to an actual case study

**Abstract:** The transport of scalars in turbulent flows is one of the major research topics in many fields of environmental engineering. In this talk, a research project carried out at the Center forTurbulence Research is presented, where scalar dispersion analysis has beenapplied to the transport and mixing of pollutants in the atmosphere and in populated areas. The first part of the talk will provide an overview of research work aimed at improving the understanding and modeling of scalar dispersion in complex flows. Numerical experiments carried out by means of Direct Numerical Simulations (DNS) and then approximated using the Reynolds-Averaged Navier-Stokes (RANS) equations, will be presented to address briefly the role of numerics, scalar flux models and surface topography in dispersion analysis and prediction. In the second part of the talk an actual case study of flow and dispersion on a complex terrain will be presented and discussed, where RANS techniques developed during the research work have been applied to the release of pollutants from the portal of a road tunnel underconstruction in the north of Italy.

**Date:** Friday Nov 2, 2012**Time: **4:00pm**Location:** CTR Conference Room**Speaker:** Prof. John Dabiri, Professor, Engineering and Applied Science, Director, Center for Bioinspired Wind Energy, California Institute of Technology**Title:** Vortex-enhanced propulsion

**Abstract:** It has been previously suggested that the generation of coherent vortical structures in the near-wake of a self-propelled vehicle can improve its propulsive efficiency by manipulating the local pressure field and entrainment kinematics. We investigated these unsteady mechanisms analytically and in experiments. A self-propelled underwater vehicle was designed with the capability to operate using either steady-jet propulsion or a pulsed-jet mode that features the roll-up of large-scale vortex rings in the near-wake. The flow field is characterized by using a combination of planar laser-induced fluorescence, laser Doppler velocimetry and digital particle-image velocimetry. These tools enable measurement of vortex dynamics and entrainment during propulsion. The concept of vortex added-mass was used to deduce the local pressure field at the jet exit as a function of the shape and motion of the forming vortex rings. The propulsive efficiency of the vehicle was computed with the aid of towing experiments to quantify hydrodynamic drag. Finally, the overall vehicle efficiency was determined by monitoring the electrical power consumed by the vehicle in steady and unsteady propulsion modes. This measurement identified conditions under which the power required to create flow unsteadiness was offset by the improved vehicle efficiency. The experiments demonstrated that substantial increases in propulsive efficiency, over 50% greater than the performance of the steady-jet mode, can be achieved by using vortex formation to manipulate the near-wake properties. At higher vehicle speeds, the enhanced performance was sufficient to offset the energy cost of generating flow unsteadiness. An analytical model explains this enhanced performance in terms of the vortex added-mass and entrainment. The results suggest a potential mechanism to further enhance the performance of existing engineered propulsion systems. In addition, the analytical methods described here can be extended to examine more complex propulsion systems such as those of swimming and flying animals, for whom vortex formation is inevitable.

**Date:** Friday Oct 19, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Mathias Baekbo Andersen, Center for Turbulence Research, Postdoctoral Fellow

Title: Ion-selective membranes and overlimiting current

Abstract: Possible mechanisms for overlimiting current through aqueous ion-selective membranes (exceeding diffusion limitation) have been debated for half a century. Flows consistent with electro-osmotic instability have recently been observed in microfluidic experiments, but the existing theory neglects chemical effects and remains to be quantitatively tested. Eelectro-osmotic instability (EOI) is an electrohydrodynamic instability that appears as a result of a high current inducing an electrically charge boundary layer at an ion-selective surface. The quiescent state is unstable and a small perturbation grows by a positive feedback mechanism until viscous friction limits further growth. At this point the fluid instability consists of a periodic array of vortices reminiscent of Rayleigh–Bénard convection cells. At high enough currents the instability becomes chaotic as evidenced by the appearance of noise in the current through the system. Theoretical models have until now only been able explain the instability qualitatively, partly due to the highly nonlinear nature of the phenomenon and the disparate length scales. Furthermore, it may be that only a full 3D model can obtain quantitatively accurate results. One of our current research objectives is to investigate the combined effect of EOI and membrane chemistry on overlimiting current (OLC). In my talk I will therefore relate EOI to our recent model results [M.B. Andersen et al., PRL 109, 108301 (2012)] showing that chemical charge regulation and water self-ionization can lead to OLC by ‘‘current-induced membrane discharge’’ (CIMD) in ion-selective membranes. Our CIMD model shows a suppression of the extended space charge that leads to EOI and the combination of the two effects is expected to be important. In CIMD, and as in EOI, an electric current passes through the membrane, which leads to salt depletion and a locally large electric field that enhances the rate of water splitting. The resulting hydronium and hydroxide ions induce large pH variations in the system and in the membrane. Depending on its chemical nature the membrane loses its ion-selectivity and thereby allows for current above the nominal diffusion limit. CIMD is likely to act in parallel with EOI and a combined model is currently being pursued. Understanding the mechanisms related to OLC is important in technologies such as electrodialysis (water de-ionization), energy storage in supercapacitors, production of acids and bases by bipolar membranes and fuel cells.

Date: Friday Oct 5, 2012

Time: 4:00pm

Location: BLDG 530 ROOM 127

Speaker: Dr. Carlo Scalo, Center for Turbulence Research, Postdoctoral Fellow

Title: Large-eddy simulation and modeling of oxygen transport and depletion in water bodies

Abstract: The prediction of dissolved oxygen (DO) concentrations is critical for managing and monitoring marine ecosystems. Oxygen evolves in water bodies as a high-Schmidt-number passive scalar. It is entrained at the surface and transported down the water column by turbulent motions. Many natural factors can interfere with this mixing process such as stratification, which damps the turbulence, reducing the supply of oxygen to the near-bed region.Here, in the sediment layer, decomposition of dead organic matter (resulting from eutrophication, i.e. the ecosystem response to the introduction of excess artificial or natural nutrients) by oxygen-consuming bacteria can cause the DO concentration to drop to unsustainable levels for aquatic life making the sediment-oxygen uptake (SOU) the most important factor affecting oxygen depletion. Despite the considerable environmental and economical impacts of oxygen depletion the interaction among all the physical processes involved is still not well understood, with obvious consequences on the quality of modeling. A computational model for DO transfer from water to underlying flat and cohesive sediment beds populated with DO-absorbing bacteria has been developed and will be discussed in this talk. The model couples Large-Eddy Simulation of turbulent transport in the water-column, a biogeochemical model for DO transport and consumption in the sediment, and Darcy's Law for the pore water-driven solute dispersion and advection. The model's predictions compare well against experimental data for low friction-Reynolds numbers, Ret. The disagreement for higher Ret is investigated by progressively increasing the complexity of the model. A sensitivity analysis shows that the SOU is approximately proportional to the bacterial content of the sediment layer, and varies with respect to fluid dynamics conditions, in accordance to classic high-Schmidt-number mass-transfer laws. The non-linear transport dynamics responsible for sustaining a statistically steady SOU are investigated by temporal-and-spatial correlations and with the aid of instantaneous visualizations: the near-wall coherent structures modulate the diffusive sublayer, which exhibits complexspatial and temporal filtering behaviors; its slow and quasi-periodic regeneration cycle determines the streaky structure of the DO field at the sediment-water interface, retained in the deeper layers of the porous medium.– Finally, the model has been tested in a real geophysical framework, reproducing published in-situ DO measurements of a transitional flow in the bottom boundary layer of lake Alpnach (Switzerland). A simple model for the SOU is then derived for eventual inclusion in RANSE biogeochemical management-type models for similar applications.

Date: Monday Jul 30, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Martin Bazant, Departments of Chemical Engineering and Mathematics, Massachusetts Institute of Technology, Cambridge, MA

Title: Overlimiting Current and Shock Electrodialysis in Porous Media

Abstract: Salt transport in bulk electrolytes occurs by diffusion and convection, but in microfluidic devices and porous media, surface conduction and electro-osmotic flow also contribute to ionic fluxes. The classical theory of electrokinetic phenomena in porous media assumes linear response to a small voltage, where the electrolyte concentration is only weakly perturbed. When a large voltage or concentration gradient is imposed, some surprising nonlinear electrokinetic phenomena result from the competition between bulk and interfacial transport in a microstructure. At constant voltage, the microstructure can sustain an over-limiting current (exceeding diffusion limitation) without any hydrodynamic or chemical instability. At constant current, a “deionization shock” can propagate through the microstructure, leaving behind a macroscopic region depleted of ions and particles. This talk will present the mathematical theory and new experimental evidence for surface-driven overlimiting current and deionization shocks in porous glass frits, as well as applications to water deionization by ``shock electrodialysis”. This work builds on the seminal 2009 paper of Mani, Zangle and Santiago at Stanford on concentration shocks in microchannels by extension to porous media, where current and flow directions are decoupled.

Date: Tuesday Jul 17, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Yury S. Kachanov, Institute of Theoretical and Applied Mechanics, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russian Federation

Title: The Deterministic Turbulence – Philosophical Abstraction or Reality?

Abstract: This paper concentrates on physical and philosophical aspects of the problem of deterministic turbulence, the idea of which has been suggested by the author about a decade ago and realized later experimentally together with two younger colleagues. The answer to the question formulated in the paper title is: “The deterministic turbulence is a reality.” In contrast to the usual (random) turbulence, the deterministic turbulent flows have reproducible instantaneous structure, representing one particular realization from infinite number of possible ones. The discovery of the deterministic turbulence is a breakthrough in the fields of: fluid dynamics, nonlinear science on turbulence and chaos, and in philosophy. It is shown in the present paper that the deterministic turbulence can be regarded as an inherent consequence of the idea on determinism of laws of nature and predictability of real physical systems, even unstable ones, occurred at least in cases of relatively short time intervals. These features enable reproducibility of behavior of even very complex real systems having infinite, practically, number of degrees of freedom. Such behavior only appears chaotic, due to its great complexity, but, simultaneously, can be deterministic, i.e. reproducible, in the main, from one realization to another. One has note that the deterministic turbulence can exist (and can be produced) not in all flows (in the majority of them only usual, stochastic turbulence is possible).

Date: Thursday Jul 5, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Barry Koren, Numerical Mathematics, Mathematical Institute, Leiden University

Title: Runge-Kutta methods for the incompressible Navier-Stokes equations

Abstract: Time integration of the incompressible Navier-Stokes equations with Runge-Kutta methods is not straightforward due to the differential-algebraic nature of the equations.We consider the temporal order of accuracy of velocity and pressure and focus on one specific class of Runge-Kutta methods, namely symplectic Runge-Kutta methods. We apply existing theory on Runge-Kutta methods for differential algebraic equations [1] to the incompressible Navier Stokes equations. Explicit Runge-Kutta methods are in general a good choice for integrating non-stiff differential equations. They combine good stability, high accuracy and low cost. If explicit Runge-Kutta methods are used, both velocity and pressure can be computed to the classical (‘ODE’) order of accuracy, except in two important cases [2]. First, if the boundary conditions for the normal velocity component depend on time, the order of accuracy of the pressure is affected. Second, if the mesh is time-dependent, the order of accuracy of both velocity and pressure is affected.We propose a number of new explicit Runge-Kutta methods for these cases that are third- and fourth-order accurate for the velocity and second-order for the pressure. The second-order accuracy of the pressure is obtained with a new technique based on reconstruction of instantaneous pressure values from time-averaged values. This technique is also applicable to obtain a second-order accurate pressure when implicit Runge-Kutta methods are used. Implicit Runge-Kutta methods are better suited for stiff problems, as is for example the case with flows featuring boundary layers. Another attractive feature is that they can be constructed such that the (kinetic) energy of the flow is conserved (in absence of diffusion) [3]. Such ‘energy-conserving’ or ‘symplectic’ Runge-Kutta methods [1] do not distort the energy cascade from large to small scales, an important property for turbulence simulations. Moreover, these Runge-Kutta methods, combined with energy-conserving space discretizations [4, 5], lead to methods that are stable for any mesh size and any step size. As a test case we consider the roll-up of an inviscid shear layer, see figure 1. The spatial discretization is a second-order finite volume method on a staggered grid, which is energy-conserving in absence of viscosity. Figure 1a shows the vorticity field at t = 8. Small wiggles appear because there is no viscosity to damp the smallest scales, but the wiggles do not lead to blow-up of the solution if the initial energy of the flow is bounded. Figure 1b shows that the symplectic single-stage Gauss 2 (implicit midpoint, second-order), two-stage Gauss 4 (fourth-order) and two-stage Radau IIB (third-order) methods conserve energy to machine precision, independent of the time step. Well known implicit methods such as Crank-Nicolson (second-order) or two-stage Radau IIA (third-order) do not have this favorable property. In ongoing work we combine these energy-conserving Runge-Kutta methods with methods that are suitable for handling very stiff diffusive terms. This results in a new class of additive Runge-Kutta methods, of which the two-stage, third-order Radau IIA/B method is an example. Other test cases involving both convection- and diffusion-dominated flow regions will show the effectiveness of these new methods.

Date: Friday Jun 22, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. H. Dogus Akaydin. The Grove School of Engineering, The City College of The City University of New York

Title: Flow-Powered Piezoelectric Energy Harvesters: A Case of Aeroelectromechanics

Abstract: "Smart structures” that respond to electrical stimuli are used in fluidic environments for many purposes including energy harvesting, sensing, active flow control and bio-inspired locomotion. In this talk, the use of flexible piezoelectric structures for flow-powered energy harvesting will be introduced. The design and performance of several harvesters, which exploit the wakes of bluff bodies, turbulent boundary layers, vortex rings and self-excited aeroelastic oscillations, will be discussed. Due to their mechanical simplicity and low speed operation, the aeroelastic harvesters can become an alternative to micro-turbines for powering small wireless sensor nodes. According to our wind tunnel tests, a self-excited harvester utilizing vortex-induced vibrations on circular cylinder generated a maximum of 0.1 mW of electrical power while resonating at a flow speed of 1.2 m/s. However, when the shape of the cylinder is changed to induce galloping instability, the harvested power rapidly increased with flow speed from 0.11 mW at 1.2 m/s to nearly 2mW at 2.7 m/s. The inclusion of a flexible piezoelectric structure in a fluid flow adds an electrical component to the complexity of the fluid-structure interaction problem and results in an “aeroelectromechanical” system. In the remainder of the talk, a computational framework to model the behavior of aeroelectromechanical systems will be introduced. The application of this framework to model the behavior of a piezoelectric beam in the wake of a circular cylinder will be demonstrated. Finally, future extensions of the framework for modeling other aeroelectromechanical configurations, including active flow control and bio-inspired locomotion, will be briefly discussed.

Date: Friday Jun 8, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Alexander Bihlo, Postdoctoral Fellow, Centre de recherches mathematiques at the Universite de Montreal, Canada

Title: Invariant turbulence modeling

Abstract: Symmetries are among the most successfully employed concepts in science and mathematics. They form a cornerstone of various branches of physics, such as classical and quantum mechanics, particle physics and relativity. The governing equations of hydrodynamics generally possess wide symmetry groups and therefore there is a great potential to exploit these symmetries so as to derive similarity solutions, conservation laws and invariants or to study the effects of symmetry breaking due to the presence of boundaries or additional body-forces. However, to date, symmetries are often used in a non-explicit or indirect way in hydrodynamics and turbulence theory. On the other hand, there exist powerful and general methods introduced in the field of group analysis of differential equations which, when suitably adapted, can be readily applied to the aforementioned fields. In this talk we will introduce an algorithmic method which allows associating with a given object its invariant counterpart. The object under consideration can be, e.g., a turbulence closure model or a finite-difference discretization of a differential equation, which can then be invariantized to yield a turbulence model or a finite-difference discretization that is invariant under the same Lie point symmetry group as admitted by the original governing equations of hydrodynamics. This method can therefore be used to correct artificial symmetry breaking due to non-appropriately designed turbulence models. As an example it is shown that classical hyperdiffusion as used in two-dimensional (decaying) turbulence simulations violates the symmetries of the incompressible Euler equations. Invariantization of these hyperdiffusion terms yields symmetry-preserving but nonlinear diffusion-like terms. Using the notion of differential invariants it is demonstrated that the invariantized hyperdiffusion models can be modified with quite some flexibility while still preserving their desired invariance characteristics. First numerical tests show that the invariant hyperdiffusion schemes which can be obtained by this method might be able to reproduce the -3 slope of the energy spectrum in the enstrophy inertial range.

Date: Thursday Jun 7, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Ahmad Shakibaeinia, Research Associate, Faculty of Engineering & Applied Science University of Regina, Canada

Title: Mesh-free particle modeling of multiphase flows

Abstract: The major challenge for numerical modeling of the multiphase systems has long been handeling the large deformation and fragmentation of the interfaces and boundaries. A recent strong interest in the field of computational fluid dynamics is focused on development of the next generation of numerical methods, the mesh-free particle (Lagrangian) method. The main goal of these methods is to facilitate the simulation of increasingly demanding problems that are involved large interfacial deformations, complex geometries and discontinuities. Mesh-free particle methods (also called particle methods) use a set of particles without any connectivity to represent the continuum and record the state and movement of the system. Each particle possesses a set of field variable (e.g., mass, momentum) and moves in a Lagrangian frame based on its velocity. The objective of this study is to develop a mesh-free particle methods based on the Moving particle Semi-implicit (MPS) method for modeling of multiphase systems in a broad verity of conditions. In this talk the MPS mathematical fundamentals and numerical algorithms for modeling of single-phase and multiphase fluid flow systems are presented. Moreover, the accuracy and performance of the MPS method is addressed through some fundamental and practical sample problems. Finally, the future works on MPS and mesh-free particle modeling of multiphase flows are briefly explained.

Date: Friday May 18, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Ricardo Garcia-Mayoral, Center for Turbulence Research, Postdoctoral Fellow

Title: THE INTERACTION OF RIBLETS WITH WALL-BOUNDED TURBULENT FLOWS

Abstract: Riblets are small surface protrusions that can reduce turbulent friction when aligned in the direction of the overlying flow. The reduction is a function of their size, measured in wall units. For vanishingly small sizes, the reduction is linear, but this linear behavior breaks down for a given, geometry-dependent size, limiting riblet performance. The mechanism of this breakdown was pre- viously not understood, and it was thus difficult to predict riblet optimum size and performance. Analyzing the available experimental data, the linear and breakdown regimes are found to collapse well for diverse configurations, when the size is scaled with the length ?+g = (A+g )1/2, based in the groove cross-section Ag. The linear drag reduction is found to break down for an optimum ?+g ,opt 11. To understand the breakdown, we analyze DNSs of turbulent channels with riblets, with sizes covering the full drag reducing range, and up to Re 550. As the size increases from ?+g,opt, coherent spanwise rollers begin to appear immediately above the riblets, growing rapidly in intensity. The extra Reynolds stress that they generate accounts quantitatively for the drag degradation. The rollers are similar to those found over porous surfaces and plant canopies, and can be traced to a Kelvin–Helmholtz-like instability, associated with the relax- ation of the impermeability condition for the wall-normal velocity. An inviscid model for the instability confirms its nature, agreeing well with the observed perturbation wavelengths and shapes. Using this model, the key parameter for the onset of the instability is identified as a ‘penetration’ length L+w , which for conventional riblet geometries is proportional to ?+g . The instability onset, L+w 4, actually corresponds to the empirical breakdown size ?+g 11.

Date: Friday May 11, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Hiroyuki Abe, Japan Aerospace Exploration Agency

Title: DNS and modeling of a separated turbulent boundary layer

Abstract: Separation of a turbulent boundary layer is one of the most challenging research topics in aeronautics. The data, which can serve as reference data for developing turbulence models, are however still limited. Over the past several years, we have performed DNS of a separated turbulent boundary layer, where a separation bubble is created by imposing blowing and suction at the upper boundary. The inlet Reynolds numbers based on the freestream velocity and the momentum thickness are 300, 600 and 900, the latter value being three times larger than that of the seminal DNS works (Spalart and Coleman 1997; Na and Moin 1998). The objectives of the present study are to clarify effects of the Reynolds number in a separation bubble and also to develop turbulence models. In this talk, some representative DNS results are first described (e.g. negative production of the turbulent kinetic energy, large-scale spanwise meandering of the separation line and large-scale structures in the bubble). Next, results regarding the RANS (low Re k-epsilon) model testing are presented; the model tends to predict the separation point earlier than the DNS. Finally, the future works on DNS and modeling (RANS and LES) are briefly described.

Date: Friday Apr 20, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dr. Takao Suzuki, The Boeing Company

Title: Hybrid unsteady-flow simulation combining PIV/PTV and DNS: Introduction to the reduced-order Kalman filter

Abstract: Capability of state-of-the-art techniques integrating experimental and computational fluid dynamics is recently expanding. In particular, time-resolved particle image velocimetry (PIV) has opened the door to hybrid simulation driven by experimental data. We have developed a hybrid unsteady-flow simulation technique combining particle tracking velocimetry (PTV) and direct numerical simulation (DNS), and demonstrated its capability by solving laminar planar-jet flows. Unsteady velocity fields on a laser sheet in a water tunnel are measured with time-resolved PTV; subsequently, PTV velocity fields are rectified in a least-squares sense so that the equation of continuity is satisfied, and they are transplanted to a two-dimensional incompressible Navier-Stokes solver by setting a multiple of the computational time-step equal to the frame rate of the PTV system. In this parental algorithm, the hybrid velocity field is given by a linear combination between the PIV/PTV velocity field and that marched from a previous time-step with the DNS. The constant weight between them is determined such that all growing modes of the numerical system are suppressed. As a result, the unsteady hybrid velocity field is converged to that of the measured one over time, and the unsteady pressure field can be simultaneously acquired. The resultant set of flow quantities satisfies the governing equations, and their resolution becomes comparable to that of numerical simulation with the noise level much lower than the original PTV data. To further improve the filtering capability for noisy PIV/PTV data at higher Reynolds numbers, this study introduces the reduced-order Kalman filter, which is widely used for data assimilation. The new algorithm temporally and spatially optimizes the weight function for the time-resolved PTV feedback; thus, it improves the spatial-filtering function, and the hybrid velocity fields converge faster on time-step bases, while the drawback is huge computational time and memory.

Date: Friday Apr 6, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Jan Nordström, Dept. of Mathematics, Linkoping University, Sweden

Title: Recent Developments in Computational Mathematics at Linköping University

Abstract: The current activities in computational mathematics at the mathematics department of Linköping University regarding methods for fluid flow, wave-propagation, heat conduction and signal propagation will be discussed. The talk will give a brief presentation of the specific theoretical problems, the specific research issues and possibly provide a ground for a discussion about future collaborations in related application fields.

Date: Friday Mar 23, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Brian Launder, Turbulence Mechanics Group, University of Manchester, UK

Title: Back to the Future: Flettner-Thom Rotors for low-carbon maritime propulsion

Abstract: The seminar will first review the major steps taken over the past 90 years to exploit the Magnus effect for ship propulsion (i.e. the lift imparted to a rotating cylinder past which a fluid stream passes) and the ambitious plan to use such vessels to combat the effects of global warming. The current computational work being undertaken by the Turbulence Mechanics Group at Manchester to examine this problem is then summarized. A range of strategies is being pursued from steady RANS to LES. The seminar will, however, focus on an unsteady RANS approach. The effect of the pure Flettner rotor is first examined and results compared with the available DNS, LES and experimental data (the first two being at Reynolds numbers orders of magnitude below those at which Flettner rotors would actually operate). The effect of adding discs to the rotor as first proposed by Alexander Thom is next examined. Current results have failed to reproduce the extraordinary increase in lift coefficient that Thom reported though they have captured the negative drag coefficients that he found occurred when the rotor speed exceeds four times the air speed. Finally an indication is given of immediate and medium-term further research needed.

Date: Friday Mar 9, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Dmitry Kotov, Postdoctoral Scholar, Stanford University, Center for Turbulence Research

Title: Performance of High Order Shock-Capturing Schemes for Stiff Source Terms & Discontinuities

Abstract: In compressible turbulent combustion/nonequilibrium flows, the constructions of numerical schemes for (a) stable and accurate simulation of turbulence with strong shocks, and (b) obtaining correct propagation speed of discontinuities for stiff reacting terms on ``coarse grids" share one important ingredient - minimization of numerical dissipation while maintaining numerical stability. This dual requirement to achieve both numerical stability and accuracy with zero or minimal use of numerical dissipation is most often conflicting for existing schemes that were designed for non-reacting flows. The goal of this paper is to relate numerical dissipations that are inherited in a selected set of high order shock- capturing schemes with the onset of wrong propagation speed of discontinuities for two representative stiff detonation wave problems.

Date: Friday Feb 24, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Rodney O. Fox, Anson Marston Distinguished Professor of Engineering, Iowa State University

Title: Quadrature-Based Moments Methods

Abstract: Kinetic theory is a useful theoretical framework for developing multiphase flow models that account for complex physics (e.g., particle trajectory crossings, particle size distributions, etc.). For most applications, direct solution of the kinetic equation is intractable due to the high-dimensionality of the phase space. Thus a key challenge is to reduce the dimensionality of the problem without losing the underlying physics. At the same time, the reduced description must be numerically tractable and possess the favorable attributes of the original kinetic equation (e.g. hyperbolic, conservation of mass/momentum, etc.) Starting from the seminal work of McGraw (1) on the quadrature method of moments (QMOM), we have developed a general closure approximation referred to as quadrature-based moment methods (2; 3; 4). The basic idea behind these methods is to use the local (in space and time) values of the moments to reconstruct a well-defined local distribution function (i.e. non-negative, compact support, etc.). The reconstructed distribution function is then used to close the moment transport equations (e.g. spatial fluxes, nonlinear source terms, etc.). In this seminar, I will present the underlying theoretical and numerical issues associated with quadrature-based reconstructions. The transport of moments in real space, and its numerical representation in terms of fluxes, plays a critical role in determining whether a moment set is realizable. Using selected multiphase flow examples, I will introduce recent work on realizable high-order flux reconstructions developed specifically for finite-volume schemes (5).

Date: Friday Feb 17, 2012

Time: 4:00pm

Location: CTR Conference Room

Speaker: Prof. Mihailo Jovanovic, Dept. of Electrical and Computer Engineering, University of Minnesota

Title: Model-based design of transverse wall oscillations for turbulent drag reduction

Abstract: Over the last two decades, both experiments and simulations have demonstrated that transverse wall oscillations with properly selected amplitude and frequency can reduce turbulent drag by as much as 40 percent. In this talk, we will describe a model-based approach for designing oscillations that suppress turbulence in a channel flow. We utilize judiciously selected linearization of the flow with control in conjunction with turbulence modeling to determine skin-friction drag in a simulation-free manner. The turbulent viscosity hypothesis is used to quantify the effect of fluctuations on the mean velocity in the flow subject to control. In contrast to the traditional approach, we determine the turbulent viscosity from the second order statistics of the linearized model driven by white-in-time stochastic forcing. The spatial power spectrum of the forcing is selected to ensure that the linearized model for the flow with no control reproduces the turbulent energy spectrum. The resulting correction to the turbulent mean velocity induced by small amplitude wall movements is then used to identify the optimal frequency of drag reducing oscillations. In addition, the control net efficiency and the turbulent flow structures that we obtain agree well with the results of direct numerical simulations and experiments. This demonstrates the predictive power of our model-based approach to controlling turbulent flows and is expected to pave the way for successful flow control at higher Reynolds numbers than currently possible.