-
The influence of the vorticity-scalar correlation on mixing
Authors:
Xi-Yuan Yin,
Wesley Agoua,
Tong Wu,
Wouter J. T. Bos
Abstract:
We investigate the role of the correlation between a scalar quantity and the vorticity in two-dimensional mixing at infinite Péclet number. We assess, using a diffusivity independent mixing-norm, the dynamics of both Galerkin-truncated ensembles and freely evolving two-dimensional scalar mixing. Both statistical mechanics and numerical experiments show how the mixing-rate is attenuated when vortic…
▽ More
We investigate the role of the correlation between a scalar quantity and the vorticity in two-dimensional mixing at infinite Péclet number. We assess, using a diffusivity independent mixing-norm, the dynamics of both Galerkin-truncated ensembles and freely evolving two-dimensional scalar mixing. Both statistical mechanics and numerical experiments show how the mixing-rate is attenuated when vorticity and scalar are initially correlated. Since the vorticity is shown to be a poorly mixing scalar, the results suggest that, in general, mixing can be enhanced by minimizing the correlation between vorticity and passive scalar.
△ Less
Submitted 26 February, 2024;
originally announced February 2024.
-
Gemini: A Family of Highly Capable Multimodal Models
Authors:
Gemini Team,
Rohan Anil,
Sebastian Borgeaud,
Jean-Baptiste Alayrac,
Jiahui Yu,
Radu Soricut,
Johan Schalkwyk,
Andrew M. Dai,
Anja Hauth,
Katie Millican,
David Silver,
Melvin Johnson,
Ioannis Antonoglou,
Julian Schrittwieser,
Amelia Glaese,
Jilin Chen,
Emily Pitler,
Timothy Lillicrap,
Angeliki Lazaridou,
Orhan Firat,
James Molloy,
Michael Isard,
Paul R. Barham,
Tom Hennigan,
Benjamin Lee
, et al. (1325 additional authors not shown)
Abstract:
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultr…
▽ More
This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.
△ Less
Submitted 17 June, 2024; v1 submitted 18 December, 2023;
originally announced December 2023.
-
Convergence guarantees for forward gradient descent in the linear regression model
Authors:
Thijs Bos,
Johannes Schmidt-Hieber
Abstract:
Renewed interest in the relationship between artificial and biological neural networks motivates the study of gradient-free methods. Considering the linear regression model with random design, we theoretically analyze in this work the biologically motivated (weight-perturbed) forward gradient scheme that is based on random linear combination of the gradient. If d denotes the number of parameters a…
▽ More
Renewed interest in the relationship between artificial and biological neural networks motivates the study of gradient-free methods. Considering the linear regression model with random design, we theoretically analyze in this work the biologically motivated (weight-perturbed) forward gradient scheme that is based on random linear combination of the gradient. If d denotes the number of parameters and k the number of samples, we prove that the mean squared error of this method converges for $k\gtrsim d^2\log(d)$ with rate $d^2\log(d)/k.$ Compared to the dimension dependence d for stochastic gradient descent, an additional factor $d\log(d)$ occurs.
△ Less
Submitted 20 June, 2024; v1 submitted 26 September, 2023;
originally announced September 2023.
-
Space-local Navier--Stokes turbulence
Authors:
Ryo Araki,
Wouter J. T. Bos,
Susumu Goto
Abstract:
We investigate the physical-space locality of interactions in three-dimensional incompressible turbulent flow. To that, we modify the nonlinear terms of the vorticity equation such that the vorticity field is advected and stretched by the locally induced velocity. This space-local velocity field is defined by the truncated Biot--Savart law, where only the neighboring vorticity field in a sphere of…
▽ More
We investigate the physical-space locality of interactions in three-dimensional incompressible turbulent flow. To that, we modify the nonlinear terms of the vorticity equation such that the vorticity field is advected and stretched by the locally induced velocity. This space-local velocity field is defined by the truncated Biot--Savart law, where only the neighboring vorticity field in a sphere of radius $R$ is integrated. We conduct direct numerical simulations of the space-local system to investigate its statistics in the inertial range. We observe a standard $E(k) \propto k^{-5/3}$ scaling of the energy spectrum associated with an energy cascade for scales smaller than the space-local domain size $k \gg R^{-1}$. This result is consistent with the assumption Kolmogorov's 1941 paper made for the space-locality of the nonlinear interactions. The enstrophy amplification is suppressed for larger scales $k \ll R^{-1}$, and for these scales, the system exhibits a scaling consistent with a conservative enstrophy cascade.
△ Less
Submitted 22 November, 2023; v1 submitted 14 August, 2023;
originally announced August 2023.
-
A supervised deep learning method for nonparametric density estimation
Authors:
Thijs Bos,
Johannes Schmidt-Hieber
Abstract:
Nonparametric density estimation is an unsupervised learning problem. In this work we propose a two-step procedure that casts the density estimation problem in the first step into a supervised regression problem. The advantage is that we can afterwards apply supervised learning methods. Compared to the standard nonparametric regression setting, the proposed procedure creates, however, dependence a…
▽ More
Nonparametric density estimation is an unsupervised learning problem. In this work we propose a two-step procedure that casts the density estimation problem in the first step into a supervised regression problem. The advantage is that we can afterwards apply supervised learning methods. Compared to the standard nonparametric regression setting, the proposed procedure creates, however, dependence among the training samples. To derive statistical risk bounds, one can therefore not rely on the well-developed theory for i.i.d. data. To overcome this, we prove an oracle inequality for this specific form of data dependence. As an application, it is shown that under a compositional structure assumption on the underlying density, the proposed two-step method achieves convergence rates that are faster than the standard nonparametric rates. A simulation study illustrates the finite sample performance.
△ Less
Submitted 3 June, 2024; v1 submitted 18 June, 2023;
originally announced June 2023.
-
Inertial range scaling of inhomogeneous turbulence
Authors:
Ryo Araki,
Wouter J. T. Bos
Abstract:
We investigate how inhomogeneity influences the $k^{-5/3}$ inertial range scaling of turbulent kinetic energy spectra (with $k$ the wavenumber). For weak statistical inhomogeneity, the energy spectrum can be described as an equilibrium spectrum plus a perturbation. Theoretical arguments suggest that this latter contribution scales as $k^{-7/3}$. This prediction is assessed using direct numerical s…
▽ More
We investigate how inhomogeneity influences the $k^{-5/3}$ inertial range scaling of turbulent kinetic energy spectra (with $k$ the wavenumber). For weak statistical inhomogeneity, the energy spectrum can be described as an equilibrium spectrum plus a perturbation. Theoretical arguments suggest that this latter contribution scales as $k^{-7/3}$. This prediction is assessed using direct numerical simulations of three-dimensional Kolmogorov flow.
△ Less
Submitted 26 July, 2023; v1 submitted 26 October, 2022;
originally announced October 2022.
-
Transport, flow topology and Lagrangian conditional statistics in edge plasma turbulence
Authors:
Benjamin Kadoch,
Diego del-Castillo-Negrete,
Wouter J. T. Bos,
Kai Schneider
Abstract:
Lagrangian statistics and particle transport in edge plasma turbulence are investigated using the Hasegawa-Wakatani model and its modified version. The latter shows the emergence of pronounced zonal flows. Different values of the adiabaticity parameter are considered. The main goal is to characterize the role of coherent structures, i.e., vortices and zonal flows, and their impact on the Lagrangia…
▽ More
Lagrangian statistics and particle transport in edge plasma turbulence are investigated using the Hasegawa-Wakatani model and its modified version. The latter shows the emergence of pronounced zonal flows. Different values of the adiabaticity parameter are considered. The main goal is to characterize the role of coherent structures, i.e., vortices and zonal flows, and their impact on the Lagrangian statistics of particles. Computationally intensive long time simulations following ensembles of test particles over hundreds of eddy turnover times are considered in statistically stationary turbulent flows. The flow topology is characterized using the Lagrangian Okubo-Weiss criterion, and the flow can thus be split into topologically different domains. In elliptic and hyperbolic regions, the probability density functions (pdfs) of the residence time have self-similar algebraic decaying tails. However, in the intermediate regions the pdfs do exhibit exponentially decaying tails. Topologically conditioned pdfs of the Lagrangian velocity, and acceleration and density fluctuations are likewise computed. The differences between the classical Hasegawa-Wakatani system and its modified version are assessed and the role of zonal flows is highlighted. The density flux spectrum which characterizes the contributions of different length scales is studied and its inertial scaling is found to be in agreement with predictions based on dimensional arguments. Analyzing the angular change of particle tracers at different time scales, corresponding to coarse grained curvature, completes the study and the multiscale geometric statistics quantify the directional properties of the particle motion in the different flow regimes.
△ Less
Submitted 14 May, 2022;
originally announced May 2022.
-
LaMDA: Language Models for Dialog Applications
Authors:
Romal Thoppilan,
Daniel De Freitas,
Jamie Hall,
Noam Shazeer,
Apoorv Kulshreshtha,
Heng-Tze Cheng,
Alicia **,
Taylor Bos,
Leslie Baker,
Yu Du,
YaGuang Li,
Hongrae Lee,
Huaixiu Steven Zheng,
Amin Ghafouri,
Marcelo Menegali,
Yan** Huang,
Maxim Krikun,
Dmitry Lepikhin,
James Qin,
Dehao Chen,
Yuanzhong Xu,
Zhifeng Chen,
Adam Roberts,
Maarten Bosma,
Vincent Zhao
, et al. (35 additional authors not shown)
Abstract:
We present LaMDA: Language Models for Dialog Applications. LaMDA is a family of Transformer-based neural language models specialized for dialog, which have up to 137B parameters and are pre-trained on 1.56T words of public dialog data and web text. While model scaling alone can improve quality, it shows less improvements on safety and factual grounding. We demonstrate that fine-tuning with annotat…
▽ More
We present LaMDA: Language Models for Dialog Applications. LaMDA is a family of Transformer-based neural language models specialized for dialog, which have up to 137B parameters and are pre-trained on 1.56T words of public dialog data and web text. While model scaling alone can improve quality, it shows less improvements on safety and factual grounding. We demonstrate that fine-tuning with annotated data and enabling the model to consult external knowledge sources can lead to significant improvements towards the two key challenges of safety and factual grounding. The first challenge, safety, involves ensuring that the model's responses are consistent with a set of human values, such as preventing harmful suggestions and unfair bias. We quantify safety using a metric based on an illustrative set of human values, and we find that filtering candidate responses using a LaMDA classifier fine-tuned with a small amount of crowdworker-annotated data offers a promising approach to improving model safety. The second challenge, factual grounding, involves enabling the model to consult external knowledge sources, such as an information retrieval system, a language translator, and a calculator. We quantify factuality using a groundedness metric, and we find that our approach enables the model to generate responses grounded in known sources, rather than responses that merely sound plausible. Finally, we explore the use of LaMDA in the domains of education and content recommendations, and analyze their helpfulness and role consistency.
△ Less
Submitted 10 February, 2022; v1 submitted 20 January, 2022;
originally announced January 2022.
-
Minimal model of quasi-cyclic behaviour in turbulence driven by Taylor--Green forcing
Authors:
Ryo Araki,
Wouter J. T. Bos,
Susumu Goto
Abstract:
We attempt to formulate the simplest possible model mimicking turbulent dynamics, such as quasi-cyclic behaviour (QCB), using only three variables. To this end, we first conduct direct numerical simulations of three-dimensional flow driven by the steady Taylor--Green forcing to find a similarity between a stable periodic orbit (SPO) at a small Reynolds number ($Re$) and turbulent QCB at higher…
▽ More
We attempt to formulate the simplest possible model mimicking turbulent dynamics, such as quasi-cyclic behaviour (QCB), using only three variables. To this end, we first conduct direct numerical simulations of three-dimensional flow driven by the steady Taylor--Green forcing to find a similarity between a stable periodic orbit (SPO) at a small Reynolds number ($Re$) and turbulent QCB at higher $Re$. A close examination of the SPO allows the heuristic formulation of a three-equation model, representing the evolution of Fourier modes in three distinct scales. The model reproduces the continuous bifurcation from SPO to turbulence with QCB when $Re$ is varied. We also demonstrate that, by changing model parameters, the proposed model exhibits a discontinuous transition from steady to chaotic solutions without going through an SPO.
△ Less
Submitted 11 June, 2023; v1 submitted 6 December, 2021;
originally announced December 2021.
-
The effect of sha** on turbulent dynamics in RFP simulations
Authors:
Robert Chahine,
Kai Schneider,
Wouter J. T. Bos
Abstract:
We study the influence of the shape of the plasma container on the dynamics of the Reversed Field Pinch (RFP). The geometries we consider are periodic cylinders with elliptical and circular-shaped cross-sections. Numerical simulations of fully nonlinear visco-resistive magnetohydrodynamics are carried out to illustrate how the plasma dynamics are affected by sha**. It is shown that independent o…
▽ More
We study the influence of the shape of the plasma container on the dynamics of the Reversed Field Pinch (RFP). The geometries we consider are periodic cylinders with elliptical and circular-shaped cross-sections. Numerical simulations of fully nonlinear visco-resistive magnetohydrodynamics are carried out to illustrate how the plasma dynamics are affected by sha**. It is shown that independent of the plasma shape, the quantity $β$, comparing the hydrodynamic pressure to the magnetic pressure, decreases for increasing values of the Lundquist number, but the pressure gradient fluctuations remain roughly constant, when compared to the Lorentz force. Different elliptical shapes of the cross-section of the domain lead to the excitation of different toroidal (or axial) magnetic and dynamic modes. Furthermore, it is found that in a geometry with circular cross-section a significant local poloidal angular momentum is observed, absent in the geometries with elliptical cross-section. Since the confinement is dominantly determined by plasma movement, and the dynamics of the velocity and magnetic field are modified by the modification of the geometry, sha** can thus affect the performance of RFP-devices.
△ Less
Submitted 16 November, 2021;
originally announced November 2021.
-
Convergence rates of deep ReLU networks for multiclass classification
Authors:
Thijs Bos,
Johannes Schmidt-Hieber
Abstract:
For classification problems, trained deep neural networks return probabilities of class memberships. In this work we study convergence of the learned probabilities to the true conditional class probabilities. More specifically we consider sparse deep ReLU network reconstructions minimizing cross-entropy loss in the multiclass classification setup. Interesting phenomena occur when the class members…
▽ More
For classification problems, trained deep neural networks return probabilities of class memberships. In this work we study convergence of the learned probabilities to the true conditional class probabilities. More specifically we consider sparse deep ReLU network reconstructions minimizing cross-entropy loss in the multiclass classification setup. Interesting phenomena occur when the class membership probabilities are close to zero. Convergence rates are derived that depend on the near-zero behaviour via a margin-type condition.
△ Less
Submitted 2 August, 2021;
originally announced August 2021.
-
Scaling laws in axisymmetric magnetohydrodynamic duct flows
Authors:
A. Poyé,
O. Agullo,
N. Plihon,
W. J. T. Bos,
V. Desangles,
G. Bousselin
Abstract:
We report on a numerical study of axisymmetric flow of liquid metal in a circular duct with rectangular cross-section. The flow is forced through the combination of an axial magnetic field and a radial current. Swee** a wide range of forcing parameters, we identify the different regimes which characterize the flows and explicit the associate scaling laws. Experimental results are interpreted in…
▽ More
We report on a numerical study of axisymmetric flow of liquid metal in a circular duct with rectangular cross-section. The flow is forced through the combination of an axial magnetic field and a radial current. Swee** a wide range of forcing parameters, we identify the different regimes which characterize the flows and explicit the associate scaling laws. Experimental results are interpreted in the light of our numerical simulations.
△ Less
Submitted 23 October, 2019;
originally announced October 2019.
-
Turbulence and turbulent pattern formation in a minimal model for active fluids
Authors:
Martin James,
Wouter J. T. Bos,
Michael Wilczek
Abstract:
Active matter systems display a fascinating range of dynamical states, including stationary patterns and turbulent phases. While the former can be tackled with methods from the field of pattern formation, the spatio-temporal disorder of the active turbulence phase calls for a statistical description. Borrowing techniques from turbulence theory, we here establish a quantitative description of corre…
▽ More
Active matter systems display a fascinating range of dynamical states, including stationary patterns and turbulent phases. While the former can be tackled with methods from the field of pattern formation, the spatio-temporal disorder of the active turbulence phase calls for a statistical description. Borrowing techniques from turbulence theory, we here establish a quantitative description of correlation functions and spectra of a minimal continuum model for active turbulence. Further exploring the parameter space, we also report on a surprising type of turbulence-driven pattern formation far beyond linear onset: the emergence of a dynamic hexagonal vortex lattice state after an extended turbulent transient, which can only be explained taking into account turbulent energy transfer across scales.
△ Less
Submitted 10 August, 2018; v1 submitted 29 November, 2017;
originally announced November 2017.
-
Short-time evolution of Lagrangian velocity gradient correlations in isotropic turbulence
Authors:
Le Fang,
W. J. T. Bos,
G. D. **
Abstract:
We show by direct numerical simulation (DNS) that the Lagrangian cross correlation of velocity gradients in homogeneous isotropic turbulence increases at short times, whereas its auto-correlation decreases. Kinematic considerations allow to show that two invariants of the turbulent velocity field determine the short-time velocity gradient correlations. In order to get a more intuitive understandin…
▽ More
We show by direct numerical simulation (DNS) that the Lagrangian cross correlation of velocity gradients in homogeneous isotropic turbulence increases at short times, whereas its auto-correlation decreases. Kinematic considerations allow to show that two invariants of the turbulent velocity field determine the short-time velocity gradient correlations. In order to get a more intuitive understanding of the dynamics for longer times, heuristic models are proposed involving the combined action of local shear and rotation. These models quantitatively reproduce the effects and disentangle the different physical mechanisms leading to the observations in the DNS.
△ Less
Submitted 20 November, 2015;
originally announced November 2015.
-
Zonal flow generation and its feedback on turbulence production in drift wave turbulence
Authors:
Andrey V. Pushkarev,
Wouter J. T. Bos,
Sergey V. Nazarenko
Abstract:
Plasma turbulence described by the Hasegawa-Wakatani equations has been simulated numerically for different models and values of the adiabaticity parameter C. It is found that for low values of C turbulence remains isotropic, zonal flows are not generated and there is no suppression of the meridional drift waves and of the particle transport. For high values of C, turbulence evolves toward highly…
▽ More
Plasma turbulence described by the Hasegawa-Wakatani equations has been simulated numerically for different models and values of the adiabaticity parameter C. It is found that for low values of C turbulence remains isotropic, zonal flows are not generated and there is no suppression of the meridional drift waves and of the particle transport. For high values of C, turbulence evolves toward highly anisotropic states with a dominant contribution of the zonal sector to the kinetic energy. This anisotropic flow leads to a decrease of a turbulence production in the meridional sector and limits the particle transport across the mean isopycnal surfaces. This behavior allows to consider the Hasegawa-Wakatani equations a minimal PDE model which contains the drift-wave/zonal-flow feedback loop prototypical of the LH transition in plasma devices.
△ Less
Submitted 21 December, 2012;
originally announced December 2012.
-
Tevatron Beam Position Monitor Upgrade
Authors:
G. Annala,
B. Banerjee,
B. Barker,
T. Boes,
M. Bowden,
C. Briegel,
G. Cancelo,
G. Duerling,
B. Forster,
S. Foulkes,
B. Haynes,
B. Hendricks,
T. Kasza,
R. Kutschke,
R. Mahlum,
M. Martens,
M. Olson,
V. Pavlicek,
L. Piccoli,
P. Prieto,
J. Steimel,
K. Treptow,
M. Votava,
D. Voy,
M. Wendt
, et al. (2 additional authors not shown)
Abstract:
This paper describes the development of a digital-based Beam Position System which was designed, developed, and adapted for the Tevatron during Collider Run II.
This paper describes the development of a digital-based Beam Position System which was designed, developed, and adapted for the Tevatron during Collider Run II.
△ Less
Submitted 26 September, 2012;
originally announced September 2012.
-
The role of coherent vorticity in turbulent transport in resistive drift-wave turbulence
Authors:
Wouter J. T. Bos,
Shinpei Futatani,
Sadruddin Benkadda,
Marie Farge,
Kai Schneider
Abstract:
The coherent vortex extraction method, a wavelet technique for extracting coherent vortices out of turbulent flows, is applied to simulations of resistive drift-wave turbulence in magnetized plasma (Hasegawa-Wakatani system). The aim is to retain only the essential degrees of freedom, responsible for the transport. It is shown that the radial density flux is carried by these coherent modes. In the…
▽ More
The coherent vortex extraction method, a wavelet technique for extracting coherent vortices out of turbulent flows, is applied to simulations of resistive drift-wave turbulence in magnetized plasma (Hasegawa-Wakatani system). The aim is to retain only the essential degrees of freedom, responsible for the transport. It is shown that the radial density flux is carried by these coherent modes. In the quasi-hydrodynamic regime, coherent vortices exhibit depletion of the polarization-drift nonlinearity and vorticity strongly dominates strain, in contrast to the quasiadiabatic regime.
△ Less
Submitted 29 November, 2011;
originally announced November 2011.
-
Quasi-static magnetohydrodynamic turbulence at high Reynolds number
Authors:
B. F. N. Favier,
F. S. Godeferd,
C. Cambon,
A. Delache,
W. J. T. Bos
Abstract:
We analyse the anisotropy of homogeneous turbulence in an electrically conducting fluid submitted to a uniform magnetic field, for low magnetic Reynolds number, in the quasi- static approximation. We interpret disagreeing previous predictions between linearized theory and simulations: in the linear limit, the kinetic energy of transverse velocity components, normal to the magnetic field, decays fa…
▽ More
We analyse the anisotropy of homogeneous turbulence in an electrically conducting fluid submitted to a uniform magnetic field, for low magnetic Reynolds number, in the quasi- static approximation. We interpret disagreeing previous predictions between linearized theory and simulations: in the linear limit, the kinetic energy of transverse velocity components, normal to the magnetic field, decays faster than the kinetic energy of the axial component, along the magnetic field (Moffatt (1967)); whereas many numerical studies predict a final state characterised by dominant energy of transverse velocity components. We investigate the corresponding nonlinear phenomenon using Direct Numerical Simulations of freely-decaying turbulence, and a two-point statistical spectral closure based on the Eddy Damped Quasi-Normal Markovian model. The transition from the three-dimensional turbulent flow to a "two-and-a-half-dimensional" flow (Montgomery & Turner (1982)) is a result of the combined effects of short-time linear Joule dissipation and longer time nonlinear creation of polarisation anisotropy. It is this combination of linear and nonlinear effects which explains the disagreement between predictions from linearized theory and results from numerical simulations. The transition is characterized by the elongation of turbulent structures along the applied magnetic field, and by the strong anisotropy of directional two-point correlation spectra, in agreement with experimental evidence. Inertial equatorial transfers in both DNS and the model are presented to describe in detail the most important equilibrium dynamics. Spectral scalings are maintained in high Reynolds number turbulence attainable only with the EDQNM model, which also provides simplified modelling of the asymptotic state of quasi-static MHD turbulence.
△ Less
Submitted 18 July, 2011;
originally announced July 2011.
-
Reduction of mean-square advection in turbulent passive scalar mixing
Authors:
Wouter J. T. Bos,
Robert Rubinstein,
Le Fang
Abstract:
Direct numerical simulation data show that the variance of the coupling term in passive scalar advection by a random velocity field is smaller than it would be if the velocity and scalar fields were statistically independent. This effect is analogous to the "depression of nonlinearity" in hydrodynamic turbulence. We show that the trends observed in the numerical data are qualitatively consistent w…
▽ More
Direct numerical simulation data show that the variance of the coupling term in passive scalar advection by a random velocity field is smaller than it would be if the velocity and scalar fields were statistically independent. This effect is analogous to the "depression of nonlinearity" in hydrodynamic turbulence. We show that the trends observed in the numerical data are qualitatively consistent with the predictions of closure theories related to Kraichnan's direct interaction approximation. The phenomenon is demonstrated over a range of Prandtl numbers. In the inertial-convective range the depletion is approximately constant with respect to wavenumber. The effect is weaker in the Batchelor range.
△ Less
Submitted 20 July, 2012; v1 submitted 5 July, 2011;
originally announced July 2011.
-
Reynolds number effect on the velocity increment skewness in isotropic turbulence
Authors:
Wouter J. T. Bos,
Laurent Chevillard,
Julian F. Scott,
Robert Rubinstein
Abstract:
Second and third order longitudinal structure functions and wavenumber spectra of isotropic turbulence are computed using the EDQNM model and compared to results of the multifractal formalism. At the highest Reynolds number available in windtunnel experiments, $R_λ=2500$, both the multifractal model and EDQNM give power-law corrections to the inertial range scaling of the velocity increment skewne…
▽ More
Second and third order longitudinal structure functions and wavenumber spectra of isotropic turbulence are computed using the EDQNM model and compared to results of the multifractal formalism. At the highest Reynolds number available in windtunnel experiments, $R_λ=2500$, both the multifractal model and EDQNM give power-law corrections to the inertial range scaling of the velocity increment skewness. For EDQNM, this correction is a finite Reynolds number effect, whereas for the multifractal formalism it is an intermittency correction that persists at any high Reynolds number. Furthermore, the two approaches yield realistic behavior of second and third order statistics of the velocity fluctuations in the dissipative and near-dissipative ranges. Similarities and differences are highlighted, in particular the Reynolds number dependence.
△ Less
Submitted 25 January, 2011;
originally announced January 2011.
-
Lagrangian statistics and flow topology in forced two-dimensional turbulence
Authors:
B. Kadoch,
D. del-Castillo-Negrete,
W. J. T. Bos,
K. Schneider
Abstract:
A study of the relationship between Lagrangian statistics and flow topology in fluid turbulence is presented. The topology is characterized using the Weiss criterion that provides a simplified tool to partition the flow into topologically different regions: elliptic (vortex dominated), hyperbolic (deformation dominated), and intermediate (turbulent background). The flow corresponds to forced two-d…
▽ More
A study of the relationship between Lagrangian statistics and flow topology in fluid turbulence is presented. The topology is characterized using the Weiss criterion that provides a simplified tool to partition the flow into topologically different regions: elliptic (vortex dominated), hyperbolic (deformation dominated), and intermediate (turbulent background). The flow corresponds to forced two-dimensional Navier-Stokes turbulence in doubly periodic and circular bounded domains with non-slip boundary conditions. In the double periodic domain, the probability density function (pdf) of the Weiss field exhibits a negative skewness consistent with the fact that in periodic domains the flow is dominated by coherent vortex structures. On the other hand, in the circular domain, the elliptic and hyperbolic regions seem to be statistically similar. We follow a Lagrangian approach and obtain the statistics by tracking large ensembles of passively advected tracers. The pdfs of residence time in the topologically different regions are computed using the Lagrangian Weiss field, i.e., the Weiss field computed along the particles' trajectories. In elliptic and hyperbolic regions, the pdfs of the residence time have self-similar algebraic decaying tails. On the other hand, in the intermediate regions the pdf has exponential decaying tails. The conditional (with respect to the flow topology) pdfs of the Lagrangian velocity exhibit Gaussian behavior in the periodic and in the bounded domains. In contrast to the freely decaying turbulence case, the conditional pdfs of the Lagrangian acceleration in forced turbulence show a comparable level of intermittency in the periodic and the bounded domains. The conditional pdfs of the Lagrangian curvature are characterized, in all cases, by self-similar power law behavior with a decay exponent of order -2.
△ Less
Submitted 14 September, 2010;
originally announced September 2010.
-
Rapid generation of angular momentum in bounded magnetized plasma
Authors:
Wouter J. T. Bos,
Salah Neffaa,
Kai Schneider
Abstract:
Direct numerical simulations of two-dimensional decaying MHD turbulence in bounded domains show the rapid generation of angular momentum in nonaxisymmetric geometries. It is found that magnetic fluctuations enhance this mechanism. On a larger time scale, the generation of a magnetic angular momentum, or angular field, is observed. For axisymmetric geometries, the generation of angular momentum is…
▽ More
Direct numerical simulations of two-dimensional decaying MHD turbulence in bounded domains show the rapid generation of angular momentum in nonaxisymmetric geometries. It is found that magnetic fluctuations enhance this mechanism. On a larger time scale, the generation of a magnetic angular momentum, or angular field, is observed. For axisymmetric geometries, the generation of angular momentum is absent; nevertheless, a weak magnetic field can be observed. The derived evolution equations for both the angular momentum and angular field yield possible explanations for the observed behavior.
△ Less
Submitted 1 September, 2010;
originally announced September 2010.
-
On the unsteady behavior of turbulence models
Authors:
Robert Rubinstein,
Wouter J. T. Bos
Abstract:
Periodically forced turbulence is used as a test case to evaluate the predictions of two-equation and multiple-scale turbulence models in unsteady flows. The limitations of the two-equation model are shown to originate in the basic assumption of spectral equilibrium. A multiple-scale model based on a picture of stepwise energy cascade overcomes some of these limitations, but the absence of nonloca…
▽ More
Periodically forced turbulence is used as a test case to evaluate the predictions of two-equation and multiple-scale turbulence models in unsteady flows. The limitations of the two-equation model are shown to originate in the basic assumption of spectral equilibrium. A multiple-scale model based on a picture of stepwise energy cascade overcomes some of these limitations, but the absence of nonlocal interactions proves to lead to poor predictions of the time variation of the dissipation rate. A new multiple-scale model that includes nonlocal interactions is proposed and shown to reproduce the main features of the frequency response correctly.
△ Less
Submitted 1 September, 2010;
originally announced September 2010.
-
Large-scale bottleneck effect in two-dimensional turbulence
Authors:
Wouter J. T. Bos,
Jean-Pierre Bertoglio
Abstract:
The bottleneck phenomenon in three-dimensional turbulence is generally associated with the dissipation range of the energy spectrum. In the present work, it is shown by using a two-point closure theory, that in two-dimensional turbulence it is possible to observe a bottleneck at the large scales, due to the effect of friction on the inverse energy cascade. This large-scale bottleneck is directly r…
▽ More
The bottleneck phenomenon in three-dimensional turbulence is generally associated with the dissipation range of the energy spectrum. In the present work, it is shown by using a two-point closure theory, that in two-dimensional turbulence it is possible to observe a bottleneck at the large scales, due to the effect of friction on the inverse energy cascade. This large-scale bottleneck is directly related to the process of energy condensation, the pile-up of energy at wavenumbers corresponding to the domain size. The link between the use of friction and the creation of space-filling structures is discussed and it is concluded that the careless use of hypofriction might reduce the inertial range of the energy spectrum.
△ Less
Submitted 1 September, 2010;
originally announced September 2010.
-
Develo** Homogeneous Isotropic Turbulence
Authors:
Wouter J. T. Bos,
Colm Connaughton,
Fabien Godeferd
Abstract:
We investigate the self-similar evolution of the transient energy spectrum which precedes the establishment of the Kolmogorov spectrum in homogeneous isotropic turbulence in three dimensions using the EDQNM closure model. The transient evolution exhibits self-similarity of the second kind and has a non-trivial dynamical scaling exponent which results in the transient spectrum having a scaling whic…
▽ More
We investigate the self-similar evolution of the transient energy spectrum which precedes the establishment of the Kolmogorov spectrum in homogeneous isotropic turbulence in three dimensions using the EDQNM closure model. The transient evolution exhibits self-similarity of the second kind and has a non-trivial dynamical scaling exponent which results in the transient spectrum having a scaling which is steeper than the Kolmogorov $k^{-5/3}$ spectrum. Attempts to detect a similar phenomenon in DNS data are inconclusive owing to the limited range of scales available.
△ Less
Submitted 6 November, 2010; v1 submitted 4 June, 2010;
originally announced June 2010.
-
Inertial range scaling of scalar flux spectra in uniformly sheared turbulence
Authors:
Wouter J. T. Bos,
Jean-Pierre Bertoglio
Abstract:
A model based on two-point closure theory of turbulence is proposed and applied to study the Reynolds number dependency of the scalar flux spectra in homogeneous shear flow with a cross-stream uniform scalar gradient. For the cross-stream scalar flux, in the inertial range the spectral behavior agrees with classical predictions and measurements. The streamwise scalar flux is found to be in good…
▽ More
A model based on two-point closure theory of turbulence is proposed and applied to study the Reynolds number dependency of the scalar flux spectra in homogeneous shear flow with a cross-stream uniform scalar gradient. For the cross-stream scalar flux, in the inertial range the spectral behavior agrees with classical predictions and measurements. The streamwise scalar flux is found to be in good agreement with the results of atmospheric measurements. However, both the model results and the atmospheric measurements disagree with classical predictions. A detailed analysis of the different terms in the evolution equation for the streamwise scalar flux spectrum shows that nonlinear contributions are governing the inertial subrange of this spectrum and that these contributions are relatively more important than for the cross-stream flux. A new expression for the scalar flux spectra is proposed. It allows us to unify the description of the components in one single expression, leading to a classical K^-7/3 inertial range for the cross-stream component and to a new K^-23/9 scaling for the streamwise component that agrees better with atmospheric measurements than the K^-3 prediction of J. C. Wyngaard and O. R. Coté [Quart. J. R. Met. Soc. 98, 590 (1972)].
△ Less
Submitted 18 December, 2007;
originally announced December 2007.
-
Dynamics of spectrally truncated inviscid turbulence
Authors:
Wouter J. T. Bos,
Jean-Pierre Bertoglio
Abstract:
The evolution of the turbulent energy spectrum for the inviscid spectrally truncated Euler equations is studied by closure calculations. The observed behavior is similar to the one found in direct numerical simulations [Cichowlas, Bonaïtiti, Debbasch, and Brachet, Phys. Rev. Lett. 95, 264502 (2005)]. A Kolmogorov spectral range and an equipartition range are observed simultaneously. Between thes…
▽ More
The evolution of the turbulent energy spectrum for the inviscid spectrally truncated Euler equations is studied by closure calculations. The observed behavior is similar to the one found in direct numerical simulations [Cichowlas, Bonaïtiti, Debbasch, and Brachet, Phys. Rev. Lett. 95, 264502 (2005)]. A Kolmogorov spectral range and an equipartition range are observed simultaneously. Between these two ranges a "quasi-dissipative" zone is present in the kinetic energy spectrum. The time evolution of the wave number that marks the beginning of the equipartition range is analyzed and it is shown that spectral nonlocal interactions are governing this evolution.
△ Less
Submitted 17 December, 2007;
originally announced December 2007.
-
Spectral imbalance and the normalized dissipation rate of turbulence
Authors:
Wouter J. T. Bos,
L. Shao,
Jean-Pierre Bertoglio
Abstract:
The normalized turbulent dissipation rate $C_ε$ is studied in decaying and forced turbulence by direct numerical simulations, large-eddy simulations, and closure calculations. A large difference in the values of $C_ε$ is observed for the two types of turbulence. This difference is found at moderate Reynolds number, and it is shown that it persists at high Reynolds number, where the value of…
▽ More
The normalized turbulent dissipation rate $C_ε$ is studied in decaying and forced turbulence by direct numerical simulations, large-eddy simulations, and closure calculations. A large difference in the values of $C_ε$ is observed for the two types of turbulence. This difference is found at moderate Reynolds number, and it is shown that it persists at high Reynolds number, where the value of $C_ε$ becomes independent of the Reynolds number, but is still not unique. This difference can be explained by the influence of the nonlinear cascade time that introduces a spectral disequilibrium for statistically nonstationary turbulence. Phenomenological analysis yields simple analytical models that satisfactorily reproduce the numerical results. These simple spectral models also reproduce and explain the increase of $C_ε$ at low Reynolds number that is observed in the simulations.
△ Less
Submitted 17 December, 2007;
originally announced December 2007.
