-
Chaotic mixing in plane Couette turbulence
Authors:
John R. Elton,
Predrag Cvitanović,
Jonathan Halcrow,
John F. Gibson
Abstract:
Lagrangian tracer particle trajectories for invariant solutions of the Navier-Stokes equations confined to the three-dimensional geometry of plane Couette flow are studied. Treating the Eulerian velocity field of an invariant solution as a dynamical system, the transport of these passive scalars along Lagrangian flow trajectories reveals a rich repertoire of different types of motion that can occu…
▽ More
Lagrangian tracer particle trajectories for invariant solutions of the Navier-Stokes equations confined to the three-dimensional geometry of plane Couette flow are studied. Treating the Eulerian velocity field of an invariant solution as a dynamical system, the transport of these passive scalars along Lagrangian flow trajectories reveals a rich repertoire of different types of motion that can occur, including stagnation points, for which there is no fluid movement, and invariant tori, which obstruct chaotic mixing across the full volume of the plane Couette flow minimal cell. We determine the stability of these stagnation points, along with their stable and unstable manifolds, and find heteroclinic connections between them. These topological features produce a skeleton that shapes the passive tracer flow for a turbulent fluid, providing a first step to elucidating Lagrangian particle transport and mixing in three-dimensional Navier-Stokes turbulent flows.
△ Less
Submitted 5 December, 2024;
originally announced December 2024.
-
TF-GNN: Graph Neural Networks in TensorFlow
Authors:
Oleksandr Ferludin,
Arno Eigenwillig,
Martin Blais,
Dustin Zelle,
Jan Pfeifer,
Alvaro Sanchez-Gonzalez,
Wai Lok Sibon Li,
Sami Abu-El-Haija,
Peter Battaglia,
Neslihan Bulut,
Jonathan Halcrow,
Filipe Miguel Gonçalves de Almeida,
Pedro Gonnet,
Liangze Jiang,
Parth Kothari,
Silvio Lattanzi,
André Linhares,
Brandon Mayer,
Vahab Mirrokni,
John Palowitch,
Mihir Paradkar,
Jennifer She,
Anton Tsitsulin,
Kevin Villela,
Lisa Wang
, et al. (2 additional authors not shown)
Abstract:
TensorFlow-GNN (TF-GNN) is a scalable library for Graph Neural Networks in TensorFlow. It is designed from the bottom up to support the kinds of rich heterogeneous graph data that occurs in today's information ecosystems. In addition to enabling machine learning researchers and advanced developers, TF-GNN offers low-code solutions to empower the broader developer community in graph learning. Many…
▽ More
TensorFlow-GNN (TF-GNN) is a scalable library for Graph Neural Networks in TensorFlow. It is designed from the bottom up to support the kinds of rich heterogeneous graph data that occurs in today's information ecosystems. In addition to enabling machine learning researchers and advanced developers, TF-GNN offers low-code solutions to empower the broader developer community in graph learning. Many production models at Google use TF-GNN, and it has been recently released as an open source project. In this paper we describe the TF-GNN data model, its Keras message passing API, and relevant capabilities such as graph sampling and distributed training.
△ Less
Submitted 23 July, 2023; v1 submitted 7 July, 2022;
originally announced July 2022.
-
Equilibrium and traveling-wave solutions of plane Couette flow
Authors:
John F. Gibson,
Jonathan Halcrow,
Predrag Cvitanović
Abstract:
We present ten new equilibrium solutions to plane Couette flow in small periodic cells at low Reynolds number (Re) and two new traveling-wave solutions. The solutions are continued under changes of Re and spanwise period. We provide a partial classification of the isotropy groups of plane Couette flow and show which kinds of solutions are allowed by each isotropy group. We find two complementary…
▽ More
We present ten new equilibrium solutions to plane Couette flow in small periodic cells at low Reynolds number (Re) and two new traveling-wave solutions. The solutions are continued under changes of Re and spanwise period. We provide a partial classification of the isotropy groups of plane Couette flow and show which kinds of solutions are allowed by each isotropy group. We find two complementary visualizations particularly revealing. Suitably chosen sections of their 3D-physical space velocity fields are helpful in developing physical intuition about coherent structures observed in low Re turbulence. Projections of these solutions and their unstable manifolds from their infinite-dimensional state space onto suitably chosen 2- or 3-dimensional subspaces reveal their interrelations and the role they play in organizing turbulence in wall-bounded shear flows.
△ Less
Submitted 16 June, 2009; v1 submitted 25 August, 2008;
originally announced August 2008.
-
Heteroclinic connections in plane Couette flow
Authors:
Jonathan Halcrow,
John F. Gibson,
Predrag Cvitanović,
Divakar Viswanath
Abstract:
Plane Couette flow transitions to turbulence for Re~325 even though the laminar solution with a linear profile is linearly stable for all Re (Reynolds number). One starting point for understanding this subcritical transition is the existence of invariant sets in the state space of the Navier Stokes equation, such as upper and lower branch equilibria and periodic and relative periodic solutions,…
▽ More
Plane Couette flow transitions to turbulence for Re~325 even though the laminar solution with a linear profile is linearly stable for all Re (Reynolds number). One starting point for understanding this subcritical transition is the existence of invariant sets in the state space of the Navier Stokes equation, such as upper and lower branch equilibria and periodic and relative periodic solutions, that are quite distinct from the laminar solution. This article reports several heteroclinic connections between such objects and briefly describes a numerical method for locating heteroclinic connections. Computing such connections is essential for understanding the global dynamics of spatially localized structures that occur in transitional plane Couette flow. We show that the nature of streaks and streamwise rolls can change significantly along a heteroclinic connection.
△ Less
Submitted 13 August, 2008;
originally announced August 2008.
-
Visualizing the geometry of state space in plane Couette flow
Authors:
John F. Gibson,
Jonathan Halcrow,
Predrag Cvitanović
Abstract:
Motivated by recent experimental and numerical studies of coherent structures in wall-bounded shear flows, we initiate a systematic exploration of the hierarchy of unstable invariant solutions of the Navier-Stokes equations. We construct a dynamical, 10^5-dimensional state-space representation of plane Couette flow at Re = 400 in a small, periodic cell and offer a new method of visualizing invar…
▽ More
Motivated by recent experimental and numerical studies of coherent structures in wall-bounded shear flows, we initiate a systematic exploration of the hierarchy of unstable invariant solutions of the Navier-Stokes equations. We construct a dynamical, 10^5-dimensional state-space representation of plane Couette flow at Re = 400 in a small, periodic cell and offer a new method of visualizing invariant manifolds embedded in such high dimensions. We compute a new equilibrium solution of plane Couette flow and the leading eigenvalues and eigenfunctions of known equilibria at this Reynolds number and cell size. What emerges from global continuations of their unstable manifolds is a surprisingly elegant dynamical-systems visualization of moderate-Reynolds turbulence. The invariant manifolds tessellate the region of state space explored by transiently turbulent dynamics with a rigid web of continuous and discrete symmetry-induced heteroclinic connections.
△ Less
Submitted 11 February, 2008; v1 submitted 27 May, 2007;
originally announced May 2007.
