2006
DOI: 10.1002/fld.1196
|View full text |Cite
|
Sign up to set email alerts
|

Continuation of travelling-wave solutions of the Navier–Stokes equations

Abstract: SUMMARYAn e cient way of obtaining travelling waves in a periodic uid system is described and tested. We search for steady states in a reference frame travelling at the wave phase velocity using a ÿrst-order pseudospectral semi-implicit time scheme adapted to carry out the Newton's iterations. The method is compared to a standard Newton-Raphson solver and is shown to be highly e cient in performing this task, even when high-resolution grids are used. This method is well suited to three-dimensional calculations… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
11
0

Year Published

2008
2008
2024
2024

Publication Types

Select...
10

Relationship

3
7

Authors

Journals

citations
Cited by 21 publications
(12 citation statements)
references
References 24 publications
0
11
0
Order By: Relevance
“…3, 21 In a companion paper, 14 we have presented and discussed a set of direct numerical simulations ͑DNSs͒ which exhibit in detail a transition to baroclinic chaos via a lowdimensional route involving a succession of modulated traveling waves in an air-filled, rotating annulus. Such studies complement the use of analytical and numerical stability and continuation methods in the context of circularly symmetric systems such as the rotating annulus, [22][23][24][25][26][27] though the latter approaches are not yet able to deal with the full complexity of time-dependent, modulated waves and their nonlinear interactions found in the present work. The use of air as the working fluid is of particular interest since this represents a fluid with Prandtl number Pr= O͑1͒ which has received relatively little attention in previous experimental studies 28 or in numerical simulations of baroclinic flows but, in common with other convective flows, the Prandtl number seems to play an important role in governing the nature of nonlinear interactions in the flow, as well as strongly affecting the form of the advective-diffusive basic state from which baroclinic instabilities grow.…”
Section: Introductionmentioning
confidence: 89%
“…3, 21 In a companion paper, 14 we have presented and discussed a set of direct numerical simulations ͑DNSs͒ which exhibit in detail a transition to baroclinic chaos via a lowdimensional route involving a succession of modulated traveling waves in an air-filled, rotating annulus. Such studies complement the use of analytical and numerical stability and continuation methods in the context of circularly symmetric systems such as the rotating annulus, [22][23][24][25][26][27] though the latter approaches are not yet able to deal with the full complexity of time-dependent, modulated waves and their nonlinear interactions found in the present work. The use of air as the working fluid is of particular interest since this represents a fluid with Prandtl number Pr= O͑1͒ which has received relatively little attention in previous experimental studies 28 or in numerical simulations of baroclinic flows but, in common with other convective flows, the Prandtl number seems to play an important role in governing the nature of nonlinear interactions in the flow, as well as strongly affecting the form of the advective-diffusive basic state from which baroclinic instabilities grow.…”
Section: Introductionmentioning
confidence: 89%
“…In the small t limit, P ≈ I and the equation is not preconditioned, while for sufficiently large t, we get the so-called Stokes preconditioner P ≈ t(γ D /γ t )L. This preconditioning method is easy to implement as it only requires a first order implicit Euler time-integration scheme and its use is natural within matrix-free methods, where the Jacobian is not explicitly constructed. The Stokes preconditioner has been widely used in problems that are dominated by diffusion, like coupled convection [47][48][49]. In contrast, most shear flow studies are carried out at large Reynolds numbers and involve weakly diffusive flows.…”
Section: -7mentioning
confidence: 99%
“…Notice that the reflection R can be obtained as R = R 1 •R 2 . The system of equations (2)-(3) and boundary conditions (4)- (5) has been solved numerically using the algorithm IPS described in [24], which can be summarized as follows. To integrate the equations in time, we use a second-order time-splitting method proposed in [25] combined with a pseudospectral method for the spatial discretization, Chebyshev collocation in x and z.…”
Section: Formulation Of the Problem: Equations Symmetries And Numentioning
confidence: 99%