betaFeedback

WorksMathematics

Enhanced dissipation and nonlinear asymptotic stability of the Taylor-Couette flow for the 2D Navier-Stokes equations

Abstract

In this paper, we study the nonlinear stability of a steady circular flow created between two rotating concentric cylinders. The dynamics of the viscous fluid are described by 2D Navier-Stokes equations. We adopt scaling variables. For the rescaled equations, we prove that the steady flow (Taylor-Couette flow) is asymptotically stable up to a large perturbation of initial data. Back to the original 2D Navier-Stokes equations, this implies an improved transition threshold for the Taylor-Couette flow. The improvement is due to enhanced dissipation and new observations and constructions of weighted L2L^2 norms, which capture a hidden structure between the viscosity constant νν and (different) rotating speeds and locations of two coaxial cylinders. In particular, we allow the location of the outer cylinder to tend to infinity, which renders the initial fluid kinetic energy not uniformly bounded. Due to enhanced-dissipation effect, we also establish a sharp resolvent estimate, desired space-time bounds and optimal decaying estimates, which lead to the proof of nonlinear asymptotic stability of 2D Taylor-Couette flow.

Introduction

In this paper, we consider the two-dimensional (2D) incompressible Navier-Stokes equations:

{∂tv−νΔv+v⋅∇v+∇p=0,div⁡v=0,v(0,x)=v0(x)(1)\begin{cases} \partial_t v-\nu\Delta v+v\cdot\nabla v+\nabla p=0,\quad\operatorname{div}v=0,\\ v(0,x)=v_0(x) \tag*{(1)} \end{cases}

with x=(x1,x2)∈Ω=R2x=(x_1,x_2)\in\Omega=\mathbb{R}^2 being the space variables and t≥0t\geq0 being the time variable. Here the constant ν>0\nu>0 is called the kinematic viscosity. The unknowns to (1.1) are the velocity field v(t,x)=(v1(t,x),v2(t,x))v(t,x)=(v^1(t,x),v^2(t,x)) and the pressure p(t,x)∈Rp(t,x)\in\mathbb{R}.

If taking the 2D curl to the above Navier-Stokes equations, then (1.1) is transferred to

∂tω−νΔω+v⋅∇ω=0.(2)\partial_t\omega-\nu\Delta\omega+v\cdot\nabla\omega=0. \tag*{(2)}

Here ω=curl⁡v=∂1v2−∂2v1\omega=\operatorname{curl}v=\partial_1v^2-\partial_2v^1 is the vorticity field. For 2D case, it is a scalar. And (1.2) is called the vorticity equation formulation. In this paper, we will use (1.2) to understand the asymptotics of solutions to (1.1).

Given ω\omega being a solution to (1.2), the velocity field v:R+×R2→R2v:\mathbb{R}_+\times\mathbb{R}^2\to\mathbb{R}^2 can be derived via solving the below elliptic system

div⁡v=0,curl⁡v=ω.\operatorname{div}v=0,\quad\operatorname{curl}v=\omega.

This leads to the 2D Biot-Savart law

v(t,x)=12π∫R2(x−y)⊥∣x−y∣2ω(t,y) dy.(3)v(t,x)=\frac{1}{2\pi}\int_{\mathbb{R}^2}\frac{(x-y)^\perp}{\lvert x-y\rvert^2}\omega(t,y)\,dy. \tag*{(3)}

For notational simplicity, we write v(t,⋅)=KBS∗ω(t,⋅)v(t,\cdot)=K_{BS}*\omega(t,\cdot), where KBS(x)=12πx⊥∣x∣2K_{BS}(x)=\frac{1}{2\pi}\frac{x^\perp}{\lvert x\rvert^2} is the Biot-Savart kernel.

Taylor-Couette flow

In this article, we focus on studying the nonlinear stability and long-time dynamics around the circular flows with the vorticity ω(x)=ω(∣x∣)\omega(x)=\omega(\lvert x\rvert) being radial. In below, we denote r~\tilde r to be ∣x∣\lvert x\rvert.

An important physical circular-flow solution to (1.2) is called the Taylor-Couette flow. It describes a steady (circular) flow of viscous fluid bounded between two rotating infinitely long coaxial cylinders and has wide applications ranging from desalination to viscometric analysis. Now we derive its expression.

Denote ω(x)=ω(r~)\omega(x)=\omega(\tilde r) and ϕ(x)=ϕ(r~)\phi(x)=\phi(\tilde r) to be the radial vorticity and the stream function, respectively. Then for 2D case we have

ω(r~)=Δϕ=ϕ′′(r~)+1r~ϕ′(r~),(4)\omega(\tilde r)=\Delta\phi=\phi''(\tilde r)+\frac{1}{\tilde r}\phi'(\tilde r), \tag*{(4)}
v(x,y)=(−∂2ϕ∂1ϕ)=∂r~ϕeθ~=(−x2x1)ϕ′(r~)r~.(5)v(x,y)= \begin{pmatrix} -\partial_2\phi\\ \partial_1\phi \end{pmatrix} =\partial_{\tilde r}\phi e_{\tilde\theta} =\begin{pmatrix} -x_2\\ x_1 \end{pmatrix}\frac{\phi'(\tilde r)}{\tilde r}. \tag*{(5)}

One can easily check ∇⋅v=0\nabla\cdot v=0 and via ∂1=cos⁡θ~∂r~−1r~sin⁡θ~∂θ~\partial_1=\cos\tilde\theta\partial_{\tilde r}-\frac{1}{\tilde r}\sin\tilde\theta\partial_{\tilde\theta}, ∂2=sin⁡θ~∂r~+1r~cos⁡θ~∂θ~\partial_2=\sin\tilde\theta\partial_{\tilde r}+\frac{1}{\tilde r}\cos\tilde\theta\partial_{\tilde\theta}, one also has

v⋅∇ω=−sin⁡θ~ϕ′∂1ω+cos⁡θ~ϕ′∂2ω=0.v\cdot\nabla\omega=-\sin\tilde\theta\phi'\partial_1\omega+\cos\tilde\theta\phi'\partial_2\omega=0.

Hence, a radial vorticity ω(r~)\omega(\tilde{r}) in (1.4) is a stationary solution to the below 2D incompressible Euler equation

∂tω+v⋅∇ω=0.(6)\partial_t\omega+v\cdot\nabla\omega=0. \tag*{(6)}

In particular, if we choose ω=const\omega=\mathrm{const}, then we have Δω=0\Delta\omega=0 and ω\omega is also a solution to the 2D incompressible Navier–Stokes equation (1.2). For this case, by (1.4) the stream function ϕ(r~)\phi(\tilde{r}) reads

ϕ′′(r~)+1r~ϕ′(r~)=2A1.(7)\phi^{\prime\prime}(\tilde{r})+\frac{1}{\tilde{r}}\phi^{\prime}(\tilde{r})=2A_1. \tag*{(7)}

where A1,A2A_1,A_2 are constants. This implies (r~ϕ′)′=2A1r~(\tilde{r}\phi')'=2A_1\tilde{r} and ϕ′r~=A1+A2r~2\frac{\phi'}{\tilde{r}}=A_1+\frac{A_2}{\tilde{r}^2}.

Via (1.7) we now derive a solution v(x,y)v(x,y) to (1.1). And with polar coordinate, we rename v(x,y)v(x,y) and ω(r)\omega(r) to be U(r~,θ~)U(\tilde{r},\tilde{\theta}) and Ω(r~)\Omega(\tilde{r}), respectively. Now we have

U(r~,θ~)=(U1U2)=(−sin⁡θ~cos⁡θ~)(A1r~+A2r~)=vθ~eθ~,Ω(r~)=2A1,(8)U(\tilde{r},\tilde{\theta})=\begin{pmatrix}U^1\\U^2\end{pmatrix}=\begin{pmatrix}-\sin\tilde{\theta}\\\cos\tilde{\theta}\end{pmatrix}\left(A_1\tilde{r}+\frac{A_2}{\tilde{r}}\right)=v_{\tilde{\theta}}e_{\tilde{\theta}},\qquad\Omega(\tilde{r})=2A_1, \tag*{(8)}

where A1,A2A_1,A_2 are constants, eθ~=(−sin⁡θ~cos⁡θ~)e_{\tilde{\theta}}=\begin{pmatrix}-\sin\tilde{\theta}\\\cos\tilde{\theta}\end{pmatrix}, and vθ~=A1r~+A2r~v_{\tilde{\theta}}=A_1\tilde{r}+\frac{A_2}{\tilde{r}} is the azimuthal velocity component. Note that U(r~,θ~)U(\tilde{r},\tilde{\theta}) in (1.8) is a stationary solution to (1.1) and it is called the Taylor-Couette flow.

Connection and comparison to the Lamb-Oseen vortex

It is well known that the 2D Navier–Stokes equations (1.2) have a family of self-similar solutions called the Lamb-Oseen vortices, which are of the following form

ω(t,x)=ανtG(xνt),v(t,x)=ανtvG(xνt).(9)\omega(t,x)=\frac{\alpha}{\nu t}G\left(\frac{x}{\sqrt{\nu t}}\right),\qquad v(t,x)=\frac{\alpha}{\sqrt{\nu t}}v^G\left(\frac{x}{\sqrt{\nu t}}\right). \tag*{(9)}

Here the vorticity profile and the velocity profile are given by

G(ξ)=14πe−∣ξ∣2/4,vG(ξ)=12πξ⊥∣ξ∣2(1−e−∣ξ∣2/4).(10)G(\xi)=\frac{1}{4\pi}e^{-|\xi|^2/4},\qquad v^G(\xi)=\frac{1}{2\pi}\frac{\xi^\perp}{|\xi|^2}\left(1-e^{-|\xi|^2/4}\right). \tag*{(10)}

In particular we have vG=KBS∗Gv^G=K_{BS}*G with KBSK_{BS} being the Biot-Savart kernel. And we further denote space

References

  1. [1]M. Beck and C. E.Wayne, Metastability and rapid convergence to quasi-stationary bar states for the two-dimensional Navier-Stokes equations, Proc. Roy. Soc. Edinburgh Sect. A 143 (2013), no. 5, 905-927.DOI
  2. [7]J. Bedrossian and M. Zelati, Enhanced dissipation, hypoellipticity, and anomalous small noise inviscid limits in shear flows, Arch. Ration. Mech. Anal. 224 (2017), no. 3, 1161-1204.arxiv.org/abs/1510.08098
  3. [10]Q. Chen, D. Wei and Z. Zhang, Linear inviscid damping and enhanced dissipation for monotone shear flows, preprint.
  4. [15]W. Deng, J. Wu and P. Zhang, Stability of Couette flow for 2D Boussinesq system with vertical dissipation, Journal of Functional Analysis, Volume 281, Issue 12 (2021).arxiv.org/abs/2004.09292
  5. [18]T. Gallay, Stability and interaction of vortices in two-dimensional viscous flows, Discr. Cont. Dyn. Systems Ser. S 5 (2012), 1091-1131.arxiv.org/abs/1203.0846
  6. [19]T. Gallay, Enhanced dissipation and axisymmetrization of two-dimensional viscous vortices, Archive for Rational Mechanics and Analysis, volume 230 (2018), 939-975.DOI
  7. [20]T. Gallay and L. M. Rodrigues, Sur le temps de vie de la turbulence bidimensionnelle (in French), Ann. Fac. Sci. Toulouse 17 (2008), 719-733.DOI
  8. [21]T. Gallay and V. Šverák, Arnold's variational principle and its application to the stability of planar vortices, arXiv:2110.13739.
  9. [22]T. Gallay and C. E. Wayne, Invariant manifolds and the long-time asymptotics of the Navier-Stokes and vorticity equations on R², Arch. Ration. Mech. Anal., 163 (2002), 209-258.DOI
  10. [23]T. Gallay and C. E. Wayne, Global stability of vortex solutions of the two-dimensional Navier-Stokes equation, Comm. Math. Phys., 255 (2005), 97-129.DOI
  11. [24]S. Ibrahim, Y. Maekawa, and N. Masmoudi, On pseudospectral bound for non-selfadjoint operators and its application to stability of Kolmogorov flows, Annals of PDE, 5(2019), 5:2.arxiv.org/abs/1710.05132
  12. [25]A. Ionescu and H. Jia, Nonlinear inviscid damping near monotonic shear flows, arXiv:2001.03087, to appear in Acta Mathematica.
  13. [27]T. Kato, Perturbation Theory for Linear Operators. Grundlehren der Mathematischen Wissenschaften 132. Berlin: Springer, 1966.
  14. [29]T. Li, D. Wei and Z. Zhang, Pseudospectral and spectral bounds for the Oseen vortices operator, Annales Scientifiques de Ecole Normale Supérieure, 53 (2020), 993-1035.
  15. [30]Z. Lin and M. Xu, Metastability of Kolmogorov flows and inviscid damping of shear flows, Arch. Rat. Mech. Anal., 231(2019), 1811-1852.DOI
  16. [34]A. Pazy, Semigroups of linear operators and applications to partial differential equations, Applied Mathematical Sciences 44, Springer-Verlag, New York, 1983.
  17. [39]D. Wei, Diffusion and mixing in fluid flow via the resolvent estimate, Science China Mathematics, 64 (2021), 507-518.
  18. [41]D. Wei, Z. Zhang, and W. Zhao, Linear inviscid damping for a class of monotone shear flow in Sobolev spaces, Comm. Pure Appl. Math., 71(2018), 617-687.
  19. [43]D. Wei, Z. Zhang and W. Zhao, Linear inviscid damping and enhanced dissipation for the Kolmogorov flow, Adv. Math., 362 (2020), 106963.arxiv.org/abs/1711.01822
  20. [45]M. Zelati, M. Delgadino and T. Elgindi, On the Relation between Enhanced Dissipation Timescales and Mixing Rates, Comm. Pure Appl. Math., 73(2020), 1205-1244.arxiv.org/abs/1806.03258
  21. [46]M. Zelati, C. Zillinger, On degenerate circular and shear flows: the point vortex and power law circular flows, Communications in Partial Differential Equations, 44 (2019), 110-155.arxiv.org/abs/1801.07371
  22. [47]C. Zillinger, On circular flows: linear stability and damping, J. Differential Equations, 263 (2017), no. 11, 7856-7899.arxiv.org/abs/1605.05959

Paper details

vlnc:W_mmgw3gsp
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 6

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents