An iterative approach to a fluid-rigid body interaction problem

Charles M. Elliott ⋆, Thomas Sales † · 2026

Read An iterative approach to a fluid-rigid body interaction problem by Charles M. Elliott ⋆, Thomas Sales † (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2601-13004.txt.

Introduction

Charles M. Elliott and Thomas Sales study a hard mix of fluid flow and a stiff solid body. The fluid cannot be squeezed, and the action takes place in three space dims. Such pairs turn up in blood flow, sea craft, and soft machines.

The pair put forward a fresh route to show that answers to the pair of laws exist. Their route steps through a chain of close tasks. Each task in the chain is set on a domain whose shape path is fixed in advance. By passing from one task to the next, they build toward the full moving case. The result adds a new proof tool for this class of flow tasks.

Catalog record

CREATORCharles M. Elliott ⋆, Thomas Sales †
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE190 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2601-13004

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

An iterative approach to a fluid-rigid body interaction problem Charles M. Elliott ⋆, Thomas Sales † Abstract . We study a novel approach for the existence of solutions to an incompressible fluid-rigid body interaction problem in three dimensions. Our approach introduces an iteration based on a sequence of related problems posed on domains with prescribed evolution. In particular we prove the short-time existence of strong solutions to a system coupling the incompressible Navier–Stokes equations to the ordinary differential equations governing the motion of a rigid body, with no slip boundary conditions on the boundary of the rigid body, provided that the relative density, ρ ρ B \frac{\rho}{\rho_{B}} , is sufficiently small. We also discuss the use of our iterative approach in numerical methods for the moving boundary problem, and complement this with some numerical experiments in two dimensions which demonstrate the necessity of the smallness assumption on ρ ρ B \frac{\rho}{\rho_{B}} . 2020 Mathematics Subject Classification 35R37, 35Q30, 76D05 1. Introduction This paper considers a novel iterative approach to construct solutions to a moving boundary problem where the incompressible Navier–Stokes equations are coupled to the equations of motion for a rigid body. The movement of the boundary comes from the motion of the rigid body, and so one must simultaneously solve for both the solutions to the system of differential equations and the (time-dependent) fluid domain. In particular we study the short time existence of functions (\mathbf{q},\boldsymbol{\omega},\mathbf{u},p) which solve, in a suitable weak sense, (1.1a) \displaystyle\rho\left(\frac{\partial\mathbf{u}}{\partial t}+(\mathbf{u}\cdot\nabla)\mathbf{u}\right)=-\nabla p+\mu\Delta\mathbf{u}, (1.1b) \displaystyle\nabla\cdot\mathbf{u}=0, on a non-cylindrical space-time domain, \bigcup_{t\in[0,T]}\Omega(t)\times\{t\}\subset\mathbb{R}^{3}\times[0,T] , which is not known a priori. Here \rho,\mu are the constant density and (dynamic) viscosity of the fluid respectively. At each time the fluid domain is given by \Omega(t):=(\Omega\setminus B(t))^{\circ} for B(t) a rigid body with motion determined by its the centre of mass, \mathbf{q} , and angular momentum, \boldsymbol{\omega} , solving (1.2a) \displaystyle m\frac{\mathrm{d}^{2}\mathbf{q}}{\mathrm{d}t^{2}}=-\int_{\partial B(t)}\mathbb{T}\boldsymbol{\nu}, (1.2b) \displaystyle\frac{\mathrm{d}}{\mathrm{d}t}\left(\mathbb{J}\boldsymbol{\omega}\right)=-\int_{\partial B(t)}(\mathbf{x}-\mathbf{q}(t))\times\mathbb{T}\boldsymbol{\nu}, subject to appropriate initial conditions. We will often omit the measures in integrals when it is clear from context. Here \boldsymbol{\nu} denotes the outward unit normal vector, \mathbb{T} denotes the viscous stress tensor, and \mathbb{J} the inertial tensor, \displaystyle\mathbb{T}:=-p\mathbb{I}+\mu\left(\nabla\mathbf{u}+(\nabla\mathbf{u})^{T}\right), \displaystyle\mathbb{J}:=\int_{B(t)}\rho_{B}\left(|\mathbf{x}-\mathbf{q}(t)|^{2}\mathbb{I}-(\mathbf{x}-\mathbf{q}(t))\otimes(\mathbf{x}-\mathbf{q}(t))\right), where \mathbb{I} is the identity matrix, and \rho_{B} is the constant density of the rigid body. We note that one considers the unit normal vector, \boldsymbol{\nu} , to be pointing out of the fluid domain \Omega(t) , so that on \partial B(t) the normal vector is pointing into B(t) . Although the pressure, p , is only unique up to a constant this does not change how we understand the boundary integrals in ( 1.2 ) since the divergence theorem yields \int_{\partial B(t)}\boldsymbol{\nu}=0=\int_{\partial B(t)}(\mathbf{x}-\mathbf{q}(t))\times\boldsymbol{\nu}. Thus, without loss of generality, we shall consider p such that \int_{\Omega(t)}p=0 at all times. Here ( 1.1 ) are the usual incompressible Navier–Stokes equations, and ( 1.2 ) are the equations of motion for a rigid body (in three dimensions). To avoid confusing notation we define the boundary of \Omega(t) as a union of disjoint sets \partial\Omega(t)=\Gamma\cup\partial B(t) , where \Gamma=\partial\Omega is the (stationary) outer portion of the boundary and \partial B(t) is the boundary of the rigid body. B(t) \Omega(t) \mathbf{u}=0 \mathbf{u}=\boldsymbol{q}^{\prime}(t)+\boldsymbol{\omega}(t)\times(\boldsymbol{x}-\boldsymbol{q}(t)) \Gamma Figure 1 . A diagram showing a fluid-rigid body interaction problem. The system ( 1.1 ), ( 1.2 ) is complemented with no-slip boundary conditions (1.3a) \displaystyle\mathbf{u}=0,\text{ on }\Gamma, (1.3b) \displaystyle\mathbf{u}=\mathbf{q}^{\prime}(t)+\boldsymbol{\omega}(t)\times(\mathbf{x}-\mathbf{q}(t)),\text{ on }\partial B(t), and initial conditions for \mathbf{u} , \mathbf{q} , and \boldsymbol{\omega} (1.4) \displaystyle\mathbf{u}(0)=\mathbf{u}_{0},\quad\mathbf{q}(0)=\mathbf{q}_{0},\quad\frac{\mathrm{d}\mathbf{q}}{\mathrm{d}t}(0)=\mathbf{v}_{0},\quad\boldsymbol{\omega}(0)=\boldsymbol{\omega}_{0}, where \nabla\cdot\mathbf{u}_{0}=0 . We shall assume throughout that the initial data are compatible in the sense that \mathbf{u}_{0}|_{\partial B(0)}=\mathbf{v}_{0}+\boldsymbol{\omega}_{0}\times(\mathbf{x}-\mathbf{q}_{0}),\text{ on }\partial B(0). We shall consider \mathbf{u}_{0}\in\mathbf{H}_{0}^{1}(\Omega) (where we are using boldface letters to denote vector-valued Sobolev spaces) so that this equality is understood in the trace sense. We assume throughout that \Gamma and \partial B(0) are both C^{2,1} surfaces, so that \Omega(0) is a C^{2,1} domain. We also assume that the rigid body and the outer boundary, \Gamma , are not initially in contact — that is to say there exists some \delta>0 such that \mathrm{dist}(\Gamma,B(0))>\delta>0. This system models the motion of a rigid body immersed in a viscous, incompressible fluid, with no slip on the boundary of the rigid body. The no-slip boundary conditions ( 1.3 ) are typically considered in the literature, but one could alternatively consider Navier-slip boundary conditions on the boundary of the rigid body, that is we replace ( 1.3b ) with (1.5a) \displaystyle\mathbf{u}\cdot\boldsymbol{\nu}=\left(\mathbf{q}^{\prime}(t)+\boldsymbol{\omega}(t)\times(\mathbf{x}-\mathbf{q}_{0})\right)\cdot\boldsymbol{\nu}, (1.5b) \displaystyle\left(\mu\left(\nabla\mathbf{u}+(\nabla\mathbf{u})^{T}\right)\boldsymbol{\nu}\right)\times\boldsymbol{\nu}=-\beta\left(\mathbf{u}-\mathbf{q}^{\prime}(t)+\boldsymbol{\omega}(t)\times(\mathbf{x}-\mathbf{q}_{0})\right)\times\boldsymbol{\nu}, on \partial B(t) for some slip length \beta>0 . With these boundary conditions the system has been studied in [ 4 , 20 , 44 ] . Fluid-structure interaction problems naturally find application in the biomedical sciences, namely to the study of blood flow around the heart [ 9 , 18 , 75 ] and the design of stents [ 19 ] . More generally, free/moving boundary problems are often motivated by applications in applied sciences and industry, as discussed in [ 32 ] . For further details on fluid-structure interactions and their applications we refer the reader to [ 17 , 41 ] and the references therein. Before stating our main result we firstly we define what our notion of a strong solution of ( 1.1 ), ( 1.2 ) entails. Definition 1.1 . A strong solution to ( 1.1 ), ( 1.2 ) is a quadruple of functions (\mathbf{q},\boldsymbol{\omega},\mathbf{u},p) , defined on a time interval [0,T] , such that the following hold: (1) \mathbf{q}\in\mathbf{H}^{2}([0,T]) , and \boldsymbol{\omega}\in\mathbf{H}^{1}([0,T]) define a time-dependent domain \Omega(t) , and a parametrisation \Phi(t):\Omega(0)\rightarrow\Omega(t) . The rigid body evolves via B(t)=\Phi(B(0),t) and is such that \mathrm{dist}(\Gamma,B(t))>\frac{\delta}{2}>0, for all t\in[0,T] . (2) The pairs (\mathbf{W}^{k,q}(\Omega(t)),\Phi(t))_{t\in[0,T]} are compatible, as defined in § 2.1 , for all k=0,1,2 and q\in[0,\infty] . (3) The fluid quantities are such that \mathbf{u}\in H^{1}_{\mathbf{L}^{2}}(\Phi)\cap L^{2}_{\mathbf{H}^{2}}(\Phi) , p\in L^{2}_{H^{1}\cap L^{2}_{0}}(\Phi) . Moreover (\mathbf{u},p) solve ( 1.1 ) pointwise almost everywhere on \bigcup_{t\in[0,T]}\Omega(t)\times\{t\} , along with the boundary conditions ( 1.3 ), and initial condition \mathbf{u}(0)=\mathbf{u}_{0} on \Omega(0) . (4) (\mathbf{q},\boldsymbol{\omega}) solve ( 1.2 ) almost everywhere on [0,T] , and satisfy the initial conditions ( 1.4 ). For a similar notation of a strong solution to this problem we refer the reader to [ 73 , Definition 2.1] . We defer further details on function spaces and notation to Section 2 . With this definition of a strong solution, our main result is the following short time existence theorem. Theorem 1.2 . Let the relative density, \frac{\rho}{\rho_{B}} , be sufficiently small 1 1 1 The smallness of \frac{\rho}{\rho_{B}} is determined by the geometry of \Omega(0) and \mathrm{diam}(B(0)) (see the proofs of Lemma 4.4 and Lemma 4.11 ). . Then there exists some time T , depending only on the initial data and the geometry of \Omega(0) , such that there exists a unique quadruple (\mathbf{q},\boldsymbol{\omega},\mathbf{u},p)\in\mathbf{H}^{2}([0,T])\times\mathbf{H}^{1}([0,T])\times L^{2}_{\mathbf{H}^{2}}(\Phi)\cap H^{1}_{\mathbf{L}^{2}}(\Phi)\times L^{2}_{H^{1}}(\Phi) of strong solutions solving ( 1.1 ), ( 1.2 ) in the sense of Definition 1.1 . Remark 1.3 . Some basic numerical experiments (see Section 5 ) indicate that the smallness condition on the relative density, \frac{\rho}{\rho_{B}} , is necessary for our iterative approach to be convergent. The novelty of this present work is not in establishing the existence of strong solutions, as indeed this has been known since the early 2000s

Read the complete text