A New approach for the unsteady Stokes equations with time fractional derivative in Bounded Domains
Juan C. Oyola Ballesteros, Paulo M. Carvalho-Neto · 2025
Read A New approach for the unsteady Stokes equations with time fractional derivative in Bounded Domains by Juan C. Oyola Ballesteros, Paulo M. Carvalho-Neto (2025) 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-2511-13896.txt.
Introduction
Juan Carlos Oyola Ballesteros and Paulo M. Carvalho-Neto build a new framework for the unsteady Stokes equations, which govern slow viscous fluid flow. Their version uses a Caputo fractional derivative in time, and that nonlocal operator breaks the standard proof tools. Classical Galerkin methods cannot be applied in their usual form.
The pair design new function spaces tailored to the fractional setting and recast weak solutions inside them. With those spaces, a generalized Galerkin scheme goes through rigorously. The work extends the classical theory of these flow equations and lays groundwork for wider classes of fractional partial differential equations in fluid mechanics and nearby fields.
Catalog record
| CREATOR | Juan C. Oyola Ballesteros, Paulo M. Carvalho-Neto |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 85 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-13896
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A New approach for the unsteady Stokes equations with time fractional derivative in Bounded Domains Juan C. Oyola Ballesteros, Paulo M. Carvalho-Neto Abstract . In this work, we introduce a novel variational framework for the study of the unsteady Stokes equations in a bounded, open Lipschitz domain Ω ⊂ ℝ n \Omega\subset\mathbb{R}^{n} , involving a Caputo fractional derivative in time. The nonlocal nature of the fractional derivative presents significant analytical challenges, making classical methods, such as the Faedo–Galerkin approach, inadequate in their standard form for a full analysis of the problem. To address these difficulties, we develop and rigorously analyze new functional spaces specifically designed for the fractional setting. These spaces allow us to reformulate weak solutions in a manner consistent with the fractional dynamics, thereby enabling the successful implementation of a generalized Galerkin scheme. Our formulation not only extends the classical theory but also provides a foundation for the study of broader classes of fractional partial differential equations in fluid mechanics and related areas. 2020 Mathematics Subject Classification 35R11, 26A33, 76D07, 35A15, 35A01, 35D30, 35Q35 1. Introduction One of the most renowned mathematical models in fluid dynamics is the Navier-Stokes equations. These equations aim to determine the velocity and pressure fields within a fluid. They are among the most widely used equations, describing the physics of numerous phenomena of both economic and academic interest. The Navier-Stokes equations find applications in modeling weather patterns, ocean currents, water flow in canoes, and many other areas. These equations can be derived directly from Newton’s laws under the assumption of incompressibility, where pressure does not affect the fluid’s volume. For more details, we refer to [ 13 , Chapter 1] . A significant breakthrough in the theory of partial differential equations was the introduction of the concept of weak solutions by J. Leray, as presented in [ 22 ] , particularly in the context of the Navier–Stokes equations. Leray’s theory establishes the existence of solutions in a new variational framework, which may potentially be irregular. This approach relies on energy estimates and specific limiting procedures involving the weak and weak- \ast topologies in certain Bochner–Lebesgue spaces (see [ 14 , 15 , 16 ] for classical references). On the other hand, fractional calculus, with its applications to solving fractional-order differential equations (see [ 19 , 26 , 30 ] for early developments), has become essential for addressing problems in fields such as fluid dynamics, porous media, control theory, and dynamical systems (see [ 27 , 31 , 32 , 36 ] for classical applications). Unlike standard calculus, which deals with derivatives and integrals of integer order, fractional calculus extends these operations to real or complex orders. Consequently, the combined importance of the two topics mentioned above highlights the growing interest in recent studies on the Navier–Stokes equations with Caputo fractional derivatives of order \alpha\in(0,1) in time. These works aim to generalize the classical problem and explore the limiting case \alpha=1 (see [ 12 , 11 , 25 , 33 , 37 , 38 , 39 ] for some examples). However, it is important to note that most existing studies focus on establishing the existence and uniqueness of mild solutions for the abstract Cauchy problem associated with the fractional Navier–Stokes equations. In contrast, only a few works address the existence of weak solutions, mainly due to the technical difficulties of applying the Faedo–Galerkin method in the fractional context. For instance, [ 38 ] defines and proves the existence of a weak solution to the Navier–Stokes equations with a Caputo fractional derivative of order \alpha\in(0,1) , using the same definition as in the classical case when \alpha=1 . Similarly, [ 25 ] introduces a new functional space (see [ 25 , Item (4.5)] ) to study weak solutions of the multidimensional fractional Burgers equations (a special case of the compressible fractional Navier–Stokes equations) while still relying on the classical definition (see [ 25 , Definition 5.1] ). In conclusion, this work aims to go beyond traditional approaches by proposing a new formulation for weak solutions of partial differential equations with a Caputo fractional derivative in time. We not only introduce this new formulation, but also analyze a modified version of the functional spaces introduced in [ 25 ] , proving why they are more appropriate for the variational framework we adopt. Although all discussions up to this point have consistently led back to the fractional Navier–Stokes problem, it is important to clarify that the present work does not aim to establish the existence or uniqueness of its solutions. Rather, our focus lies in introducing new functional spaces and proving key properties related to them, inspired by those proposed in [ 25 ] . Due to the additional difficulties introduced by the nonlinear term in the Navier–Stokes equations, we restrict our attention in this initial study to its classical linearization: the unsteady Stokes equations. This choice simplifies the analysis while still serving to illustrate the proposed framework. The treatment of the full nonlinear case is left for future work. 1.1. A Novel Weak Formulation Let us first recall the fractional order approach to the unsteady Stokes equations. Let \Omega be a bounded Lipschitz subset of \mathbb{R}^{n} , for some n\geq 2 , and let \alpha\in(0,1) , T>0 , and \nu>0 be fixed constants. Consider f:[0,T]\times\Omega\rightarrow\mathbb{R}^{n} as the forcing term and u_{0}:\Omega\rightarrow\mathbb{R}^{n} as the initial condition. The unsteady Stokes equations with a Caputo fractional derivative in the time variable are given by (1) \left\{\begin{array}[]{rll}cD_{t}^{\alpha}u(t,x)-\nu\Delta u(t,x)+\nabla p(t,x)=f(t,x),&&\textrm{ in }(0,T)\times\Omega,\\ \operatorname{div}u(t,x)=0,&&\textrm{ in }(0,T)\times\Omega,\\ u(t,x)=0,&&\textrm{ on }[0,T]\times\partial\Omega,\\ u(0,x)=u_{0}(x),&&\textrm{ in }\Omega,\end{array}\right. where cD_{t}^{\alpha} stands for the Caputo fractional derivative of order \alpha (for more details see Section 2 ). It is common to use some classical notations in the study of the unsteady Stokes equation (see [ 34 ] for more details). Let us first recall the function space \mathcal{V}:=\left\{v\in\big[\mathcal{D}(\Omega)\big]^{n}:\operatorname{div}\;v=0\right\}, where \mathcal{D}(\Omega) denotes the test functions in \Omega . Also consider (2) H:=\overline{\mathcal{V}}^{\mathbb{L}^{2}(\Omega)}=\left\{v\in\mathbb{L}^{2}(\Omega):\operatorname{div}v=0\textrm{ in }\Omega\textrm{ and }v\cdot\vec{n}=0\textrm{ on }\partial\Omega\right\}, which with the induced topology of \mathbb{L}^{2}(\Omega):=\big[{L}^{2}(\Omega)\big]^{n} becomes a Hilbert space, and finally (3) V:=\overline{\mathcal{V}}^{\mathbb{H}_{0}^{1}(\Omega)}=\{v\in\mathbb{H}_{0}^{1}(\Omega):\;\operatorname{div}v=0\textrm{ and }\Omega\}, that equipped with the inner product inherited from \mathbb{H}_{0}^{1}(\Omega)=\big[{H}_{0}^{1}(\Omega)\big]^{n} is also a Hilbert space. Now suppose that u and p are sufficiently regular solutions of ( 1 ), and that f is at least in L^{2}(0,T;V^{\prime}) . Then, if \eta\in\mathcal{V} we have (4) (cD_{t}^{\alpha}u(t),\eta)_{H}+\nu(u(t),\eta)_{V}=\langle f(t),\eta\rangle_{V^{\prime},V},\textrm{ for a.e. }t\in[0,T]. Above the symbol \langle\cdot,\cdot\rangle_{V^{\prime},V} denotes the duality pairing between V^{\prime} and V . First, let \eta=u(t) in ( 4 ). By performing a formal calculation and applying [ 7 , Corollary 19] , we derive the energy inequality (5) \frac{1}{2}cD_{t}^{\alpha}\|u(t)\|^{2}_{H}+\nu\|u(t)\|^{2}_{V}\leq\langle f(t),u(t)\rangle_{V^{\prime},V},\text{ for a.e. }t\in[0,T], which becomes an equality when \alpha=1 . Following the classical ideas introduced by Leray in [ 22 , 23 , 24 ] , below we present a series of formal arguments to explore the regularity that can be expected from the weak solution of the unsteady Stokes equations, which is derived from ( 5 ). (i) Recall that when \alpha=1 , as is classically justified in the literature, integrating both sides of ( 5 ) from 0 to t and applying Young’s inequality allows us to deduce that \|u(t)\|^{2}_{H}+\nu\int_{0}^{t}\|u(s)\|^{2}_{V}\,ds\leq\|u_{0}\|^{2}_{H}+\frac{1}{\nu}\int_{0}^{t}\|f(s)\|^{2}_{V^{\prime}}\,ds,\textrm{ for a.e. }t\in[0,T]. The above inequality suggests that requiring u\in L^{2}(0,T;V)\cap L^{\infty}(0,T;H) is a