Well-Posedness and Asymptotic Decay of Solutions to the Three-Dimensional Euler Equations with Damping

Gui-Qiang G. Chen, Feimin Huang, Houzhi Tang, Shuxing Zhang, Weiyuan Zou · 2025

Read Well-Posedness and Asymptotic Decay of Solutions to the Three-Dimensional Euler Equations with Damping by Gui-Qiang G. Chen, Feimin Huang, Houzhi Tang, Shuxing Zhang, Weiyuan Zou (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-2502-12457.txt.

Introduction

Gui-Qiang Chen and co-authors settle in 2025 a long-standing question about gas flow equations with friction. The compressible Euler equations with damping describe gases that lose momentum to resistance. Whether smooth solutions exist for all time in three dimensions had resisted proof, except for small data in some regimes.

The team proves global existence and decay for both the standard and the borderline isothermal cases, even when part of the starting data is large. A new framework derives all needed estimates from one smallness condition on the flow's gradients. The sharp decay rates follow, and the isothermal case is also reached as a limit.

Catalog record

CREATORGui-Qiang G. Chen, Feimin Huang, Houzhi Tang, Shuxing Zhang, Weiyuan Zou
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE82 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2502-12457

Browse the full library · About the Tower of Babel

Full text

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

Well-Posedness and Asymptotic Decay of Solutions to the Three-Dimensional Euler Equations with Damping Gui-Qiang G. Chen, Feimin Huang, Houzhi Tang, Shuxing Zhang, Weiyuan Zou Abstract . The global well-posedness of the multi-dimensional compressible Euler equations with damping remains a longstanding open problem. This problem has been partially resolved in the isentropic regime ( i.e. , the adiabatic exponent γ > 1 \gamma>1 ) for small smooth initial data (see [ 31 , 29 ] ). In this paper, we establish the global well-posedness and asymptotic decay of smooth solutions of the Cauchy problem of the three-dimensional compressible Euler equations with damping for the isentropic regime γ > 1 \gamma>1 and the isothermal regime γ = 1 \gamma=1 , allowing for partially large initial data. More precisely, the L 2 L^{2} -norm of the initial data is allowed to be large, while the third-order Sobolev norm of the initial data is assumed to be small. For the isentropic case, we develop a new analytical framework in which all required a priori estimates of solution ( ρ , u ) (\rho,u) can be derived under the condition that ∫ 0 T ( ‖ ∇ ρ ‖ L ∞ + ‖ ∇ u ‖ L ∞ ) ​ 𝑑 t \int_{0}^{T}\big(\|\nabla\rho\|_{L^{\infty}}+\|\nabla u\|_{L^{\infty}}\big)\,\mathrm{d}t remains sufficiently small. Moreover, we obtain the optimal algebraic decay rates of global solutions. Furthermore, we study the isothermal limit of solutions of the isentropic regime as γ → 1 \gamma\to 1 , and establish the global well-posedness and asymptotic decay of solutions to the isothermal Euler equations with damping. 2020 Mathematics Subject Classification 35Q31, 35B40, 35B44, 76N15 1. Introduction The three-dimensional Euler equations with damping can be written as (1.1) \left\{\begin{aligned} &\partial_{t}\rho+\mathrm{div}(\rho u)=0,\\ &\partial_{t}(\rho u)+\mathrm{div}(\rho u\otimes u)+\nabla P(\rho)=-\rho u,\\ \end{aligned}\right. which model compressible gas flow through a porous medium. Here, the unknown functions \rho=\rho(t,x) and u=u(t,x) denote the density and velocity of the fluid in the domain (t,x)\in\mathbb{R}_{+}\times\mathbb{R}^{3} , respectively. The pressure P:[0,\infty)\times\mathbb{R}^{3}\to\mathbb{R} is given by the barotropic pressure law P=P(\rho)=a_{0}\rho^{\gamma}\qquad\mbox{\text{for $a_{0}>0$}}, where \gamma>1 is the adiabatic exponent, while the case \gamma=1 corresponds to the isothermal case. Without loss of generality, we normalize a_{0}=1 for simplicity. System ( 1.1 ) is equipped with the initial data: (1.2) (\rho,u)(0,x)=(\rho_{0},u_{0})(x), and the far-field states: (1.3) \lim_{|x|\rightarrow\infty}(\rho,u)(t,x)=(\rho_{*},0), where \rho_{*}>0 is a given constant. The global existence of solutions to the \gamma -law Euler equations is a challenging problem because the solution develops singularities, such as shocks, in finite time, regardless of how small and smooth the initial data are. For the one-dimensional case, Nishida [ 26 ] used the Glimm difference scheme to prove the global existence of BV solutions with large initial data for \gamma=1 as the gas is away from the vacuum. This result was extended to the case \gamma>1 by Nishida-Smoller [ 28 ] when \gamma is close 1. When vacuum occurs, DiPerna [ 6 ] first established the global existence of solutions with large initial data in L^{\infty} for \gamma=1+\frac{2}{2n+1} , where n\geq 2 denotes the number of degrees of freedom of the molecules. Later, the existence problem was solved by Ding-Chen-Luo [ 3 , 4 ] for \gamma\in(1,\frac{5}{3}] , Lions-Perthame-Tadmor [ 20 ] for \gamma\in[3,\infty) , and Lions-Perthame-Souganidis [ 19 ] for \gamma\in(\frac{5}{3},3) . The \gamma=1 case was treated by Huang-Wang [ 17 ] . On the other hand, the global existence of solutions with large initial data in the multidimensional (M-D) case remains open. It is known that the damping can improve the regularity of solutions; see [ 8 , 10 , 11 , 12 , 13 , 27 , 29 , 16 , 18 , 33 ] . That is, the damping prevents the development of singularities in some sense and leads to the global existence of smooth solutions for sufficiently small initial data. However, the damping cannot prevent the formation of singularities for large initial data; see [ 21 , 22 ] . The entropy solution has to be considered; see [ 1 , 2 , 5 ] . It is fundamental to further study the regularity of the damping structure. In this paper, we establish the global well-posedness and asymptotic decay of smooth solutions of the Cauchy problem of the M-D compressible Euler equations with damping for some partially large initial data, i.e. , \|(\rho_{0}-\rho_{*},u_{0})\|_{L^{2}} is allowed to be large, but \|D^{3}(\rho_{0}-\rho_{*},u_{0})\|_{L^{2}} is necessarily small. Our results consist of two parts. In the first part, concerning the isentropic regime ( \gamma>1 ), the precise statement is following: Theorem 1.1 . Let the initial data (\rho_{0}-\rho_{*},u_{0})\in H^{3}(\mathbb{R}^{3}) satisfy 0<\underline{\rho}\leq\rho_{0}\leq\bar{\rho}<\infty for some positive constants \underline{\rho} and \bar{\rho} , (1.4) \|(\rho_{0}-\rho_{*},u_{0})\|_{L^{2}}\leq M_{0}\;,\qquad\qquad\|D^{3}(\rho_{0}-\rho_{*},u_{0})\|_{L^{2}}\leq\delta_{0}, where \delta_{0}>0 is small and 0<M_{0}\leq\delta_{0}^{-\frac{1}{11}} . Then the Cauchy problem ( 1.1 )–( 1.3 ) for the isentropic case, i.e., \gamma>1 , admits a unique global classical solution (\rho-\rho_{*},u)\in C([0,\infty);H^{3}(\mathbb{R}^{3})) satisfying (1.5) \displaystyle\|D^{k}(\rho-\rho_{*})(t)\|_{L^{2}}\leq C(1+t)^{-\frac{k}{2}}, (1.6) \displaystyle\|D^{k}u(t)\|_{L^{2}}\leq C(1+t)^{-\frac{k+1}{2}},\quad for k=0,1,2,3 , where C=C(M_{0},\underline{\rho},\bar{\rho},\delta_{0}) is a positive constant independent of t\geq 0 . The equation of state for isothermal gases can be regarded as the limit of that for isentropic gases as \gamma\to 1 . Hence, the second main result of this paper addresses the isothermal limit of the isentropic damped Euler system ( 1.1 )-–( 1.3 ). Theorem 1.2 (Isothermal Limit) . Assume that the conditions in Theorem 1.1 hold. Then, as \gamma\to 1 , the solutions of the Cauchy problem ( 1.1 )–( 1.3 ) for the isentropic regime ( \gamma>1 ) strongly converge to the solution of the corresponding problem for \gamma=1 . Moreover, the solution of the Cauchy problem ( 1.1 )–( 1.3 ) for \gamma=1 is a unique global classical solution (\rho-\rho_{*},u)\in C([0,\infty);H^{3}(\mathbb{R}^{3})) satisfying ( 1.5 )–( 1.6 ). Remark 1.1 . Theorems 1.1 – 1.2 admit the following initial data: (1.7) \displaystyle\rho_{0}=\rho_{*}+\varepsilon^{\frac{5}{11}}e^{-\varepsilon^{\frac{8}{11}}|x|^{2}},\qquad\qquad u_{0}=\varepsilon^{\frac{5}{11}}e^{-\varepsilon^{\frac{8}{11}}|x|^{2}}, where \varepsilon<\delta_{0} is small. Indeed, \|(\rho_{0}-\rho_{*},u_{0})\|_{L^{2}}=O(\varepsilon^{-\frac{1}{11}})\rightarrow\infty\;\text{as $\varepsilon\rightarrow 0$} , thus the L^{2} -norm of the initial data can be large. Remark 1.2 . The decay rates obtained in Theorems 1.1 – 1.2 are optimal, which are the same as the ones of the solution to the corresponding linear system. The strategies of the proofs are outlined as follows: Firstly, for the isentropic case \gamma>1 , we introduce a new function (1.8) \phi=\frac{2}{\gamma-1}\big(\sigma(\rho)-\sigma(\rho_{*})\big), where \sigma(\rho)=\sqrt{P^{\prime}(\rho)} is the sound speed and \sigma(\rho_{*})=\bar{\sigma} corresponds to the sound speed at the background density \rho_{*} . Then system ( 1.1 )–( 1.3 ) can be reformulated as a symmetric hyperbolic system with partial damping: (1.9) \displaystyle\begin{cases}\partial_{t}\phi+\bar{\sigma}\,{\mathrm{div}}u=-u\cdot\nabla\phi-\frac{\gamma-1}{2}\phi\,{\mathrm{div}}u,\\[2.84526pt] \partial_{t}u+\bar{\sigma}\,\nabla\phi+u=-u\cdot\nabla u-\frac{\gamma-1}{2}\phi\nabla\phi.\end{cases} For this system, the key observation is that all L^{2} -norm of the k -order derivatives of the solution can be controlled by the k -order derivative of the initial data. More precisely, there exists a constant C>0 such that (1.10) \|D^{k}(\phi,u)(t)\|_{L^{2}}\leq C\|D^{k}(\phi_{0},u_{0})\|_{L^{2}}\qquad\text{for \,$k=0,1,2,3,$} provided that (1.11) \int_{0}^{T}\big(\|\nabla\phi\|_{L^{\infty}}+\|\nabla u\|_{L^{\infty}}\big)\,\mathrm{d}t is small. Then, by the analysis of the Green function G_{ij} for the linearized system of ( 1.9 ) (see ( 3.33 ) below), we obtain that, for any test function \zeta , (1.12) \|D^{k}G_{ij}^{\ell}\ast\zeta\|_{L^{2}}\leq\begin{cases}C(1+t)^{-\frac{k}{2}}\|\zeta\|_{L^{2}}&\text{for \;$i+j=2$},\\[2.84526pt] C(1+t)^{-\frac{k+1}{2}}\|\zeta\|_{L^{2}}&\text{for \;$i+j=3$},\\ C(1+t)^{-\frac{k+2}{2}}\|\zeta\|_{L^{2}}\quad&\text{for \;$i+j=4$},\end{cases} where G^{\ell}_{ij} corresponds to the low-frequency part. Based on ( 1.12 ) and the Duhamel principle, we can further obtain \|D^{k}\phi^{\ell}\|_{L^{2}}\leq C(1+t)^{-\frac{k}{2}}\qquad\text{for \;$k=1,2,3.$} Using the energy method again, we have (1.13) \frac{\mathrm{d}}{\mathrm{d}t}\|(D^{k}\phi,D^{k}u)\|_{H^{3-k}}^{2}+\|(D^{k}\phi,D^{k}u)\|_{H^{3-k}}^{2}\leq C\|D^{k}\phi^{\ell}\|_{L^{2}}^{2}\qquad\text{for \;$k=1,2,3,$} which implies (1.14) \|D^{k}(\phi,u)(t)\|_{L^{2}}\leq C(1+t)^{-\frac{k}{2}}\qquad\text{for \;$k=1,2,3.$} On the other hand, since

Read the complete text