Nonlinear parabolic problem with time fractional derivative

Nikolai Kutev, Tsviatko Rangelov · 2026

Read Nonlinear parabolic problem with time fractional derivative by Nikolai Kutev, Tsviatko Rangelov (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-2603-13534.txt.

Introduction

Nikolai Kutev and Tsviatko Rangelov study a nonlinear parabolic problem with a time fractional derivative. The equation pairs a p-Laplacian term with a double singular Hardy-type potential. They prove a comparison principle for its weak solutions, plus a priori estimates that bound them. They then ask when global weak solutions exist and when solutions blow up in finite time. The answer turns on the value of the optimal Hardy constant, they show. Their work sets out the line between the two regimes for this class of problems.

Catalog record

CREATORNikolai Kutev, Tsviatko Rangelov
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32179
LICENSEhttp://creativecommons.org/publicdomain/zero/1.0/
RIGHTSCC0
SOURCEarXiv

Source: https://arxiv.org/abs/2603.13534

Browse the full library · About the Tower of Babel

Full text

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

Nonlinear parabolic problem with time fractional derivative Nikolai Kutev, Tsviatko Rangelov Abstract Time fractional parabolic problem for p-Laplacian with double singular Hardy-type potential is considered. Comparison principle and appriory estimates for the weak solutions are proved. Existence of global weak solutions and finite-time blow-up are investigated depending on the optimal Hardy constant. † † Corresponding author: T. Rangelov, [email protected] Keywords Fractional derivative on time, p-Laplacian, Hardy inequality, Comparison principle, Global existence, Finite time blow-up. Math. Subj. Class. 35K92, 26A33, 35B44 1 Introduction Let \displaystyle D_{t}^{\alpha} , \displaystyle\alpha\in(0,1) be the Liouville-Caputo fractional derivative with respect to \displaystyle t\in[0,\infty) , D_{t}^{\alpha}h(t)=\frac{\partial}{\partial t}\frac{1}{\Gamma(1-\alpha)}\int_{0}^{t}(t-s)^{-\alpha}\left[h(s)-h(0)\right]ds, where \displaystyle\Gamma is Gamma function, see Kiryakova 1994 and Kilbas et al. 2006 . Denote by \displaystyle\Omega=\{x\in\mathbb{R}^{n},|x|\leq\varphi(x)\} a bounded star-shaped domain with respect to a small ball centered at the origin and \displaystyle\varphi(x)\in C^{0,1}(\mathbb{R}^{n}) is a positive homogeneous function of 0-th order. We consider the parabolic problem with time fractional derivative \left\{\begin{array}[]{ll}&D_{t}^{\alpha}u-\hbox{div}\left(|\nabla u|^{p-2}\nabla u\right)-\mu W(x)|u|^{p-2}u=0,\hbox{ in }Q=(0,T)\times\Omega,\\ &u(t,x)=0,\hbox{ for }(t,x)\in[0,T]\times\partial\Omega,\\ &u(0,x)=u_{0}(x),\hbox{ for }x\in\Omega\subset\mathbb{R}^{n},u_{0}(x)\in L^{2}(\Omega),\end{array}\right. (1.1) where \displaystyle n>p , \displaystyle p>2 , \displaystyle\mu>0 , \displaystyle\Delta_{p}=\hbox{div}\left(|\nabla u|^{p-2}\nabla u\right) is p-Laplacian and the potential \displaystyle W(x) is singular on the boundary \displaystyle\partial\Omega and at the origin W(x)=|x|^{-p}\left[1-\left(\frac{|x|}{\varphi(x)}\right)^{\frac{n-p}{p}}\right]^{-p}. (1.2) The theory of fractional evolutionary PDEs is intensively investigated in the last decades. Fractional time derivative PDEs, including Laplace, p-Laplace, porous medium, mean curvature, Kirchhoff operator and slow diffusion are studied in Dipierro et al. 2019 , Tarasov 2013 and Uchaikin 2000 . In Allen et al. 2016 , Djida et al. 2019 , Dong and Kim 2021 , Kubica and Yamamoto 2018 , Li et al. 2014 , Luchko 2009 , Vergara and Zacher 2010 , Vergara and Zacher 2015 existence and uniqueness of the solutions is proved, while in Smadiyeva and Torebek 2023 optimal decay estimates are established. Let us mention that evolutionary PDEs with fractional Laplace and p-Laplace operator are considered in Mazón et al. 2016 , Fu and Pucci 2016 , Boudjeriou 2022 where existence of local solutions, global solvability and final time blow-up by means of the potential well method are proved. The applications of fractional calculus appear in the theory of mechanics of fluids, probability and statistics, dynamical processes in self-similar and and porous structures, viscoelasticity, electrochemistry of corrosion, optic and signal processing and so on, see Kilbas et al. 2006 . For example, in mathematical physics when describing dynamic processes in material with memory in the theory of heat conduction, models of anomalous diffusion, in the context of spatially disordered systems, porous media, fractional media, sites of micro-molecular crowding, stock price movements ( Ben-Avraham and Havlin 2000 , Kiselev et al. 2007 , Metzler and Klafter 2000 ), magneto-thermoelastic heat conduction ( Ezzat and Karamany 2011 ) and hydrodynamics ( Atangana and Kilicman 2014 ). As for classical parabolic problems with singular potentials, they are models in molecular physics ( Lèvy-Leblon 1967 ), quantum cosmology ( de Castro 2004 ), non-relativistic quantum mechanics ( Peral and Vazquez 1995 ) and others. Global existence or finite time blow-up is proved for inverse-square potential in Baras and Goldstein 1984 , Cabré and Martel 1999 and for more general singularities at an interior point in Cabré and Martel 1999 , Fernandez-Cara 1997 , Anita and Tataru 2000 and Anita and Tataru 2002 . Singular potentials on the whole boundary are treated in Boccardo and Murat 1992 , while double singular potentials at an interior point and on the boundary are investigated in Kutev and Rangelov 2022 . The goal of the present paper is to extend the result for classical parabolic problems in the works mention above to time fractional evolutionary PDEs with p-Laplace operator and double singular potential. We prove finite time blow-up or global existence of the weak solutions to ( 1.1 ) and thresholds between them is the optimal Hardy constant, see Theorem 9.3 in Kutev and Rangelov 2022 . Note that for potential \displaystyle W(x) the following Hardy inequality holds \int_{\Omega}|\nabla u|^{p}dx\geq\Lambda(n,p)\int_{\Omega}W(x)|u|^{p}dx, (1.3) with optimal constant \displaystyle\Lambda(n,p)=\left(\frac{n-p}{p}\right)^{p} . Let us construct the truncated problem \left\{\begin{array}[]{ll}&D_{t}^{\alpha}u_{N}-\Delta_{p}u_{N}-\mu W_{N}(x)|u_{N}|^{p-2}u_{N}=0,\hbox{ in }Q,\\ &u_{N}(t,x)=0,\hbox{ for }(t,x)\in[0,T]\times\partial\Omega,\\ &u_{N}(0,x)=u_{0}(x),\hbox{ for }x\in\Omega.\end{array}\right. (1.4) for every positive integer \displaystyle N\geq 1 , where \displaystyle W_{N}(x)=\min\{N,W(x)\} . Definition 1.1 . The function \displaystyle u(t,x) is a weak solution in distributional sense of ( 1.1 ) if u(t,x)\in C^{0}\left([0,\tau];L^{2}(\Omega)\right)\cap L^{p}\left([0,\tau];W^{1,p}_{0}(\Omega)\right),\ \ \hbox{ for some }\tau>0, and is a limit of the solutions \displaystyle u_{N}(t,x) of the truncated problem ( 1.4 ), for \displaystyle N\rightarrow\infty . Definition 1.2 . The solution of ( 1.1 ) defined in \displaystyle[0,\tau) does not blow-up at \displaystyle\tau if there exists a constant \displaystyle C_{0} such that \|u(t,x)\|_{L^{2}(Q_{\tau})}\leq C_{0},\quad\|u(t,x)\|_{W^{1,p}(Q_{\tau})}\leq C_{0}\hbox{ for }t\in[0,\tau),x\in\Omega, (1.5) where \displaystyle Q_{\tau}=(0,\tau)\times\Omega . Our aim is to prove the following two theorems Theorem 1.1 . Let \displaystyle\Omega\subset\mathbb{R}^{n} , \displaystyle n>p , \displaystyle p>2 , \displaystyle 0\in\Omega , be a star-shaped domain with respect to a small ball centered at the origin, and \displaystyle\mu<\Lambda(n,p) . If \displaystyle u_{0}(x)\in W_{0}^{1,p}(\Omega) then every weak solution of the problem problem ( 1.1 ) defined in the maximal existence interval \displaystyle(0,T_{m}) satisfies ( 1.5 ) for every \displaystyle\tau\in(0,T_{m}) and does not blow-up at \displaystyle T_{m} . Theorem 1.2 . Suppose \displaystyle\mu>\Lambda(n,p) , \displaystyle p>2 , \displaystyle n>p , \displaystyle u_{0}(x)>0 in \displaystyle\bar{\Omega} , \displaystyle u_{0}(x)\in L^{\infty}(\Omega)\cap W_{0}^{1,p}(\Omega) . Then problem ( 1.1 ) has no global in time weak solution because it blows-up for a finite time. The plan of the paper is the following. In Sect. 2 we consider fractional ordinary differential equation (FODE) and prove comparison principle (Theorem 2.1 ) for Eq. ( 2.1 ). Then, for initial-value problem of special type FODE ( 2.8 ) and using the comparison principle (Theorem 2.2 , Lemma 2.1 in Lakshmikantham and Vatsala 2008 ) we prove Theorem 2.3 - the existence of blow-up solution of the problem ( 2.8 ). Sect. 2 is essential for the proof of the main results and has its own interest. In Sect. 3.1 we establish a comparison principle (Theorem 3.1 for time fractional p-Laplace equation ( 3.1 ) using the ”fundamental identity” (Lemma 3.1 in Zacher 2019 ). In Sect. 3.2 for initial boundary value problem with truncated potential ( 3.13 ) we prove some useful appriory estimates. Finally, in Sect. 4 the main results Theorem 1.1 and Theorem 1.2 are proved. 2 Time fractional ordinary differential equations 2.1 Comparison principle Consider the ordinary fractional differential equation D^{\alpha}_{t}w(t)=f(t,w(t)),\hbox{ for }t\in[0,T]. (2.1) Theorem 2.1 . Suppose \displaystyle u(t),v(t) are sub- and supersolutions of ( 2.1 ) and \displaystyle|u(t)|\leq M,\quad|v(t)|\leq M,\hbox{ for }t\in[0,T] , D^{\alpha}_{t}u(t)\leq f(t,u(t)),\quad D^{\alpha}_{t}v(t)\geq f(t,v(t)),\hbox{ for }t\in[0,T], (2.2) |f_{\omega}(t,\omega)|\leq K,\quad\hbox{ for }t\in[0,T],|\omega|\leq M. (2.3) If \displaystyle u(0)\leq v(0) then u(t)\leq v(t),\hbox{ for }t\in[0,T]. (2.4) Proof. Suppose by contradiction that ( 2.4 ) fails. Define the function z(t)=\left(u(t)-v(t)\right)_{+}=\left\{\begin{array}[]{lll}&u(t)-v(t)&\hbox{ for }u(t)-v(t)>0,\\ &0&\hbox{ for }u(t)-v(t)\leq 0,\end{array}\right. Then there exists \displaystyle T_{0} , \displaystyle 0\leq T_{0}<T , \displaystyle T_{0}=\inf\{t;z(t)>0\} and function \displaystyle z(t)=0 for \displaystyle t\in[0,T_{0}] . Moreover, from ( 2.2 ) \displaystyle D^{\alpha}_{t}z(t) satisfies the inequality D^{\alpha}_{t}z(t)\leq f(t,u(t))-f(t,v(t))\hbox{ for }t\in[0,T]. (2.5) Since f(t,u(t))-f(t,v(t))=\left(u(t)-v(t)\right)\int_{0}^{1}\frac{\partial f}{\partial\omega}\left(t,(1-\theta)v(t)+\theta u(t)\right)d\theta, and from ( 2.3 ) and \displaystyle|(1-\theta)v(t)+\theta u(t)|\leq M , \displaystyle|f(t,u(t))-f(t,v(t))|\leq Kz(t) for \displaystyle t\in[0,T] inequality ( 2.5 ) becomes D^{\alpha}_{t}z(t)\leq Kz(t),z(0)=0. Thus D^{\alpha}_{t}z(t)=\frac{1}{\Gamma(1-\alpha)}\frac{\partial}{\partial t}\int_{0}^{t}(t-s)^{-\alpha}z(s)ds\leq Kz(t),\hbox{ for }t\in[0,T], (2.6) and integrating ( 2.6 ) from

Read the complete text