Boundedness of solutions to quasilinear parabolic equations (Q329249)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Boundedness of solutions to quasilinear parabolic equations
scientific article

    Statements

    Boundedness of solutions to quasilinear parabolic equations (English)
    0 references
    0 references
    0 references
    0 references
    21 October 2016
    0 references
    The authors prove boundedness properties of the weak solutions to the quasilinear parabolic problem \(u_{t}-\mathrm{div}(a(x,t,u,Du))=b(x,t,u,Du)\) posed in \(\Omega \times (0,T)\) with the homogeneous Dirichlet boundary condition \(u=0\) on \(\partial \Omega \times (0,T)\) and the initial condition \( u=u_{0}\in L^{2}(\Omega )\) in \(\Omega \times \{0\}\). Here \(\Omega \) is a bounded domain of \(\mathbb{R}^{n}\), \(n\geq 2\). The nonlinear terms \(a\) and \(b \) are supposed to satisfy sub-controlled growth properties \(\left| a(x,t,z,\xi )\right| \leq \Lambda (\varphi (x,t)+\left| z\right| ^{\frac{\ell (m-1)}{m}}+\left| \xi \right| ^{m-1})\), \(\left| b(x,t,z,\xi )\right| \leq \Lambda (\psi (x,t)+\left| z\right| ^{\ell -1}+\left| \xi \right| ^{\frac{m(\ell -1)}{\ell }})\), with \( \frac{m(n+2)}{n+m}<\ell <\frac{m(n+2)}{n}\), \(m>\frac{2n}{n+2}\) and a coercivity property \(a(x,t,z,\xi )\cdot \xi \geq \gamma \left| \xi \right| ^{m}-\Lambda \left| \xi \right| ^{\ell }-\Lambda \varphi ^{\frac{m}{m-1}}(x,t)\). Here \(\varphi \) (resp. \(\psi \)) is a nonnegative function in the space \(L^{q_{1}}(0,T;L^{p_{1},\lambda _{1}}(\Omega ))\) (resp. \(L^{q_{2}}(0,T;L^{p_{2},\lambda _{2}}(\Omega ))\)) with hypotheses on the parameters \(q_{i},p_{i},\lambda _{i}\) in terms of \(n\) and \(m\). The authors first define the notion of weak solution in this context, introducing a variational formulation of the Cauchy-Dirichlet problem. The main result proves that every weak solution to this Cauchy-Dirichlet problem is essentially bounded in \(\Omega \times (\sigma T,T)\) for some \(\sigma \in (0,1)\). Moreover if \(u_{0}\) is essentially bounded in \(\Omega \) the weak solution is globally essentially bounded in \(\Omega \times (0,T)\). For the proof, the authors estimate the integral \(\int_{t_{1}}^{t_{2}}\int_{\Omega }v^{s_{0}}\phi dxdt\) with \(v\in L^{\infty }(0,T;L^{2}(\Omega ))\cap L^{m}(0,T;W_{0}^{1,m}(\Omega ))\) and \(\phi \in L^{q}(0,T;L^{p,\lambda }(\Omega ))\) for appropriate values of the different parameters in terms of the Dirichlet energy of \(v\) in the same time strip \(\Omega \times (t_{1},t_{2})\). This requires an Adams trace inequality for functions in \( W_{0}^{1,r}(\Omega )\). The authors also use the De Giorgi approach in order to get boundedness properties, extending the results of \textit{O. A. Ladyzhenskaya} et al. [Линейные и квазилинейные уравнения параболического типа (Russian). Moskva: Izdat `Nauka' (1967; Zbl 0164.12302)].
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    quasilinear parabolic operator
    0 references
    Cauchy-Dirichlet problem
    0 references
    weak solution
    0 references
    boundedness
    0 references
    sub-controlled controlled growth
    0 references
    coercivity
    0 references
    Morrey space
    0 references
    mass estimate
    0 references
    time strip
    0 references
    Adams trace inequality
    0 references
    De Giorgi approach
    0 references
    0 references