Regularity for Orlicz phase problems

Homogenization in context of PDEs; PDEs in media with periodic structure (35B27) Smoothness and regularity of solutions to PDEs (35B65) Variational methods for second-order elliptic equations (35J20) Research exposition (monographs, survey articles) pertaining to calculus of variations and optimal control (49-02) Regularity of solutions in optimal control (49N60)

Review text

Let \(\Omega\) be a bounded domain in \(\mathbb{R}^n\), \(n\in\mathbb{N}\setminus\{1\}\), and consider functionals of the type\N\[\NW^{1,1}(\Omega)\ni v\mapsto\mathcal{F}(v,\Omega):=\int_\Omega F(x,v,Dv)\mathrm{d}x,\N\]\Nwhere \(F:\Omega\times\mathbb{R}\times\mathbb{R}^n\rightarrow\mathbb{R}\) is a Carathéodory function satisfying the double-sided bound\N\[\N\nu\Psi(x,|z|)\leq F(x,y,z)\leq L\Psi(x,|z|)\qquad\mbox{in }\Omega\times\mathbb{R}\times\mathbb{R}^n,\N\]\Nfor some \(L\geq\nu>0\), where\N\[\N\Psi(x,t):=G(t)+a(x)H_a(t)+b(x)H_b(t),\qquad(x,t)\in\Omega\times[0,\infty),\N\]\Nfor sufficiently smooth Young functions \(G,H_a,H_b\) possessing superlinear indexes (see Definition 2.1) and nonnegative \(a\in C^{0,\alpha}(\Omega),b\in C^{0,\beta}(\Omega)\), \(\alpha,\beta\in(0,1]\).\N\NA function \(u\in W^{1,1}_{\mathrm{loc}}(\Omega)\) is a \textit{local minimizer} of the above functional if\N\begin{itemize}\N\item[\(\bullet\)] \(\Psi(x,|Du|)\in L^1(\Omega)\), and\N\item[\(\bullet\)] \(\mathcal{F}(u,\mathrm{supp}\{u-v\})\leq\mathcal{F}(v,\mathrm{supp}\{u-v\})\) for all \(v\in W^{1,1}_{\mathrm{loc}}(\Omega)\) with \(\mathrm{supp}\{u-v\}\) compactly contained in \(\Omega\).\N\end{itemize}\NSet\N\[\N\Lambda(s,t):=\frac{s^\alpha}{1+s^\alpha}\frac{H_a(t)}{G(t)}+\frac{s^\beta}{1+s^\beta}\frac{H_b(t)}{G(t)},\qquad s,t>0.\N\]\NUnder minimal/natural smoothness, convexity, monotonicity, growth and other structural conditions on \(F\) (hence on \(H_a,H_b,G\)), the authors provide several optimal regularity results for local minimizers \(u\) which lie in the Musielak-Orlicz-Sobolev space \(W^{1,\Psi}(\Omega)\). For instance,\N\begin{itemize}\N\item[(i)] if \(\sup_{t>0}\Lambda(t,G^{-1}(t^{-n}))<\infty\), where \(G^{-1}\) is the inverse function of \(G\), then\N\[\Nu\in C^{0,\theta}_{\mathrm{loc}}\mbox{ for every }\theta\in(0,1),\qquad Du\in C^{0,\theta}_{\mathrm{loc}}\mbox{ for some }\theta\in(0,1).\N\]\N\item[(ii)] If in addition \(u\) is bounded, then the conclusions of \((i)\) are true, provided\N\[\N\sup_{t>0}\Lambda(t,t^{-1})<\infty.\N\]\N\end{itemize}\NThe case \(a\in C^{\omega_\alpha}(\Omega)\) and \(b\in C^{\omega_\beta}(\Omega)\), where \(\omega_\alpha\) and \(\omega_\beta\) are real functions satisfying conditions that makes them moduli of continuity, is also investigated. This changes the definition of \(\Lambda\) accordingly.

Cites work

Cited in

(8)

This page was built for publication: Regularity for Orlicz phase problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6986768)