On \(L^ {p}\)-solutions of semilinear stochastic partial differential equations. (Q1879484)

From MaRDI portal
scientific article
Language Label Description Also known as
English
On \(L^ {p}\)-solutions of semilinear stochastic partial differential equations.
scientific article

    Statements

    On \(L^ {p}\)-solutions of semilinear stochastic partial differential equations. (English)
    0 references
    0 references
    0 references
    22 September 2004
    0 references
    Let \(D\subseteq \mathbb R^ {d}\) be a bounded convex domain with a smooth boundary \(\partial D\), \(W\) a \(k\)-dimensional standard Wiener process defined on a stochastic basis \((\Omega , \mathfrak F, (\mathfrak F_ {t}), P)\). Existence and uniqueness of solutions to a semilinear stochastic parabolic equation \[ {\partial u\over \partial t}= \sum ^ {d}_ {i=1}{\partial \over \partial x_ {i}}\left (\sum ^ {d}_ {j=1} a_ {ij}(x){\partial u \over \partial x_ {j}} + g_ {i}(t,x,u)\right ) + f(t,x,u)+ \sum ^ {k}_ {i=1} \sigma _ {i}(t,x,u)\dot W^ {i},\tag{1} \] \(u=0\) on \(\partial D\), \(u(0,\cdot ) = u_ {0}\), with nonlinear terms of a polynomial growth is studied. More precisely, assume that \(a_ {ij}\in C^ 2(\bar D)\), the matrix \((a_ {ij}(x))\) is symmetric for each \(x\in D\) and the uniform ellipticity condition is satisfied. The functions \(f,g_ {i}, \sigma _ {i}: \mathbb R_ {+}\times D\times \mathbb R\to \mathbb R\) are Borel, \(f(t,x,\cdot )\), \(\sigma _ {i}(t,x,\cdot )\) are globally Lipschitz continuous and satisfy the linear growth condition uniformly with respect to \((t,x)\). The functions \(g_ {i}\) are locally Lipschitz and of a polynomial growth: there exist \(\nu \geq 1\) and \(L<\infty \) such that \[ | g_ {i}(t,x,r)-g_ {i}(t,x,s)| \leq L(1 + | r| ^ {\nu -1} + | s| ^ {\nu -1})| r-s|, \quad g_ {i}(t,x,r) = g^ {(1)}_ {i}(t,x,r) + g^ {(2)}_ {i}(t,r), \] \[ | g^ {(1)}_ {i} (t,x,r)| \leq L(1+| r| ),\quad | g^ {(2)}_ {i}(t,r)| \leq L(1+| r| ^ {\nu }) \] for every \((t,x)\) and all \(r,s\in \mathbb R\). Under these hypotheses it is shown that \(\rho _ 0\), depending on \(\nu \) and \(d\), may be found such that for every \(\rho >\rho _ 0\) and every \(\mathfrak F_ 0\)-measurable \(L^ \rho (D)\)-valued random variable \(u_ 0\) there exists a unique continuous \(L^ \rho (D)\)-valued solution \(u\) to (1). The process \(u\) has a modification continuous in \((t,x)\in \mathbb R_ {+}\times D\) provided that \(u_ 0\) has a continuous modification. Moreover, a comparison theorem is established.
    0 references
    0 references
    0 references
    stochastic partial differential equations
    0 references
    stochastic Burgers equation
    0 references