On nondegenerate quasilinear stochastic partial differential equations (Q1347115): Difference between revisions

The author considers the stochastic PDE \[ {\partial \over {\partial t}} u(t, x)= {{\partial^2} \over {\partial x^2}} u(t,x)+ f(t,x,u(t,x))+ g(t,x, u(t, x)) {{\partial^2} \over {\partial t\partial x}} w \] with \(u(t, 0)= u(t,1)=0\) \((t\in [0,T ])\), \(u(0,x)= u_0 (x)\) \((x\in [0,1 ])\), where \((\partial^2/ \partial t \partial x)w\) is a space-time white noise. The solution is understood in the generalized sense defined by \textit{J. B. Walsh} [in: Ecole d'été de probabilités de Saint Flour XIV-1984. Lect. Notes Math. 1180, 265-437 (1986; Zbl 0608.60060)]. The author obtains existence and uniqueness and the comparison theorems for the above problem when \(f\) is a locally bounded measurable function and \(g\) is only locally Lipschitz in the third variable \(r\). For a proof he adopts the Krylov technique of passage to limit based on the uniform integrability of the densities of the occupation measures of the converging random fields, and so he does not require \(L_p\) estimate for these densities such as in N. V. Krylov's work (from 1969) and the work of \textit{V. Bally}, the author and \textit{E. Pardoux} [J. Funct. Anal. 120, No. 2, 484-510 (1994; Zbl 0801.60049)].