A stochastic optimal control problem for the heat equation on the halfline with Dirichlet boundary-noise and boundary-control (Q607563)

From MaRDI portal
scientific article
Language Label Description Also known as
English
A stochastic optimal control problem for the heat equation on the halfline with Dirichlet boundary-noise and boundary-control
scientific article

    Statements

    A stochastic optimal control problem for the heat equation on the halfline with Dirichlet boundary-noise and boundary-control (English)
    0 references
    0 references
    22 November 2010
    0 references
    This paper deals with an optimal control problem for a state equation of parabolic type on the halfline \((0,+\infty)\): \({\partial y\over\partial s}(s,\xi)= {\partial^2 y\over\partial \xi^2}(s,\xi)+f(s,y(s,\xi))\), \(s\in[t,T]\), \(\xi\in(0,+\infty)\), \(y(t,\xi)=x(\xi)\), \(y(s,0)=u\_s+\dot W_s\). Here \(\{W_t\), \(t\geq0\}\) is a standard real Wiener process, \(y(s,\xi,\omega)\) is the unknown real-valued process and represents the state of the system; the control is given by the real-valued processes \(u(s,\xi,\omega)\) acting at 0; \(x: (0,+\infty)\to\mathbb R\). The optimal control problem in this paper consists in minimizing the following finite horizon cost \(J(t,x,u)=E\int_t^T\int_0^{+\infty}l(s,\xi,y(s,\xi),u_s)\,d\xi\,ds+ E\int_0^{+\infty}\varphi(\xi,y(T,\xi))\,d\xi\), and also minimizing the following infinite horizon cost \(J(x,u)=E\int_0^{+\infty}e^{-\mu s}\int_0^{+\infty} l(s,\xi,y(s,\xi),u_s)\,d\xi\,ds \). The authors transpose the controlled state equation in the infinite dimensional framework and study regularity properties of the solution of this forward state equation. The backward equation associated to the problem is studied. Existence and uniqueness of the Hamilton-Jacobi-Bellman equation are proved. The case of infinite horizon is also considered.
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    heat equation
    0 references
    Dirichlet boundary conditions
    0 references
    boundary noise
    0 references
    boundary control
    0 references
    stochastic optimal control
    0 references
    backward stochastic differential equations
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references