Approximate solution of a class of optimal control problems in domains of arbitrary form (Q1592012)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Approximate solution of a class of optimal control problems in domains of arbitrary form |
scientific article |
Statements
Approximate solution of a class of optimal control problems in domains of arbitrary form (English)
0 references
7 April 2002
0 references
The authors solve the optimal control problem with an elliptic-type operator of the second order \(A\), the state of a system \(y(u)\) is determined as a solution of the problem \(Ay(u)=f+u\), \(x=(x_1,x_2)\in \Omega\), \(y(u)|_\Gamma = 0\), \(Ay=-\Delta y+q(x)\), \(0<q_0\leq q(x)\in L_{\infty}(\Omega)\), \(f\in L_2(\Omega)\), \(\Omega\) is a bounded simply connected domain with the boundary \(\Gamma\in C^2.\) The problem of the optimal control consists in obtaining \[ \inf_{v\in U_g} J(v), \] \[ J(v)=\int_\Omega [y(v)-z_g]^2 dx +\mu \int_\Omega v^2 dx,\;\mu >0,\;U_g=\{v\in L_2(\Omega)\mid v\geq 0 \text{ a.e. in }\Omega\}. \] Using a conjugate state, the problem is converted to the nonlinear boundary value problem \(Ay+\frac 1\mu \inf(0,p)=f(x)\), \(x\in \Omega\), \(y(x)=0\), \(x\in \Gamma\), \(Ap-y=-z_g(x)\), \(x\in \Omega\), \(p(x)=0\), \(x\in \Gamma.\) The optimal control \(u\) can be found from the equality \(u=-\frac 1\mu \inf(0,p).\) The theorem on the existence and uniqueness of the above problem is verified. Using the method of fictitious domains the numerical solution by the difference scheme on the rectangular domain is given. The convergence theorem is verified.
0 references
optimal control
0 references
elliptic systems
0 references
difference scheme
0 references
net method
0 references