Uniform convergence of the POD method and applications to optimal control (Q255853)

A proper orthogonal decomposition (POD) based on Galerkin approximations is considered. A POD basis is used to get reduced-order models (ROM). The authors derive error estimates for linear abstract parabolic evolution problems that establish the uniform convergence to optimal control problems and they also give a convergence rate of the POD ROM solution with respect to forcing functions. They show how to avoid problems related to unmodeled dynamics. The results are demonstrated on some numerical examples.