Invited Lecture - Abstract

Optimization problems in chemical engineering often involve complex systems of nonlinear DAE as the model equations. The direct multiple shooting method has been known for a while as a fast off-line optimization method in ODE and later in DAE. In recent years this approach has been successfully adapted to the specific requirements of real-time optimization in case of perturbations. NMPC is a real-time optimization problem with a specific structure. Special strategies have been developed in which a progress of the optimization iterations is nested with the progress of the real process. Multiple shooting combined with an initial value embedding effectively reduces the on-line computations. Precalculated information - gradients, Hessians and QP factorizations for iterated reference trajectories - minimizes response time. In typical real-time problems these approaches have proven much faster than fast off-line strategies. Experimental results for NMPC of a distillation column are presented.

(Based on joint work with M. Diehl, A. Schaefer, J. Schloeder, F. Allgoewer, R. Findeisen, I. Uslu, S. Schwarzkopf)