Reaction mechanism simplification using mixed-integer nonlinear programming

For large, complex reacting systems, computational efficiency becomes a critical issue in process simulation, optimization, and model-based control. Mechanism simplification is often a necessity to improve computational speed. We present a novel approach to reaction mechanism simplification that formulates the model reduction problem as a mixed-integer nonlinear programming problem and solves it using DICOPT + + (discrete and continuous optimizer). Two formulations of the mechanism simplification problem are considered, one involving the elimination of reactions and the other the elimination of species. Both steady-state and dynamic problem formulations are developed. Solutions for example problems having six reactions and six species are presented. (C) 2000 Elsevier Science Ltd. All rights reserved.