- Neu
Buch, Englisch, Band 2026,2, 200 Seiten, Format (B × H): 148 mm x 210 mm, Gewicht: 268 g
Reihe: Schriftenreihe des Lehrstuhls für Systemdynamik und Prozessführung
Buch, Englisch, Band 2026,2, 200 Seiten, Format (B × H): 148 mm x 210 mm, Gewicht: 268 g
Reihe: Schriftenreihe des Lehrstuhls für Systemdynamik und Prozessführung
ISBN: 978-3-8191-0775-7
Verlag: Shaker
Iterative real-time optimization (RTO) aims at computing improved operating points of processing plants despite plant-model mismatch by repeatedly correcting a model-based optimization problem using plant measurements. This thesis focuses on iterative RTO methods that combine model adaptation with modifier adaptation. In this way, plant measurements are used both to reduce the mismatch between model predictions and plant behavior and to assess whether the model used for optimization is adequate for the measurement-based correction of the optimality conditions. The goal of this thesis is to investigate how iterative RTO can be made more efficient and reliable when structural and parametric plant-model mismatch limit the reliability of the available model. In particular, the thesis investigates how parameter estimation can be integrated into modifier adaptation so that the model used in the modified optimization problem is not only fitted to plant measurements, but also adequate for convergence to a plant optimum. This requires addressing the model-adequacy conditions that are relevant for the modified optimization problem. Based on this analysis, approaches are developed that improve the use of available plant information and enhance the convergence behavior of iterative RTO methods using modifier adaptation. The proposed approaches are demonstrated using representative simulation examples, including the Williams–Otto reactor and a fed-batch reactor, as well as a practical membrane process case study.
