KIT | KIT-Bibliothek | Impressum | Datenschutz

Feedforward and Feedback Tracking Control of Nonlinear Diffusion−Convection−Reaction Systems Using Summability Methods

Meurer, T. ORCID iD icon 1; Zeitz, M.
1 Institut für Mechanische Verfahrenstechnik und Mechanik (MVM), Karlsruher Institut für Technologie (KIT)

Abstract (englisch):

Motion planning and design of feedforward and feedback tracking control are studied for a tubular reactor modeled by nonlinear parabolic diffusion−convection−reaction equations. The approach is based on formal power series parametrizations of the system states and inputs, whereby the domain of convergence of the formal series solutions can be greatly extended using appropriate summation methods. As a result, feedforward controls can be determined for a wide range of trajectories and system parameters. Furthermore, on the basis of a reinterpretation of the formal power series parametrization, state feedback with asymptotic tracking control and an observer can be designed for the tubular reactor. On the other hand, following the 2-degrees-of-freedom approach, the feedforward control can be supplemented by a standard output feedback to obtain robust tracking control for tubular reactors with parameter sets reflecting the unpacked tubular reactor up to the real fixed-bed reactor.


Originalveröffentlichung
DOI: 10.1021/ie0495729
Scopus
Zitationen: 64
Web of Science
Zitationen: 47
Dimensions
Zitationen: 61
Zugehörige Institution(en) am KIT Institut für Mechanische Verfahrenstechnik und Mechanik (MVM)
Publikationstyp Zeitschriftenaufsatz
Publikationsdatum 01.04.2005
Sprache Englisch
Identifikator ISSN: 0888-5885, 1520-5045
KITopen-ID: 1000156674
Erschienen in Industrial & Engineering Chemistry Research
Verlag American Chemical Society (ACS)
Band 44
Heft 8
Seiten 2532–2548
Vorab online veröffentlicht am 16.10.2004
Schlagwörter Approximation, Computational chemistry, Order, Palladium, Power
Nachgewiesen in Dimensions
Web of Science
Scopus
OpenAlex
KIT – Die Universität in der Helmholtz-Gemeinschaft
KITopen Landing Page