This paper focuses on the dynamic behavior of process networks consisting of a reactor with highly exothermic chemical reactions that occur with different rates, connected via a high material recycle stream to an external heat exchanger. By employing singular perturbation arguments, we show that the dynamics of the considered process networks typically exhibits three distinct time scales, the fastest one, in which the temperature dynamics evolve, a fast one, owing to the presence of the fast reactions, and a slow time scale due to the presence of the reactions with slow reaction rates. We derive reduced-order models for the dynamics in each time scale and outline a rational controller design framework that accounts for this time scale separation. Finally, we provide an example and illustrative numerical simulation results.
|Original language||English (US)|
|Title of host publication||Proceedings of the 16th IFAC World Congress, IFAC 2005|
|Number of pages||6|
|ISBN (Print)||008045108X, 9780080451084|
|State||Published - 2005|
|Name||IFAC Proceedings Volumes (IFAC-PapersOnline)|
Bibliographical noteFunding Information:
1 Partial support for this work by ACS-PRF, grant 38114-AC9 and NSF-CTS, grant 0234440 is gratefully acknowledged. ‡P. Daoutidis is currently at Aristotle University of Thessaloniki, Thessaloniki, Greece.
- Energy recycle
- Model reduction
- Nonlinear control
- Singular perturbations