Control of simulated moving bed (SMB) chromatographic processes

Introduction and State of the Art

SMB chromatographic processes are nowadays frequently used for difficult separation problems like the separation of enantiomers, for example. When operated at the economic optimum with maximum productivity and minimum solvent consumption the process has maximum sensitivity to disturbances. Therefore, in practice, these processes are operated often in suboptimal way to introduce some robustness against disturbances. Main objective of this project is to develop feedback controllers to allow robust operation at the economic optimum.

SMB processes are operated cyclically by switching the inlets and outlets from column to column. Chromatographic columns are represented by nonlinear partial differential equations. For the control of these processes different concepts were proposed during the last decade. Most of these were based on dynamic online optimization, which is computationally challenging in the case of nonlinear models or has limitations when simpler linearized models are used.

Results

In this project we follow a different approach which has its roots in the PhD thesis of Marco Fütterer developed in our group [1]. The control concept is an adaptive cycle to cycle control method [2], [3], [4]. It is based on an online parameter estimator, which automatically delivers optimal operating conditions for total separation from rough initial guesses corresponding to suboptimal operation without much a priori knowledge. Parameter estimates can be used for a manual readjustment of operating conditions or for closed loop control, which allows the automatic rejection of disturbances and a precise adjustment of reduced outlet purities. In practice, reduced outlet purities, if acceptable, will lead to an increase of productivity. The parameter estimator and the control concept were validated experimentally in cooperation with the PCF group for bicalutamide enantiomers [4], [5].

Outlook

Current work focuses on control of ternary SMB processes which play an important role in biotechnological and pharmaceutical industries for the isolation of desired key components (component B on the chromatogram).

Most common configurations for ternary center-cut separation are 8-zone SMB with raffinate recycle or 8-zone SMB with extract recycle.

This project is supported by the International Max Planck Research School for Advanced Methods in Process and Systems Engineering (IMPRS ProEng).

Cooperation Partners

• PCF Group

• Prof. Alain Vande Wouwer, University of Mons, Mons, Belgium

 

References

 [1] Fütterer (2010). On Design and Control of Simulated Moving Bed Processes. PhD Thesis, Otto-von-Guericke-Universität Magdeburg.

 [2] P. Suvarov, A. Vande Wouwer and A. Kienle (2012). A simple robust control for simulated moving bed chromatographic separation. Proceedings of 8th IFAC International Symposium on Advanced Control of Chemical Processes, pp. 137-142.

 [3] P. Suvarov, A. Kienle, C. Nobre, G. De Weireld and A. Vande Wouwer (2014). Cycle to cycle adaptive control of simulated moving bed chromatographic separation processes. Journal of Process Control. 24 (2), pp. 357-367.

 [4] P. Suvarov, A. Vande Wouwer, J. W. Lee, A. Seidel-Morgenstern, A. Kienle (2016). Control of incomplete separation in simulated moving bed chromatographic processes. IFAC-Papers-Online, 49 (7), pp. 153-158.

 [5] P. Suvarov, A. Vande Wouwer, J. W. Lee, A. Seidel-Morgenstern, A. Kienle (2018). Online estimation of optimal operating conditions for simulated moving bed chromatographic processes. In preparation.