Achim Küpper

Online identification of adsorption isotherms in SMB processes via efficient moving horizon state and parameter estimation

Abstract In this paper, a concept to identify the adsorption isotherm of Simulated Moving Bed processes online during the operation is proposed. The influence of model parameters on the measurements is analyzed by a sensitivity analysis which enables to identify the set of parameters that can be estimated simultaneously. The parameters are estimated in real-time by a moving horizon state and parameter estimation scheme. Numerical simulations of validated models for separation problems with nonlinear isotherms of Langmuir type are presented. Furthermore, it is it shown that a structural plant/model mismatch in the void fraction can be compensated to a large extent by modifying the Henry coefficients of the adsorption isotherm such that the estimated model can be used in the context of online optimizing control.

Non-linear Model Predictive Control of the Hashimoto Simulated Moving Bed Process

In recent years, continuous Chromatographic processes have been established as an efficient separation technology in industry, especially when temperature sensitive components or species with similar thermodynamic properties are involved. In SMB processes, a counter-current movement of the liquid and the solid phases is achieved by periodically switching the inlet and the outlet ports in a closed loop of Chromatographic columns. The integration of reaction and separation in one single plant is a promising approach to overcome chemical or thermodynamic equilibria and to increase process efficiency. Reactive Chromatographie SMB processes in which the columns are packed with catalyst and adsorbent have been proposed and demonstrated successfully. However, a full integration often is not efficient because in the columns in the separating zones, the catalyst is not used or even counterproductive. By placing reactors between the separation columns at specific positions around the feed port, a more efficient process, the Hashimoto SMB process, is established. In this contribution, a non-linear predictive control concept for the Hashimoto SMB process is presented. The controller computes optimal control variables (flow rates and the switching time) to optimize an economic objective over a moving horizon. The purity requirements of the product streams are implemented as constraints and not as controlled variables. The optimization-based controller is combined with a scheme to estimate selected model parameters in order to reduce the influence of the inevitable model errors. Simulative results are presented for the example of the racemization of Troger’s base.

OPTIMIZING CONTROL OF THE HASHIMOTO SMB PROCESS: EXPERIMENTAL APPLICATION

Abstract In this contribution, a non-linear predictive control strategy for the Hashimoto SMB process that combines reaction and continuous chromatographic separation is discussed and applied to a pilot plant. The controller computes optimal control variables (flow rates and the switching time) such that an economic objective is optimized over a moving horizon. The purity requirements of the product streams are implemented as constraints and not as controlled variables. The concept is extended to the case of high purities. Simulative and experimental results are presented for the example of the racemization of Trogers base.

PARAMETER AND STATE ESTIMATION IN CHROMATOGRAPHIC SMB PROCESSES WITH INDIVIDUAL COLUMNS AND NONLINEAR ADSORPTION ISOTHERMS

Abstract In this paper, measurement based parameter and state estimation in Simulated Moving Bed plants with nonuniform columns is investigated. The estimation strategy presented uses the available measurements of the concentrations in the product flows and in one internal flow which is realistic for industrial applications. The estimation task is solved in a decentralized fashion. The correction of the parameters and the state is performed only for the column positioned in front of the respective measurement. Convergence is achieved by the shift of the product concentration measurements. The local estimation problems are solved by Extended Kalman filters. The scheme is validated for a propanolol isomers system with nonlinear adsorption isotherms.

Optimierungsbasierte Regelung des Hashimoto-SMB-Prozesses (Optimizing Control of the Hashimoto SMB Process).

Zusammenfassung Simulierte Gegenstromprozesse (SMB-Prozesse) sind leistungsfähige chromatographische Trennverfahren, bei denen die physikalischen Eingangsgrößen (Zufluss- und Abflussströme) periodisch von einer Trennsäule zur nächsten geschaltet werden, sodass sich ein effektiver Gegenstrom von Feststoff und Flüssigkeit ergibt. Der sog. Hashimoto-Prozess integriert chemische Reaktoren in diesen Prozess, deren Position sich aber relativ zu den Zuflüssen nicht ändert. SMB-Trennprozesse und insbesondere Prozesse mit zusätzlichen Reaktoren zeigen ein komplexes nichtlineares Verhalten. In diesem Beitrag wird eine direkt optimierende prädiktive Regelung vorgestellt, die auf der Grundlage eines rigorosen, nichtlinearen Prozessmodells hoher Ordnung unter Einbeziehung aller Freiheitsgrade des Prozesses (Flussraten und Taktzeit) online den Lösungsmittelverbrauch minimiert und gleichzeitig die Produktreinheit und die gewünschte Ausbeute einhält. Es wird gezeigt, dass der Einfluss von Modellfehlern zu einem unbefriedigenden Regelverhalten führen kann und es wird eine Modifikation der Formulierung der Regelaufgabe vorgestellt, die zu einem robusten Verhalten führt. Das Regelungskonzept wird anhand von Simulationen und experimentellen Ergebnissen für die Razemisierung von Trögerscher Base verifiziert.

Engineering of Online Optimizing Control - A Case Study: Reactive SMB Chromatography

Abstract This contribution discusses the application of the idea of online optimizing control to a complex process from chemical engineering. The idea of online optimizing control is to minimize an economic objective over a finite moving horizon during plant operation based upon a rigorous nonlinear dynamic model. Plant limitations and product specifications are included in the optimization as constraints. The process discussed here is the so-called Hashimoto SMB process that combines reaction and continuous chromatographic separation in one plant. The degrees of freedom (flow rates and the switching time) are used to minimize the solvent consumption and to keep the product purity and product recovery above the specified values. The emphasis of this paper is on the modifications of the formulation of the optimization problem in order to cope with plant/model mismatch.

Efficient Moving Horizon State and Parameter Estimation for the Varicol SMB Process

Abstract Abstract In this paper, a moving horizon state and parameter estimation (MHE) scheme for the Varicol process is presented. The Varicol process is an extension of the Simulated Moving Bed (SMB) process that realizes non-integer column distributions over the separation zones by an asynchronous switching of the inlet and outlet ports (the ports are shifted individually). These additional degrees of freedom can be used to yield an improvement in economical performance compared to SMB operation. The proposed estimation scheme is based on a rigorous SMB model that incorporates rigorous chromatographic columns and port switching. The absence of model simplifications allows the extension of the estimation scheme to the more complex Varicol process. The goal of the estimation scheme is to reconstruct the full state of the system, i.e. the concentration profiles along all columns, and to identify critical model parameters in the presence of noisy measurements. The estimation is based on measurements of the concentrations of the components at the two outlet ports (which are asynchronously switched from one column to the next) and at one fixed location between two columns. The state estimation scheme utilizes a deterministic model within the prediction horizon. State noise is only considered in the state and in the parameters up to the beginning of the horizon. By applying a multiple-shooting method and a real-time iteration scheme for solving the resulting optimization problem, the computation times are reduced and the scheme can be applied online. A numerical simulation for an enantiomer separation system with nonlinear adsorption isotherm is presented.

Determining SMB Superstructures by Mixed-Integer Optimal Control

We treat a simplified model of a Simulated Moving Bed (SMB) chromatographic separation process that contains time-dependent discrete decisions. SMB processes have been gaining increased attention lately, see [3], [4] for further references. The related optimization problems are challenging from a mathematical point of view, as they combine periodic nonlinear optimal control problems in partial differential equations (PDE) with time-dependent discrete decisions. For this problem class of mixed-integer optimal control problems (MIOCP) a novel numerical method, developed in [5], is applied.

Optimization of Simulated Moving Bed Processes

In this contribution, the optimization of periodic chromatographic simulated moving bed SMB processes is discussed. The rigorous optimization is based on a nonlinear pde model which incorporates rigorous models of the chromatographic columns and the discrete shifts of the inlet and outlet ports. The potential of the optimization is demonstrated for a separation problem with nonlinear isotherm of the Langmuir type for an SMB process and the ModiCon process. Here, an efficient numerical approach based on multiple shooting is employed. An overview of established optimization approaches for SMB processes is given.

Real-Time Moving Horizon State and Parameter Estimation for SMB Processes

Abstract Advances in numerical algorithms have rendered the application of advanced process control schemes feasible for complex chemical processes that are described by high-order first-principles models. Applying real-time iteration schemes reduces the CPU requirement such that rigorous models can be applied that enable a precise forecast of the system behaviour. In this paper, a moving horizon state and parameter estimation scheme for chromatographic simulated moving bed SMB processes is presented. The simultaneous state and parameter estimation is based on a high-order nonlinear SMB model which incorporates rigorous models of the chromatographic columns and the discrete shifting of the inlet and outlet ports. The estimation is performed using sparse measurement information: the concentrations of the components are only measured at the two outlet ports (which are periodically switched) and at one fixed location between two columns. The goal is to reconstruct the full state of the system, i.e. the concentration profiles along all columns, and to identify model parameters reliably. The state estimation scheme assumes a deterministic model within the prediction horizon, state noise is only present in the state and in the parameters prior to and at the beginning of the horizon. The scheme can be applied online. The advantage of this estimation scheme is that it is applicable to all process scenarios encountered during the real operation of an SMB plant, e.g. start up, transition periods, varying flows and switching times, since no model simplification nor a state reduction scheme are applied. Numerical simulations (start up of the SMP process) of a validated model for a separation problem with nonlinear isotherms of the Langmuir type demonstrate the efficiency of the algorithm.