Achim Kienle

Steady-state multiplicities in reactive distillation columns for the production of fuel ethers MTBE and TAME: theoretical analysis and experimental verification

In this contribution, the nonlinear dynamic behaviour of reactive distillation columns for the production of MTBE and TAME is studied. The focus is on steady-state multiplicity and a rigorous bifurcation analysis of pilot plant reactive distillation columns for both processes is presented. The different sources and physical causes for the existence of multiple steady-states in MTBE and TAME synthesis are discussed. Further, a rigorous experimental verification of steady-state multiplicity in a pilot plant reactive distillation column for the production of TAME is presented. Finally, some remarks on the implications of multiple steady-states on column operation are made.

Nonlinear computation in DIVA — methods and applications

Methods for one-parameter continuation and stability analysis of periodic solutions as well as two-parameter continuation of Hopf and limit bifurcation points have been added to the dynamic flowsheet simulation environment DIVA. They are specially tailored for large sparse systems of differential algebraic equations with arbitrary structural properties usually arising in dynamic flowsheet simulation of chemical processes and plants. The application of these methods combined with the other capabilities of such an integrated tool for nonlinear system analysis is demonstrated for two different types of processes. The first is the so-called circulation-loop reactor, which has been used for catalytic combustion of waste air. Through periodic operation of the reactor, abatement of small amounts of volatile organic compounds (VOCs) in waste air is possible autothermally or with minimum supply of energy. Control strategies are studied for extending the region of desired periodic operating points, when the amount of VOCs is too small for autonomous periodic operation. The second application is concerned with the nonlinear dynamic behaviour of reactive distillation processes for the production of fuel ethers MTBE and TAME. In particular, the existence of multiple steady states is analyzed for the practical important case, when pre-reaction is carried out prior to the reactive distillation column. Further, these results are compared with the case without pre-reaction and a physical interpretation is given.

Simulated moving bed process with cyclic modulation of the feed concentration

The improvement of the simulated moving bed (SMB) process based on the introduction of a cyclic modulation of the feed concentration is described. It is demonstrated that such a feed concentration gradient during the shifting cycle can improve the performance significantly. The productivity and the product concentrations can be increased while simultaneously the solvent consumption can be decreased compared to the conventional SMB process with constant feed parameters.

Low-order dynamic models for ideal multicomponent distillation processes using nonlinear wave propagation theory

A new approach to the development of low-order dynamic models for multicomponent distillation processes is presented. This approach makes direct use of well-known spatio-temporal pattern formation phenomena also termed as nonlinear wave propagation. It takes into account the typical features of multicomponent systems, i.e. coexistence of different constant pattern waves within a single section of a distillation column and the resulting wave interactions. In a first step, constant molar holdups and flow rates, constant pressure and constant relative volatilities are assumed. Under these conditions a rigorous analytical treatment is possible and a comparably simple but sound and robust method for nonlinear model reduction is developed. The approach applies to packed as well as staged columns provided the number of column stages is sufficiently large. Application is demonstrated for two different distillation processes with a three and a five component mixture, respectively. It is shown that the dynamic behaviour of the low-order model is in good agreement with corresponding reference model for a large set of operating conditions. Finally, extensions of the present approach to processes with variable molar flow rates as well as variable volatilities are discussed.

Improving simulated moving bed processes by cyclic modulation of the feed concentration

A new mode of operation for simulated moving bed (SMB) chromatographic processes is presented. While conventional SMB processes are operated with constant feed concentration, a modulation of the feed concentration during the switching cycles is introduced in the new mode of operation. Thereby the performance compared to the conventional SMB process is improved significantly. The productivity and the product concentration is increased. Simultaneously, the solvent consumption is decreased.

Nonlinear control of a reactive distillation column

Abstract Control of reactive distillation columns is a challenging task due to the complex dynamics arising from the coupling of reaction and separation. In this paper, asymptotically exact input/output-linearization is applied in simulation studies to an industrial reactive distillation column which is operated by Bayer AG. The resulting control law is rather general and can be easily adopted for other reactive distillation columns. This control scheme requires knowledge of the complete state of the process and therefore an observer is designed. Asymptotically exact input/output-linearization inherits robust stability from a robust observer. It is intuitively argued that the proposed observer is robust w.r.t. both model structure and parameter errors. In order to compensate for steady state observer offsets an outer control loop with simple PI-controllers is implemented. Simulation studies evidence that in comparison with a well-tuned linear controller the nonlinear controller shows a superior performance with respect to setpoint-changes and disturbances, even in the presence of unknown input delays.

Optimal operation of simulated moving bed chromatographic processes by means of simple feedback control

In this contribution, simple methods are presented for controlling a simulated moving bed (SMB) chromatographic process with standard PI (proportional integral) controllers. The first method represents a simple and model-free inferential control scheme which was motivated from common distillation column control. The SMB unit is equipped with UV detectors. The UV signals in the four separation zones of the unit are fixed by four corresponding PI controllers calculating the ratio of liquid and solid flow in the respective separation zone. In order to be able to adjust the product purity a second, model-based control scheme is proposed. It makes use of the nonlinear wave propagation phenomena in the apparatus. The controlled chromatographic unit is automatically working with minimum solvent consumption and maximum feed throughput--without any numerical optimization calculations. This control algorithm can therefore also be applied for fast optimization of SMB processes.

Nonlinear behaviour of an ideal reactor separator network with mass recycle

Abstract In this work the nonlinear behavior of a coupled reactor–separator network is analyzed using singularity and bifurcation theory. The reactor is modeled as a CSTR, which sustains an exothermic first-order reaction. The separator is modeled as a flash process. The effluent from the reactor is fed to the separator. The reactant-rich stream (assumed to be the bottoms) from the separator is recycled back to the reactor. Focus is on pure mass recycle, as the two units are decoupled energetically via heat exchangers. Two different modes of operation are considered. These are compared with each other and with the well-known stand alone reactor with a first-order exothermic reaction. For the first mode of operation, the feed rate is fixed and for the second the recycle rate is fixed. In practice, these different modes of operation can be achieved by a suitable control strategy. It is shown that the behavior crucially depends on the mode of operation. Fixing the feed rate can lead to severe operational problems including monotonic and oscillatory unstable steady states over a wide range of operating conditions. Further, parameter regions are identified, where no steady state exists at all. On the other hand when the recycle rate is fixed the coupled system admits at least one stable steady-state solution for a fixed set of operating conditions.

On the dynamics of the circulation loop reactor : numerical methods and analysis

The Circulation Loop Reactor has been developed during the last years as an alternative to the well-known switch flow reactor. In contrast to the latter the CLR is intended to work as an autonomous periodic system. Suitable operating conditions are determined by a bifurcation analysis. Further, we study domains of attraction and the transient behavior towards the different attractors. The analysis reveals a rich dynamic and steady-state behavior. The desired periodic regime is shown to be achievable for a large subset of operating conditions and to be robust with respect to disturbances. For the analysis numerical methods have been developed and integrated into the dynamic simulator DIVA. The methods are suitable for bifurcation and stability analysis of general, large multi unit plants and processes usually arising in the chemical process industries.

Molten Carbonate Fuel Cells : Modeling, Analysis, Simulation, and Control

Preface. List of Contributors. Part I Design and Operation. 1 MTUs Carbonate Fuel Cell HotModule (Gerhard Huppmann). 1.1 The Significance of Fuel Cells. 1.2 Basic Statements of Power Production and Combined Heat and Power Systems. 1.3 Fuels for Fuel Cells. 1.3.1 Fuels Containing Gaseous Hydrocarbons. 1.3.2 Synthesis Gases. 1.3.3 Group of Gasified Hydrocarbons. 1.3.4 Secondary Fuel. 1.4 Why Molten Carbonate Fuel Cells. 1.5 The Carbonate Fuel Cell and its Function. 1.6 Optimisation by Integration: The HotModule Concept. 1.7 Manufacturing. 1.8 Advantages of the MCFC and its Utilization in Power Plants. 1.8.1 Electrical Efficiency. 1.8.2 Modularity. 1.8.3 Inherent Safety. 1.8.4 Environmentally Friendly - Pollution Free. 1.8.5 Silent. 1.9 History. 1.9.1 The European MCFC Development Consortium. 1.9.2 Continuing of the HotModule Development at MTU CFC Solutions. 1.10 Possible Applications of MCFC Systems. 1.10.1 Different Applications Using Different Fuels. 1.10.2 Different Applications Using the Diffrent Products of the MCFC System. 1.11 Economical Impacts. 2 Operational Experiences (Mario Koch, Joachim Berndt, and Matthias Gundermann). 2.1 Combined Heat and Power Plant of the Company IPF in Magdeburg. 2.2 The HotModule in Magdeburg. 2.3 Operation Experience. 2.4 Results and Outlook. Part II Model-based Process Analysis. 3 MCFC Reference Model (Peter Heidebrecht, and Kai Sundmacher). 3.1 Model Hierarchy. 3.2 General. 3.3 Model Equations. 3.3.1 Indirect Internal Reformer. 3.3.2 Anode Channel. 3.3.3 Combustion Chamber. 3.3.4 Reversal Chamber. 3.3.5 Cathode Channels. 3.3.6 Electrode Pores. 3.3.7 Solid Phase. 3.3.8 Electric Potential. 3.3.9 Reaction Kinetics. 3.3.10 Thermodynamics. 3.4 Summary. Bibliography. 4 Index Analysis of Models (Kurt Chudej, Hans Josef Pesch, and Joachim Rang). 4.1 Differential Time Index. 4.2 MOL Index. 4.3 Perturbation Index. 4.3.1 Transformation to Homogenous Dirichlet Boundary Conditions. 4.3.2 Abstract Problem. 4.3.3 Perturbation Index. 4.3.4 Garding-Type Inequality. 4.3.5 Estimate for v and v. 4.3.6 Estimate for u, w and w with Garding-Type Inequality. 4.4 Conclusion. Bibliography. 5 Parameter Identification (Matthias Gundermann and Kai Sundmacher). 5.1 Experimental Work. 5.1.1 Measurement of Cell Current and Cell Voltage. 5.1.2 Temperature Measurement. 5.1.3 Measurement of Concentrations. 5.1.4 Measurement of Flow Rates. 5.1.5 Conversion of the Measurements into Dimensionless Values. 5.1.6 Measurement Errors. 5.1.7 Measuring Campaigns. 5.2 Strategy for Parameter Estimation. 5.2.1 Determination of Relevant Parameters. 5.2.2 Balancing of the Fuel Cell Plant. 5.2.3 Sensitivity Analysis. 5.2.4 Parameter Estimation for a Single Load Case. 5.2.5 Parameter Estimation for the Whole Operating Range. 5.2.6 Temperature Dynamics. 5.3 Results of the Parameter Identification. 5.3.1 Steady State Measurements. 5.3.2 Plant Balancing and Error Minimisation. 5.3.3 Parameter Estimation. 5.3.4 Dynamic Measurements. 5.3.5 Estimation of the Solid Heat Capacity. 5.3.6 Evaluation of the Results. 5.4 Summary. Bibliography. 6 Steady State and Dynamic Process Analysis (Peter Heidebrecht, Matthias Gundermann, and Kai Sundmacher). 6.1 Steady State Simulation. 6.2 Current-Voltage Curve. 6.3 Transient Simulation. 6.4 Summary. Bibliography. 7 Hot Spot Formation and Steady State Multiplicities (Michael Krasnyk, Michael Mangold, Achim Kienle, and Kai Sundmacher). 7.1 Introduction. 7.2 Models Nonlinear Analysis. 7.2.1 Spatially Distributed Model. 7.2.2 Lumped Model. 7.3 Analysis of the Lumped FC Model. 7.4 Analysis of the Spatially Distributed FC Model. 7.5 Analysis of a More General High Temperature Fuel Cell Model. 7.6 Conclusions. 7.7 Appendix: Model Equations for Nonlinear Analysis. 7.7.1 Equations of the Spatially Distributed Model. 7.7.2 Equations of the Lumped Model. 7.3.3 Model Parameters. Bibliography. 8 Conceptual Design and Reforming Concepts (Peter Heidebrecht and Kai Sundmacher). 8.1 Steady State Anode Model. 8.1.1 General. 8.1.2 Equations. 8.1.3 Conversion Diagram. 8.2 Applications of the Steady State Anode Model. 8.2.1 Comparison of Reforming Concepts. 8.2.2 Fuel Cell Cascades. 8.2.3 Anode Exhaust Gas Recycling. 8.2.4 Fuel Gas Sidefeed. 8.3 Summary. Bibliography. Part III Optimization and Advanced Control. 9 Model Reduction and State Estimation (Markus Grotsch, Michael Mangold, Min Sheng, and Achim Kienle). 9.1 Introduction. 9.2 Development of a Nonlinear Reduced Model. 9.2.1 Choice of Basis Functions. 9.2.2 Treatment of Boundary Conditions. 9.2.3 Resulting Reduced Model of the HotModule. 9.3 Investigation of Observability. 9.4 Design of an Extended Kalman Filter. 9.5 Simulation Results. 9.6 Experimental Results. 9.7 Conclusions. Bibliography. 10 Optimal Control Strategies (Kati Sternberg, Kurt Chudej, and Hans Josef Pesch). 10.1 Model and Simulation Setting. 10.2 Mathematical Methods. 10.3 Optimal Control of a Load Change. 10.4 Summary and Conclusion. Bibliography. 11 Optimisation of Reforming Catalyst Distribution (Peter Heidebrecht and Kai Sundmacher). 11.1 Introduction. 11.2 Objective Functions and Optimisation Parameters. 11.3 Numerical Aspects. 11.4 Results. 11.4.1 Optimisation of Input Conditions at Constant Catalyst Density. 11.4.2 Optimisation of the Reforming Catalyst Density Distribution. 11.4.3 Optimisation of the Input Conditions for a System with Optimised Catalyst Density. 11.5 Summary. Bibliography. Appendices. A List of Symbols. B Benchmark Problem: Complete Set of Equations and Parameters (Peter Heidebrecht). B.1 Equations. B.2 Parameters. Index.