(234b) A Systematic Procedure for the Optimal Design of Operable and Flexible Reactive Distillation Systems
AIChE Annual Meeting
Monday, November 16, 2020 - 8:15am to 8:30am
Different optimisation strategies have been used for the design of reactive distillation systems. One of the most rigorous concepts is optimisation based on a process superstructure, a concept which was first introduced for the optimal design of simple distillation columns (Sargent and Gaminibandara 1976). This network representation method determines all the desired column design and operation specifications by solving a mathematically complex optimization problem. In this work, the concept of distillation column superstructures will be extended to reactive distillation systems in order to determine both structural and operational decisions including but not limited to: total number of stages, feed stage locations, existence of side draw streams, reactive zones etc. In addition to a single reactive distillation column, the synthesis of more complex reactive distillation processes, e.g. including a pre-reactor, a sequence of reactive/non-reactive distillation columns etc., will also be considered in order to investigate a large number of possible process alternatives.
The optimization problem considered is a Mixed-Integer Non-Linear Problem (MINLP), as the design variables are both continuous (e.g. reflux ratio) and integer (e.g. number of stages). Flowsheets with such a high level of complexity often suffer from initialisation issues as well as from the existence of a large number of local optima due to their non-convex nature. A demanding optimisation strategy is therefore employed to ensure that initialisation issues are tackled and to increase the possibility of locating the global optimum. The solution of the MINLP firstly determines the existence of the units and secondly, their design and operating parameters in case they exist. In this way, the optimal design of various process alternatives can be determined, thus allowing the selection of the most suitable process alternative based on the given objective function, in our work a cost function. The software used in this work is gPROMS ProcessBuilder (Process Systems Enterprise, 2020), selected due to its rigorous mathematical solvers and excellent optimization capabilities.
The optimisation strategy is demonstrated using a number of different chemical systems with varying difficulty in terms of the separation task (easy/difficult separations) and different kinetics (fast/slow reactions), and chemical equilibrium (towards products or reactants), in order to illustrate how the strategy can provide insight into which factor(s) (reaction, separation or their combination) is more dominant for the design and operation of the intensified process. The MINLP considered for each case aims to minimize the total annualised (capital and operational) cost of the process, subject to constraints such as required product purity.
As real plants are generally subject to disturbances, the controllability and operability of the process also need to be evaluated before settling on a process alternative. The control characteristics of the optimal steady state process found using the superstructure strategy are therefore initially investigated in the frequency- and time-domains by considering controllability indices and the performance of conventional control schemes under various disturbances which could potentially be introduced (e.g. feed flow rate disturbance), respectively. This investigation will indicate how operable the optimal design is, how reaction kinetics and separation parameters impact on the stability and controllability of a process and will establish whether any adjustments are needed in order to render it viable from a controllability perspective. The frequency domain analysis is performed using MATLAB (MATLAB 2020) and the time-domain analysis using gPROMS ProcessBuilder (Process Systems Enterprise, 2020).
The optimal design is obtained by the rigorous optimization of a mathematical model as described above. This model, however, depends on a number of model parameters, which are key for the performance of reactive distillation systems. A robust design should tolerate a certain level of parameter uncertainty otherwise it may be considered practically infeasible. For this reason, the final step of the methodology presented in this work is the investigation of the impact of uncertainty on the optimal design to determine exactly how sensitive the optimal process is to key model parameters and if needed, will revise the optimal design accordingly.
Based on a number of industrially-relevant case studies, this work will demonstrate how the optimal design, control performance and sensitivity to parameter uncertainty can be evaluated using a Process Systems Engineering approach for reactive distillation systems. Challenges around the formulation of the superstructure (flowsheet initialisation, software limitations etc.), as well as methods employed to overcome those, will be described. The importance of computational tools which can accommodate all steps of the investigation, and therefore guarantee consistency in the results supporting the development of systematic methodologies, will be highlighted.
In conclusion, this work will provide a systematic methodology for how to select the most suitable reactive distillation process design, based on various (simple and more complex) process configuration alternatives as expressed by the superstructure, that is operable as well as flexible enough to tolerate input uncertainty. A detailed insight into the economic and energy benefits (expressed by the optimal MINLP results) as well as into the potential operability/flexibility limitations (found using the controllability/uncertainty analysis) will be provided which may hopefully pave the way for an even greater acceptance of reactive distillation systems.
Harmsen, G. J. (2007). Reactive distillation: The front-runner of industrial process intensification: A full review of commercial applications, research, scale-up, design and operation, Chem. Eng. Process., 46, 774â780.
Kiss, A. A. (2013). Advanced Distillation Technologies: Design, Control and Applications, Wiley, Chichester.
Luyben, W. L. and Yu, C. C. (2008). Reactive Distillation Design and Control, Wiley, New Jersey.
Process Systems Enterprise, gPROMS, www.psenterprise.com/gproms, 2020.
Wang, S. J. et al. (2010). Plantwide Design of Ideal Reactive Distillation Processes with Thermal Coupling, Ind. Eng. Chem. Res., 49, 3262-3274.
MATLAB (2020). 18.104.22.1683654 (R2018a). Natick, Massachusetts, The MathWorks Inc.
Sargent, R. W. H. and K. Gaminibandara (1976). Optimum Design of Plate Distillation Columns. Optimization in Action;. London, Academic Press.