Hybrid modeling distillation column

In Greaves et al. (2001), a hybrid model for an actual pilot plant batch distillation column is developed. However, taking advantage of some of the inherent properties of batch distillation processes a simpler version (new algorithm) of the general optimisation framework is developed to find optimal reflux ratio policies which minimises the batch time for a given separation task. 

Hybrid Model Development for Pilot Batch Distillation Column... 

In Greaves et al. (2001) and Greaves (2003), instead of using a rigorous model (as in the methodology described above), an actual pilot plant batch distillation column is used. The differences in predictions between the actual plant and the simple model (Type III and also in Mujtaba, 1997) are defined as the dynamic process-model mismatches. The mismatches are modelled using neural network techniques as described in earlier sections and are incorporated in the simple model to develop the hybrid model that represents the predictions of the actual column. 

For the study of the process, a set of partial differential model equations for a flat sheet pervaporation membrane with an integrated heat exchanger (see fig.2) has been developed. The temperature dependence of the permeability coefficient is defined like an Arrhenius function [S. Sommer, 2003] and our new developed model of the pervaporation process is based on the model proposed by [Wijmans and Baker, 1993] (see equation 1). With this model the effect of the heat integration can be studied under different operating conditions and module geometry and material using a turbulent flow in the feed. The model has been developed in gPROMS and coupled with the model of the distillation column described by [J.-U Repke, 2006], for the study of the whole hybrid system pervaporation distillation. 

For the model validation and the analysis of the heat integration in the hybrid pervaporation distillation process, a laboratory plant has been built at the TU -Berlin and prepared for the connection with the distillation column (see fig. 3). With this plant experiments with a flat PVA-based (Polyvinylalcohol from GKSS) hydrophilic membrane have been done. A heat exchanger has been built within the pervaporation module. The temperature in the heat exchanger has been necessary to avoid the temperature drop between feed and retentate streams in the pervaporation process. In the process a 2-Propanol/ Water mixture has been separated. The concentration of 2-Propanol in the feed is between 80 and 90 % in weight and the temperature range in the experiments was between 70 and 90°C. The feed flow is turbulent and the system fully insulated to avoid heat looses. The pressure in the permeate side has been kept at 30 mbar and the feed pressure at 1.5 bar. 

More efficient is a generic tree representation of the separations based on tasks. The sequencing can be formulated as a structural optimisation problem where standard techniques based on Mixed Integer Linear Programming (MILP) apply. The tasks consist of simple distillation columns, as well as of hybrids for complex column arrangements, modelled by appropriate shortcut or semi-rigorous methods. Details can be found in Doherty and Malone (2001). 

PrODHyS (Process Object Dynamic Hybrid Simulator) is an object-oriented environment for the hybrid dynamic simulation of chemical processes. The aim of this paper is to introduce a small part of the main reusable object components available in this library and to illustrate its potentialities through the modeling and the simulation of a reactive distillation column. 

For a fundamental understanding of the hybrid process it is necessary to describe the interactions between two different unit operations with appropriate models. Making basic parameter studies the equilibrium stage model for distillation and a short-cut model for membrane separation is sufficient. The models are well established and the model parameters are quite accessible. This combination gives an first survey on the influence of structural and operational parameters on the concentration profiles in the column and on the maximum amount of water, which can be removed. 

---