Skip to main content
European Commission logo print header

Complex distillation columns

Exploitable results

Previous work has seen (1) the development of a short cut design technique for thermally coupled distillation systems and (2) initial studies on the control and operability aspects of the Dividing Wall Column (DWC). The objective was to improve the existing short cut design techniques to cater for more realistic industrial applications. UMIST developed a novel methodology for steady-state design optimization, to be applied to a wider range of problems, and multicomponent rather than ternary separations, the use of stripping steam and the other systems. The development of software supporting this new methodology was substantial contribution. UT studied control issues. Results show that the DWC can be controlled by conventional control structures. The optimal operation point is strongly dependent on feed composition and on the product specifications. When three product composition loops are closed, adjustment of liquid and vapour flow splits can be used to minimize energy consumption. From analysis of the shape of the solution surface can be seen that at least one of these inputs should be adjustable on-line in order to maintain close to minimum energy consumption when feed parameters may be varying. PARAGON has developed: a) A Fluid mechanics code for the modelling of the flow, b) An optimization code based on Genetic Algorithms for the optimization of the operating degrees of freedom, and c) An Artificial Neural Nets (ANN) for the modelling of the DWC. UPC and INFOR have developed an intelligent system framework supporting the DWC control system, including Fault Diagnosis. First, UPC performed several steady state and dynamic studies, leading to design and control strategies that optimize energy consumption and improve controllability indexes. A novel DWC dynamic model useful for forecasting and predictive control was designed. The model, based on ANN, can be continuously actualized by a genetic algorithm (evolutionary model) and has been integrated in the centralized control strategy of the whole DWC finally developed. ULg has studied the implementation of DWC in two processes. They extended the modelling capabilities of available simulation software to DWC: the heat transfer across the dividing wall can now be simulated accurately. The modified simulation software was used to carry out a detailed case study of an existing gas separation plant. MWKL, utilizing their wide base in the chemical, gas and refining industries, completed a rigorous analysis, using actual plant/design data. It demonstrated that DWC's provide a significant reduction in energy demand. The highest energy savings of 33 % were predicted for a refinery fractionation problem producing gasoline-blending components. ASTON completed work on the air-water simulator to determine the liquid distribution in a packed section in a DWC.

Searching for OpenAIRE data...

There was an error trying to search data from OpenAIRE

No results available