(79c) An Iteratively Refined Distillation Design Method
The focus of this talk is distillation design feasibility. It is shown that all current shortcut methods for determining feasibility (e.g., the boundary value design procedure of Doherty and co-workers, the shortest stripping line approach of Lucia et al., and methods that use rectification bodies such as the feed pinch method of Marquardt and co-workers) suffer from mass balance errors. These methods typically over-specify the design problem and fail to satisfy either component mass balances around the feed stage or around the column. As a result, they can generate designs that are declared feasible when they are really infeasible or produce feasible (over)designs that waste energy.
To resolve mass balance errors, a novel iterative refinement procedure based on direct substitution is proposed within the distillation line method of Lucia et al., which is a bottom up (or top down) strategy. Iterative refinement 1) Automatically adjusts the reflux (or boil-up) ratio and all compositions in one product stream to determine feasibility. 2) Can be combined with the concept of shortest stripping line distance and find feasible minimum energy designs. 3) Can identify infeasible column specification sets. The equations of iterative refinement are briefly summarized and some discussion of implementation, robustness and computational efficiency are presented. Fourteen literature examples are used to illustrate the efficacy of iterative refinement. These numerical results clearly show that iterative refinement is a powerful procedure that is capable of finding feasible designs other methods cannot, often resulting in significant reductions in energy requirements. All designs determined by iterative refinement are validated using the Aspen Plus simulator. Many examples and geometric illustrations are used to elucidate key attributes of iterative refinement.