Optimal design of complex distillation system for multicomponent zeotropic separations

The optimal design of complex distillation system for separation of multicomponent zeotropic mixtures is studied. Super column with the ability to separate any specified products from the given feed is introduced for defining integration upper limit of design alternatives. The improved state-space (SS) superstructure incorporating all basic mass and heat transfer elements are adopted to capture all configurations in the framework of super column. Specifically, by adding stages-cascade process operator in the original representation, a series of optimal flowsheets with multiple thermal links, which have never been included within previous superstructure are easily generated. The mathematical modeling of the proposed superstructure is performed with mixed integer non-linear programming (MINLP). It advocates the use of rigorous physical model for each mass/heat transfer stage to ensure the practical reliability and optimality of the attained designs. Then the derived optimization model is solved by a modified solution procedure, in which the key item is an iterative initialization scheme. Three multicomponent zeotropic separation examples are employed to illustrate the effectiveness of the proposed approach. These optimum designs yield significant savings in total annual cost (TAC) relative to Petlyuk columns and some guidelines for distillation flowsheet retrofit based on thermodynamic analysis are proposed.

► Optimal design of complex distillation system for zeotropic mixtures.
► Rigorous model is adopted for each mass/heat transfer stage.
► Our designs even yield significant cost savings relative to Petlyuk columns.
► Guidelines for distillation retrofit based on exergy loss analysis are proposed.