Generalized Disjunctive Programming for Optimum Distillation Column Sequence: a comparison with genetic algorithm and heuristic methods

An optimization systematic framework is inroduced in this work based on superstructure optimization with mathematical programming for the synthsis of optimum distillation column sequences that is a challenging problem, due to its huge search space. The proposed optimization models are obtained from a systematic modeling procedure where generalized disjunctive programming models are derived using the state task network (STN) superstructure representations. This method has been implemented in the LOGMIP, a GAMS-based software. The results obtained for optimum sequence is compared with genetic algorithm and heuristic method for a five component typical problem.