Exergoeconomic Distillation Sequencing by Multi-Objective Optimization Through a Hybrid Genetic Algorithm

Abstract

While trying to optimize sharp distillation processes, the number of possible column sequences significantly increases as the number of components that make up the feed mixture increases. As a result, proper sequencing with maximum exergetic profit and minimum exergy destruction becomes harder to achieve. In this study, an exergoeconomic multi-objective optimization was applied to the distillation sequences of three separate hydrocarbon mixture cases, by means of a genetic-algorithm-based solver software. A computer program (DISMO) was developed in-house to achieve this functionality. The results indicate that the created algorithm was quite applicable in determining the optimum sequencing in distillation, as it successfully created the Pareto Solution Set and suggested the optimum configurations. This study also presented an opportunity to conduct a parametric investigation on various weighting factors for objective functions. As the importance given to a specific objective was increased, the optimization results had a tendency to favour that specific objective through arrangement of sequencing as expected, though the profit and sequencing converged to a single result after a certain threshold.