Batch processing has received renowned attention over the past decade particularly in the low-volume, high-value-added fine chemical and pharmaceutical industries. The importance of batch processes is increasing due to increasing cost pressures associated with over-capacity of high volume continuous plants and the ongoing preference of custom-made rather than commodity chemicals. Within these industries, batch distillation remains the most commonly used technique for separating liquid mixtures despite being an energy and capital intensive process. Many mixtures commonly encountered in the fine chemical and pharmaceutical industries are, however, difficult or impossible to separate by normal distillation due to azeotropic behaviour, tangent pinch or low relative volatilities. Pervaporation has been hailed as an alternative to distillation for such mixtures as the separation mechanism is different, relying on differences in solubility and diffusivity between the components in the mixture and not vapour-liquid equilibrium as in distillation. Recently, hybrid processes have been proposed where a distillation column unit and a pervaporation unit are integrated into one process. In such a process, the shortcomings of one method are balanced by the benefits of the other, allowing for significant savings in terms of energy consumption and cost. Single-objective optimisation of design and operation of hybrid batch distillation/pervaporation processes has been attempted before. This paper, however, presents results of the novel application of multi-objective optimisation for the design and operation of such processes.

Batch distillation processes, as most batch processes, are highly complex in nature and involve a number of operational and economical objectives that need to be considered. The performance of these processes depends on a number of different criteria that are often conflicting. An effective optimisation of such systems therefore requires the consideration of multi-criteria approaches to accommodate for the multi-objective nature of the problem and to effectively evaluate and optimise the performance in order to explore and understand the trade-offs between the multiple objectives (e.g. through the generation of a Pareto set).

Dedieu et al. (2003) considered the multi-objective design and retrofit of batch plants using a multi-objective genetic algorithm (MOGA). Their MOGA consisted of a combination of a single objective genetic algorithm and a Pareto sort procedure to optimise investment costs whilst at the same time minimising equipment sizing. Silva and Biscaia (2003) considered a constrained optimisation through a fuzzy penalty function of batch polymerisation reactors using a MOGA. They considered the maximisation of monomer conversion rate while minimising the concentration of initiator residue in the polymerisation reactor product.

This paper considers the simultaneous multi-objective optimisation of design and operation of hybrid batch distillation/pervaporation processes. The overall problem is formulated as a multi-objective mixed integer dynamic optimisation (MO-MIDO) problem. The multiple objectives include economic indices that reflect equipments capital investment, operating costs and production revenues. Several case studies for the separation of homogeneous binary mixtures are presented for a dual-criteria optimisation case in which the trade-offs between production and investment costs are explored. It is found that the genetic algorithm based multi-objective optimisation method is robust and able to adequately handle the multi-objective nature of the simultaneous design and operation of batch distillation columns.

References

1. Dedieu, S., L. Pibouleau, C. Azzaro-Pantel and S. Domenech, Design and retrofit of multiobjective batch plants via a multicriteria genetic algorithm, Computers & Chemical Engineering, 27(12), 1723-1740, 2003.

2. Silva, C. and E. Biscaia, Jr, Genetic algorithm development for multi-objective optimization of batch free-radical polymerization reactors, Computers & Chemical Engineering, 27(8-9), 1329-1344, 2003.