Design of non-equilibrium stage separation systems by a stochastic optimization approach for a class of mixtures

• A strategy for the design and optimization of non-equilibrium distillation sequences is proposed.
• A strategy is proposed for the design and optimization of non-equilibrium thermally coupled systems.
• Pareto fronts for equilibrium and non-equilibrium distillation sequences are obtained and analyzed.
• Equilibrium and non-equilibrium designs show similar heat duties but differences on configurations.

Designand analysis of distillation columns usually takes place by using rigorous equilibrium models, where efficiencies are utilized to enhance the designs obtained and have a better approach to the real performance of the column. A second alternative, based on the mass and heat transfer rates, has been proposed and it is known as the non-equilibrium (or rate-based) model. By employing this approach, the use of supposed values for efficiencies is not necessary. Nevertheless, to the authors’ knowledge, there is not a short-cut design model for analysis of distillation columns with the non-equilibrium model. In this work, the design and optimization of conventional and intensified distillation sequences, with equilibrium and non-equilibrium models for the separation of ternary mixtures, is presented. The multiobjective optimization is performed by a stochastic technique with handling constraints, which is also coupled to a process simulator. This strategy allows analyzing the distillation systems with the complete rigorous models (MESH and MERSHQ equations, respectively). Optimal designs are then compared to determinate relationships between the design parameters of the optimal equilibrium sequences and the optimal non-equilibrium sequences. It has been found that the optimal on heat duty for each analyzed systems is almost the same for both, equilibrium and non-equilibrium models, but with different number of stages. Nevertheless, other design variables have only slight differences between models for an optimal structure.