Global Optimization of Reactive Distillation Processes Using Bat Algorithm

Abstract

Reactive distillation (RD) is an important process intensification approach with several advantages. It can improve the reaction selectivity and yield, overcome the thermodynamic restrictions, and reduce the cost/energy. However, the optimal design of RD relies on highly nonlinear and multivariable optimization including continuous and integer design variables. The objective function is generally non-convex with several constraints. For this problem, the conventional derivative-based optimization algorithms are faced with convergence problems and fail to guarantee the global optimal solution.
Stochastic optimization algorithms appear to be a better alternative for the optimal design of RD because of the high robustness and efficiency. Bat algorithm (BA), which combines advantages of other existing algorithms, is a potential stochastic optimization algorithm. In this work, the BA was used to optimize the RD for the production of methyl acetate (MeAc). The link between Matlab and Aspen Plus was also created to ensure that each solution was provided from rigorous simulations. The total annual cost (TAC) was set as the objective function. Product purity constraints were achieved through Aspen plus instead of algorithms to simplify the process. BA can find the global optimal solution within less computation time than other stochastic algorithms or sequential optimization.