Abstract
Propylene glycol monomethyl ether acetate (PMA) is a critical solvent in industrial applications, recognized for its high solvency, low corrosiveness, minimal toxicity, and thermal stability. It is produced via the transesterification of propylene glycol methyl ether (PM) and methyl acetate (MEAC) using sodium methoxide as a catalyst. Since the reaction is reversible and limited by equilibrium, reactive distillation (RD) is used to improve the conversion. However, methanol (MEOH), a by-product, forms an azeotrope with methyl acetate, requiring separation by pressure-swing distillation (PSD). The conventional reactive distillation system is typically optimized using sequential iterative optimization (SIO), but this method may not fully maximize optimization potential for systems with complex internals. This study used a genetic algorithm (GA) to optimize the production system by adjusting multiple design variables simultaneously. The process was automated by integrating Aspen Plus® with MATLAB®, aiming to minimize the total annual cost (TAC). The genetic algorithm’s elimination mechanism ensured efficient steady-state designs, while the flexibility index (FI) accounted for operational uncertainties. By combining the objectives of minimizing total annual cost and maximizing flexibility index, the study aimed to identify the most cost-effective and safest design parameters for the production system, resulting in TACFI. The results showed a 48.62 % reduction in TAC as the evolutionary generations progressed. For the bi-objective optimization, TAC was reduced by 41.073 %, TACFI by 97.719 %, and the flexibility index improved from 0.027 to 1.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call