Abstract

In the present article, an optimization approach for the synthesis of reactor networks is proposed. A stage-wise superstructure containing a continuous stirred tank reactor (CSTR) and a plug flow reactor (PFR) is used to formulate the mathematical model without the use of differential equations and with a mixed-integer nonlinear programming (MINLP) formulation. To solve the optimization problem a bi-level meta-heuristic approach combining simulated annealing (SA) and particle swarm optimization (PSO) is used. Two case studies involving complex nonisothermal homogeneous reaction systems were used to test the applicability of this technique, and a methodology based on temperature profiles was developed to guide the search for operating temperatures during the optimization. The results achieved improvements in the objective function when compared to the majority of results reported in the literature, which demonstrates that the developed methodology is suitable to perform the reactor network synthesis in homogeneous systems.

摘要

在本文中，提出了一种用于合成反应器网络的优化方法。包含连续搅拌釜反应器 (CSTR) 和活塞流反应器 (PFR) 的分级上部结构用于在不使用微分方程和混合整数非线性规划 (MINLP) 公式的情况下制定数学模型。为了解决优化问题，使用了结合模拟退火 (SA) 和粒子群优化 (PSO) 的双层元启发式方法。使用涉及复杂非等温均相反应系统的两个案例研究来测试该技术的适用性，并开发了一种基于温度分布的方法来指导在优化过程中搜索操作温度。