Optimization of a Stripper Unit for the Desorption of MEA From (MEA-H2O-CO2) System

Abstract

The aim of this paper is to develop an optimization NLP mathematical model to determine the best operating conditions for the amine regeneration unit of the CO2 post-combustion process. The amine regeneration is an energy intensive process and its optimization is the most important key to obtain cost-effective designs. Thus, an objective function defined as the ratio between the heating duty and the mass of CO2 captured is proposed for minimization. The resulting model involves a high number of non-lineal constraints given by the mass and energy balances and the specific correlations used to compute physical-chemical properties. In addition, bilinear terms are also considered which lead to non-convex constraints.
Also, an initialization phase is proposed to solve the model with the aim of dealing with convergence difficulties. In this way, the model and the initialization strategy resulted to be robust and flexible enough to allowing perform all optimization cases without computational problems. Optimization results including a sensitive analysis are discussed through different case studies.