Optimization of industrial carbon capture process using gradient, weight-based, and lexicographic approach

Abstract

Natural gas (NG) is a fossil energy source and a crucial petrochemical feedstock. Raw NG contains impurities that must be removed before it can be commercially used. Carbon capture (CC) is considered a crucial step in the NG treatment process. The removal of acid gases like hydrogen sulphide (H₂S) and carbon dioxide (CO₂) from NG is of great importance. Optimization of the industrial CC process is carried out using environment, process, and energy-based objectives, five decision variables, and two constraints. Six different optimization algorithms are utilized for each of the objective functions, and their detailed convergence-specific comparison is carried out using the corresponding objective and the decision variable's values. In a gradient optimization study, the minimum energy value of 13.35 MMBtu/h is achieved by the Interior Point-Central difference algorithm. In a weight-based study, as weight increases from 0 to 0.4, the CO₂ in sweet NG decreases and remains nearly constant at an average of 8038 ppm, and the hydrocarbon recovery first decreases and remains constant at the value of 92.4 %. In a lexicographic optimization study, the total energy optimum value increases with an increase in compromise percentage, with a maximum of 8.1 % with 10 % compromise, whereas the CO₂ content in sweet NG objective values decreases by up to 7 %. This optimization study gives insight into the complex natural gas CC process using traditional optimization algorithms.