Investigation of carbon dioxide capture and storage by a novel LSSVM-GA method

Abstract

Accurate knowledge about vapor–liquid equilibrium (VLE) and phase behavior of impure carbon dioxide has noticeable effect on designing carbon dioxide capture and storage (CCS) processes. In the current work, a novel computational method namely least square support vector machine (LSSVM) was used to predict VLE of binary system of carbon dioxide with nitrogen, hydrogen, oxygen, and argon. To this end, two models based on LSSVM were developed for vapor and liquid phases. The accuracy of predicting models was investigated by different methods. The results express that the proposed algorithms are dependable for the estimation of phase behavior of impure CO2.