Determining solubility of CO2 in aqueous brine systems via hybrid smart strategies. (6th March 2021)

• Record Type: Journal Article
• Authors: Sayahi, Tofigh; Tatar, Afshin; Rostami, Alireza; Anbaz, Mohammad Amin; Shahbazi, Khalil
• Is Part Of: International journal of computer applications technology. Volume 65:Number 1(2021)
• Page Start: 1
• Page End: 13
• Publication Date: 2021-03-06
• Subjects: equilibrium CO2-water/brine -- solubility -- least squares support vector machine -- carbon capture and storage -- outliers analysis -- radial basis function neural network; Technology -- Data processing -- Periodicals; 620.00285
• Languages: English
• ISSNs: 0952-8091

Abstract

In this study, Radial Basis Function Neural Network (RBF-NN) and Least Square Support Vector Machine (LSSVM) were established for estimation of equilibrium CO2 -water/brine solubility as a function of salt molecular weight, temperature, salt molality and pressure. A reliable database was gathered from the open source literatures, and was split into two groups of testing and training subsets. Optimal structure of the proposed RBF-NN technique and the tuning coefficients of LSSVM model were determined by Cuckoo Optimisation Algorithm (COA). Accordingly, the proposed approaches here can accurately prognosticate CO2 solubility with determination factor (R²) of 0.9966 and average absolute relative deviation (AARD%) of 2.5885% for COA-LSSVM, and AARD% = 3.8832% and R² = 0.9962 for COA-RBF-NN; therefore, the proposed COA-LSSVM gives more accurate results for estimating CO2 solubility. Williams' outliers detection technique reveals that less than 3% of database are outliers. Salt molality is the most affecting variable based on sensitivity analysis.