Modeling solubility of CO2–N2 gas mixtures in aqueous electrolyte systems using artificial intelligence techniques and equations of state

Abstract

Determining the solubility of non-hydrocarbon gases such as carbon dioxide (CO2) and nitrogen (N2) in water and brine is one of the most controversial challenges in the oil and chemical industries. Although many researches have been conducted on solubility of gases in brine and water, very few researches investigated the solubility of power plant flue gases (CO2–N2 mixtures) in aqueous solutions. In this study, using six intelligent models, including Random Forest, Decision Tree (DT), Gradient Boosting-Decision Tree (GB-DT), Adaptive Boosting-Decision Tree (AdaBoost-DT), Adaptive Boosting-Support Vector Regression (AdaBoost-SVR), and Gradient Boosting-Support Vector Regression (GB-SVR), the solubility of CO2–N2 mixtures in water and brine solutions was predicted, and the results were compared with four equations of state (EOSs), including Peng–Robinson (PR), Soave–Redlich–Kwong (SRK), Valderrama–Patel–Teja (VPT), and Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT). The results indicate that the Random Forest model with an average absolute percent relative error (AAPRE) value of 2.8% has the best predictions. The GB-SVR and DT models also have good precision with AAPRE values of 6.43% and 7.41%, respectively. For solubility of CO2 present in gaseous mixtures in aqueous systems, the PC-SAFT model, and for solubility of N2, the VPT EOS had the best results among the EOSs. Also, the sensitivity analysis of input parameters showed that increasing the mole percent of CO2 in gaseous phase, temperature, pressure, and decreasing the ionic strength increase the solubility of CO2–N2 mixture in water and brine solutions. Another significant issue is that increasing the salinity of brine also has a subtractive effect on the solubility of CO2–N2 mixture. Finally, the Leverage method proved that the actual data are of excellent quality and the Random Forest approach is quite reliable for determining the solubility of the CO2–N2 gas mixtures in aqueous systems.

High-pressure electrochemistry: a new frontier in decarbonization

“…For example, CO 2 solubility starts deviating from linear behavior from pressures exceeding 60 bar. 43 Another example is the gases that exist as a vapor (e.g., H 2 O). These vapor gases follow Henry's law only up to one-half to two-thirds of their saturation point.…”

“…For example, CO 2 solubility starts deviating from linear behavior from pressures exceeding 60 bar. 43 Another example is the gases that exist as a vapor ( e.g. , H 2 O).…”

Modeling the solubility of light hydrocarbon gases and their mixture in brine with machine learning and equations of state

“…The following statistical factors viz., determination coefficient (R 2 ), average absolute percent relative error (AAPRE), root mean square error (RMSE), and standard deviation (SD) were employed to assess the accuracy of the machine learning models. The mathematical formula of these statistical criteria is defined below 96 , 97 : where N refers to the count of data, η i,exp shows the experimental hydrocarbon gases solubility, and η i,pred is predicted hydrocarbon gases solubility in the liquid phase by presented models.…”

“…The following statistical factors viz., determination coefficient (R 2 ), average absolute percent relative error (AAPRE), root mean square error (RMSE), and standard deviation (SD) were employed to assess the accuracy of the machine learning models. The mathematical formula of these statistical criteria is defined below 96,97 :…”

Toward Predicting Interfacial Tension of Impure and Pure CO2-Brine Systems Using Robust Correlative Approaches

“…Utilizing the alternating condition expectation algorithm, IFT predictions spanning a broad pressure and temperature range of 0.1–60 MPa and 5.25–175 °C were conducted by Li et al However, despite the model’s intricacy, it shows a lack of accuracy, coupled with an error exceeding 10%, emphasizing the critical necessity for the development of a more dependable predictive approach. Therefore, researchers explored artificial intelligence modeling due to its capacity to effectively represent intricate systems encompassing diverse included parameters. , Zhang et al employed a neural network to model CO 2 -brine IFT, utilizing a database containing a total of 1716 data points. Kamari et al employed the identical database of 1716 data points, utilizing multiple machine learning (ML) models to establish a predictive model based on four input variables, including temperature, pressure, monovalent cation molality, and divalent cation molality for CO 2 -brine IFT.…”