Compositional modeling of solution gas–oil ratio (Rs): a comparative study of tree-based models, neural networks, and equations of state

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: March 11, 2025

Accurate knowledge of crude oil pressure–volume–temperature (PVT) properties is essential for both industrial and academic applications. However, traditional experimental methods determining these properties, particularly the solution gas–oil ratio (Rs), are time-intensive costly. In this study, advanced compositional models were developed using a broad range machine learning (ML) techniques to predict Rs efficiently reliably. A comprehensive database 1,154 data points was utilized modeling. Among tested models, extra trees (ET) algorithm demonstrated superior performance, achieving an average absolute percent relative error (AAPRE) approximately 3%, indicating its high reliability prediction. Additionally, estimated seven different equations state (EoS). Systematic graphical statistical evaluations revealed that Schmidt-Wenzel (SW) EoS most accurate among conventional methods, with 11%. The robustness ET validated across various temperature ranges, detailed trend analysis confirming their ability accurately capture physical relationship between pressure. relevancy factor quantified influence each input parameter on model outputs, whereas Leverage technique identified outliers defined ranges optimal performance. While ML achieved predictive reliability, computational demands complexity may limit deployment in resource-constrained environments decision-critical Nevertheless, study represents significant advancement modeling, providing robust, scalable, cost-effective tools

Language: Английский

Accurate modeling of nano-enhanced polyethylene glycol thermal conductivity using soft computing methods: application to thermal energy storage

International Journal of Polymer Analysis and Characterization, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 28

Published: April 7, 2025

Language: Английский