Open Access
PROCEEDINGS
Fast and Accurate Calculation on Competitive Adsorption Behavior in Shale Nanopores by Machine Learning Model
1 Department of Modern Mechanics, University of Science and Technology of China, Hefei, 230027, China
* Corresponding Author: Hao Yu. Email:
The International Conference on Computational & Experimental Engineering and Sciences 2024, 30(2), 1-1. https://doi.org/10.32604/icces.2024.011120
Abstract
Understanding the competitive adsorption behavior of CO2 and CH4 in shale nanopores is crucial for enhancing the recovery of shale gas and sequestration of CO2, which is determined by both the inherent characteristics of the molecules and external environmental factors such as pore size, temperature, and partial pressures of CO2 and CH4. While the competitive adsorption behavior of CO2/CH4 has been analyzed by previous studies, a comprehensive understanding from the perspective of molecular kinetic theory and the efficient calculation for competitive adsorption behavior considering various geological situations is still challenging, limited by the huge computation cost of classical molecular dynamics (MD) methods. In this work [1], the theoretical connection between inherent characteristics of molecules and adsorption behavior is firstly built to reveal the general laws in the behavior of CO2/CH4 competitive adsorption through posture analysis of the molecules. A machine learning algorithm, aided by molecular kinetic theory, is proposed to facilitate the fast and accurate predictions of competitive adsorption behavior, and detailed analyses of the influencing factors are conducted accordingly. The insights gained from this work provide a foundation for expeditiously optimizing the competitive adsorption behavior of CO2/CH4, with potential implications for CO2 sequestration and enhanced gas recovery (CSEGR) process.Keywords
Cite This Article
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.