Enhancing wettability prediction in the presence of organics for hydrogen geo-storage through data-driven machine learning modeling of rock/H2/brine systems

Authors

Zeeshan Tariq
Muhammad Ali
Nurudeen Yekeen, Edith Cowan UniversityFollow
Auby Baban, Edith Cowan UniversityFollow
Bicheng Yan
Shuyu Sun
Hussein Hoteit

Document Type

Journal Article

Publication Title

Fuel

Volume

354

Publisher

Elsevier

School

School of Engineering

RAS ID

61919

Comments

Tariq, Z., Ali, M., Yekeen, N., Baban, A., Yan, B., Sun, S., & Hoteit, H. (2023). Enhancing wettability prediction in the presence of organics for hydrogen geo-storage through data-driven machine learning modeling of rock/H2/brine systems. Fuel, 354, article 129354. https://doi.org/10.1016/j.fuel.2023.129354

Abstract

The success of geological H2 storage relies significantly on rock–H2–brine interactions and wettability. Experimentally assessing the H2 wettability of storage/caprocks as a function of thermos-physical conditions is arduous because of high H2 reactivity and embrittlement damages. Data-driven machine learning (ML) modeling predictions of rock–H2–brine wettability are less strenuous and more precise. They can be conducted at geo-storage conditions that are impossible or hazardous to attain in the laboratory. Thus, ML models were utilized in this research to accurately model the wettability behavior of a ternary system consisting of H2, rock minerals (quartz and mica), and brine at different operating geological conditions. The results revealed that the ML models accurately captured the wettability behavior at different geo-storage conditions by yielding less than 5% mean absolute percent error and above 0.95 coefficient of determination values. The partial dependency or sensitivity plots were generated to evaluate the impact of individual features on the trained models. These plots revealed that the models accurately captured the physics behind the problem. Furthermore, a mathematical equation is derived from the trained ML model to predict the wettability behavior without using any ML software. The accuracy of the predictions of the ML model can be beneficial for exactly predicting the H2 geo-storage capacities and assessing of H2 containment security of storage and caprocks for large-scale geo-storage projects.

DOI

10.1016/j.fuel.2023.129354

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.