Physics-informed gaussian process regression for reproducible and uncertainty-aware CO2 injectivity prediction
PLOS One · 01.07.2026 · 21 (7) · e0352178
DOI: 10.1371/journal.pone.0352178 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 01.07.2026. Verfasst von Shamsuddeen Adamu und 11 weiteren Autor:innen. Beteiligte Einrichtung: Computer Information Science Department, Universiti Teknologi PETRONAS, Seri Iskandar, Malaysia. Hauptthemen laut MEDLINE: Carbon Dioxide. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Scientific Research, King Khalid University, Yayasan UTP.
Abstract (Originalsprache)
Reliable prediction of CO 2 injectivity decline is essential for safe geological carbon storage, yet existing machine learning models often provide deterministic point predictions that lack uncertainty quantification. This paper presents a physics-informed Gaussian process regression (PC-GPR) framework for Relative Injectivity Change (RIC) prediction, embedding constraints derived from two independently grounded physical laws: the Derjaguin–Landau–Verwey–Overbeek (DLVO) colloidal monotonicity condition and the Civan–Kozeny–Carman permeability impairment model. Four GP variants are developed and benchmarked on a curated laboratory dataset ( n = 44) under a three-tier validation protocol combining Leave-One-Out cross-validation, repeated k -fold cross-validation, and non-parametric bootstrap confidence intervals. Two complementary uncertainty quantification mechanisms are employed: GP posterior calibration via the Expected Calibration Error (ECE) and split-conformal prediction intervals. The GP-Base model achieves strong predictive performance (LOO R 2 = 0.9401, 95% CI: [0.882, 0.978]) with well-calibrated uncertainty (ECE = 0.026) and reliable coverage (97.7% at the nominal 95% level). The PC-GPR-M variant reduces DLVO monotonicity violations to 1.5% across the input domain, demonstrating effective soft constraint enforcement. Operationally, the proposed framework translates predictive uncertainty into actionable injection scheduling guidance, identifying high-risk regions at salinity >30,000 ppm and jamming ratio >0.04. These results provide an uncertainty-aware baseline for future PIML research in subsurface carbon storage.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Computer Information Science Department, Universiti Teknologi PETRONAS, Seri Iskandar, Malaysia.
- IAIICT, Ahmadu Bello University, Zaria, Nigeria.
- King Khalid University, Applied College of Muhayil Asir, Abha, Saudi Arabia.
- Department of Information Systems, College of Economics, Management and Information Systems, University of Nizwa, Nizwa, Oman.
- Environmental Engineering Department, University of Nizwa, Nizwa, Oman.
- Department of Cyber Security, Faculty of Computing, Air Force Institute of Technology, AFIT, Kaduna, Nigeria.
ORCID-Profile: Shamsuddeen Adamu
Förderung
- Deanship of Scientific Research, King Khalid University
- Yayasan UTP
Zitieren
APA
Adamu, S., Alhussian, H., Abdulkadir, S. J., Eltayeb Eltahir, M. M., Khairy, S. O. F., Hayder, G., Lathbl, M. A., Salisu Mohammed, H., Adekunle, A. O., Kamaldeen, G., Abdullahi, S. M., & Saidu, Y. (2026). Physics-informed gaussian process regression for reproducible and uncertainty-aware CO2 injectivity prediction. PLOS One, 21(7), e0352178. https://doi.org/10.1371/journal.pone.0352178
Vancouver
Adamu S, Alhussian H, Abdulkadir SJ, Eltayeb Eltahir MM, Khairy SOF, Hayder G, et al. Physics-informed gaussian process regression for reproducible and uncertainty-aware CO2 injectivity prediction. PLOS One. 2026;21(7):e0352178. doi:10.1371/journal.pone.0352178
BibTeX
@article{Adamu2026,
author = {Adamu, Shamsuddeen and Alhussian, Hitham and Abdulkadir, Said Jadid and Eltayeb Eltahir, Majdy Mohamed and Khairy, Sallam O F and Hayder, Gasim and Lathbl, Mahdi Ali and Salisu Mohammed, Hassan and Adekunle, Abdulrazak Oladeji and Kamaldeen, Gbolagade and Abdullahi, Samaila Musa and Saidu, Yahaya},
title = {Physics-informed gaussian process regression for reproducible and uncertainty-aware CO2 injectivity prediction},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0352178},
doi = {10.1371/journal.pone.0352178}
}
EUVIMED