A critical review of physics-informed machine learning applications in subsurface energy systems

• Published in: Geoenergy Science and Engineering Vol. 239; p. 212938
• Main Authors: Latrach, Abdeldjalil, Malki, Mohamed L., Morales, Misael, Mehana, Mohamed, Rabiei, Minou
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.08.2024
• Subjects: Deep learning; Energy; Petroleum engineering; Physics-informed machine learning; Subsurface; Deep learning; Physics-informed machine learning; Energy; Subsurface; Petroleum engineering
• ISSN: 2949-8910; 2949-8910
• DOI: 10.1016/j.geoen.2024.212938

Machine learning has emerged as a powerful tool in various fields, including computer vision, natural language processing, and speech recognition. It can unravel hidden patterns within large data sets and reveal unparalleled insights, revolutionizing many industries and disciplines. However, machine and deep learning models lack interpretability and limited domain-specific knowledge, especially in applications such as physics and engineering. Alternatively, physics-informed machine learning (PIML) techniques integrate physics principles into data-driven models. By combining deep learning with domain knowledge, PIML improves the generalization of the model, abidance by the governing physical laws, and interpretability. This paper comprehensively reviews PIML applications related to subsurface energy systems, mainly within the context of the oil and gas industry. The review highlights the successful utilization of PIML for tasks such as seismic applications, reservoir and fluid flow simulation, and hydrocarbons production forecasting. Additionally, it demonstrates PIML’s capabilities to revolutionize the oil and gas industry and other emerging areas of interest, such as carbon and hydrogen storage, by providing more accurate and reliable predictions for resource management and operational efficiency. •Machine Learning advances in scientific fields face limitations in scientific research.•Issues include interpretability, data needs, and adherence to physical laws.•Physics-informed machine learning (PIML) integrates domain knowledge to guide training.•Neural simulators in PIML, like neural solvers/operators, are compared to traditional methods.•Study reviews PIML in subsurface energy systems, focusing on oil and gas industry applications.