Machine Learning for Subsurface Characterization focuses on the development and application of neural networks, deep learning, unsupervised learning, reinforcement learning, and clustering methods for subsurface characterization under constraints due to financial, operational, regulatory, risk, technological and environmental challenges. The book introduces readers to methods of generating subsurface signals and analyzing the complex relationships within various subsurface signals using machine learning. Algorithmic procedures in MATLAB, R, PYTHON, and TENSORFLOW are displayed in text and through online instructional videos to assist training and learning. Field cases are also presented to demonstrate real-world applications, with a particular focus on examples involving shale reservoirs.
Explaining the concept of machine learning, advantages to the industry, and applications applied to complex subsurface rocks, this book delivers a missing piece for the reservoir engineer’s toolbox.
- Focuses on applying predictive modeling and machine learning from real case studies and Q&A sessions at the end of each chapter
- Teaches users how to develop codes, such as MATLAB, PYTHON, R and TENSORFLOW with step-by-step guides included
- Helps readers visually learn code development with video demonstrations
1. Unsupervised outlier detection techniques for well logs and geophysical data2. Unsupervised clustering methods for noninvasive characterization of fracture-induced geomechanical alterations3. Shallow neural networks and classification methods for approximating the subsurface in situ fluid-filled pore size distribution4. Stacked neural network architecture to model themultifrequency conductivity/permittivity responses of subsurface shale formations5. Robust geomechanical characterization by analyzing the performance of shallow-learning regression methods using unsupervised clustering methods6. Index construction, dimensionality reduction, and clustering techniques for the identification of flow units in shale formations suitable for enhanced oil recovery using light-hydrocarbon injection7. Deep neural network architectures to approximate the fluid-filled pore size distributions of subsurface geological formations8. Comparative study of shallow and deep machine learning models for synthesizing in situ NMR T2 distributions9. Noninvasive fracture characterization based on the classification of sonic wave travel times10. Machine learning assisted segmentation of scanning electron microscopy images of organic-rich shales with feature extraction and feature ranking11. Generalization of machine learning assisted segmentation of scanning electron microscopy images of organic-rich shales12. Characterization of subsurface hydrocarbon/water saturation by processing subsurface electromagnetic logs using a modified Levenberg-Marquardt algorithm13. Characterization of subsurface hydrocarbon/water saturation using Markov-chain Monte Carlo stochastic inversion of broadband electromagnetic logs
Siddharth Misra is currently associate professor at the Harold Vance Department of Petroleum Engineering, Texas A&M University, College Station, Texas. His research work is in the area of data-driven predictive models, machine learning, geosensors, and subsurface characterization. He earned a PhD in petroleum engineering from the University of Texas and a bachelor of technology in electrical engineering from the Indian Institute of Technology in Bombay. He received the Department of Energy Early Career Award in 2018 to promote geoscience research. Hao Li is a PhD-degree candidate in the Mewbourne College of Earth and Energy (MCEE) at the University of Oklahoma in Norman. He interned with Facebook on improving ranking models using machine learning. His research interests include machine learning, petrophysics, and data analytics. He holds an MS degree in petroleum engineering from China University of Petroleum in Beijing. Jiabo He is currently a doctoral candidate in computer science at the University of Melbourne, Australia. Jiabos research area includes deep learning, reinforcement learning, and imitation learning. He earned an MS in petroleum engineering from the University of Oklahoma and a BS in petroleum engineering from the China University of Petroleum in Beijing.
Lisainfo e-raamatute kohta