H042
Data Science for Discovery and Prediction in Porous Media

Session ID#: 77640

Session Description:
Data science brings computer science, statistics, and applied mathematics together to discover new knowledge of a complex system. Machine learning (ML) and artificial intelligence (AI) methods have shown capabilities in enhancing predictive understanding of such systems. Predicting and controlling transport phenomena through porous, granular materials is critical to a wide variety of applications. Accurate estimation of properties such as porosity and permeability is vital for improvement of many geo-related applications including management of water supplies, selection of cost-effective production schemes, CO2sequestration, developing geothermal energy schemes, etc. This session provides a forum for researchers to exchange ideas on advancing the state-of-the-art through data science for discovering phenomena in porous media. Topics within the scope of interests include but are not limited to (1) innovative applications of modern data science approaches in porous media, (2) use of ML techniques for predicting physical and chemical properties in geological media, and (3) data mining.
Index Terms:

0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
1805 Computational hydrology [HYDROLOGY]
1858 Rocks: chemical properties [HYDROLOGY]
1859 Rocks: physical properties [HYDROLOGY]

Primary Convener:  Pania Newell, University of Utah, Mechanical Engineering, Salt Lake City, UT, United States
Conveners:  Dan Lu, Oak Ridge National Laboratory, Oak Ridge, TN, United States and Wei Jia, The University of Utah, Salt Lake City, UT, United States
Primary Liaison:  Pania Newell, Sandia National Laboratories, Albuquerque, NM, United States
 
Physics-Informed AI Modeling and Forecasting in Porous Media (Invited) (487393)
Hector Klie, DeepCast.ai, CEO, Houston, TX, United States and Bicheng Yan, DeepCast.ai, Data Science, Houston, TX, United States
 
Adversarial Learning for Subsurface Flow and Transport Modeling (Invited) (489465)
Alexander Y Sun, Zhi Zhong and Hoonyoung Jeong, University of Texas at Austin, Austin, TX, United States
 
Linking Pore Network Structure to Preferential Pathway Development: Differential Evolution & Shortest Path Analysis (491669)
Zoe Kanavas, University of California Davis, Civil and Environmental Engineering, Davis, CA, United States and Veronica L Morales, University of California Davis, Davis, United States
 
A Deep Learning Based Approach for the Dynamic Reservoir Property Changes Estimation by Using Time-lapse Seismic Data (507655)
Zhi Zhong and Alexander Y Sun, University of Texas at Austin, Austin, TX, United States
 
Fast permeability estimation for sandstone and carbonate reservoir rocks by 3D Bayesian Convolutional Neural Network with uncertainty quantification (524359)
Siyan Liu, University of Kansas, Chemical and Petroleum Engineering, Lawrence, KS, United States, Chi Zhang, University of Kansas, Department of Geology, Lawrence, KS, United States, Reza Barati, University of Kansas, Chemical & Petroleum Engineering, Lawrence, United States and Dan Lu, Oak Ridge National Laboratory, Oak Ridge, TN, United States
 
Integrating artificial intelligence with the Subsurface Trend Analysis method to predict porous media properties (609860)
MacKenzie Mark-Moser1, Kelly Rose2, Jennifer Bauer3, Patrick Wingo1 and Anuj Suhag4, (1)Leidos Research Support Team, National Energy Technology Laboratory, Albany, OR, United States, (2)National Energy Technology Lab, U.S. Dept. of Energy, Albany, OR, United States, (3)National Energy Technology Laboratory, U.S. Dept. of Energy, Albany, OR, United States, (4)Oak Ridge Institute for Science and Education, National Energy Technology Laboratory, Oak Ridge, United States
 
Variational Auto-Encoders for Low-Rank Parameterization and Calibration of Subsurface Flow Models (625821)
Anyue Jiang, LOS Angeles, CALIFORNIA, United States and Behnam Jafarpour, University of Southern California, Los Angeles, CA, United States
 
A novel clustering workflow to do data mining on CO2/water coreflooding data (629132)
Hailun Ni, Stanford University, Energy Resources Engineering, Stanford, CA, United States and Sally M Benson, Stanford University, Geologic Carbon Storage, Stanford, CA, United States
 
Multi-scale Quantitative Analysis of Microporosity in Carbonate Fabrics (631684)
Ibukun Bode-Omoleye, Stillwater, OK, United States, Michael Grammer, Oklahoma State University Main Campus, Boone Pickens School of Geology, Stillwater, OK, United States and Adam Molnar, Oklahoma State University Main Campus, Statistics, Stillwater, OK, United States
See more of: Hydrology
<< Previous Session Proposal | Next Session Proposal >>