(754e) SMART: A New Initiative to Transform Subsurface Visualization and Prediction through Machine Learning | AIChE

(754e) SMART: A New Initiative to Transform Subsurface Visualization and Prediction through Machine Learning


Underwood, M. K. - Presenter, Department of Energy/ NETL
Mahajan, K., National Energy Technology Laboratory
Bromhal, G., National Energy Technology Laboratory
The U.S. Department of Energy’s (DOE) Office of Fossil Energy aims to advance the science and technology of the fossil energy industry to enable the reliable, efficient, affordable, and environmentally sound use of fossil fuels. Carbon capture, utilization, and storage (CCUS) is a prominent technology being developed by DOE to reduce carbon dioxide (CO2) emissions from anthropogenic sources (e.g. fossil fuel power generation and industrial facilities). This technology depends on safe, secure, and cost-effective storage of large volumes of CO­2 in the subsurface. The Carbon Storage Program, managed by the Office of Fossil Energy through the National Energy Technology Laboratory(NETL), has partnered with the Oil and Natural Gas Program to embark on a new initiative that focuses on further elucidating the subsurface through innovative artificial intelligence and machine learning (ML) techniques.

The SMART Initiative (Science-informed Machine Learning to Accelerate Real Time Decisions in Subsurface Applications) is an NETL-led multi-organizational effort which leverages the expertise of 15 different research organizations—national laboratories, industry partners, universities and Carbon Storage Regional Initiative Partnerships. The initiative has three primary goals: 1) real-time visualization of key subsurface features and processes, 2) virtual learning for rapid investigation of reservoir behavior, and 3) real-time forecasting of select attributes at active storage and production sites for optimization and risk mitigation. The SMART Initiative capitalizes on recent advances in geophysics and data collection methods, as well as high fidelity and rapid predictive modeling. The SMART team was built from long-term collaboration through previous initiatives and partnerships and benefits from over 15 years of data collection and analysis sponsored by DOE. This uniquely positions the team to have the resources and understanding of relevant field data to develop, train, and test ML algorithms. The initiative, currently in its feasibility phase, is expected to benefit subsurface operators, regulators, and stakeholders by providing transformative solutions to visualizing and managing the subsurface in near real-time, thus reducing risk and increasing efficiency. This presentation will focus on the latest progress from the carbon storage tasks of the SMART Initiative.


This paper has an Extended Abstract file available; you must purchase the conference proceedings to access it.


Do you already own this?



AIChE Members $150.00
AIChE Emeritus Members $105.00
AIChE Graduate Student Members Free
AIChE Undergraduate Student Members Free
Non-Members $225.00