An NETL-led team secured first place in the engineering track of the Geothermal Energy from Oil and Gas Demonstrated Engineering (GEODE) Datathon for their development of the FlowDash Geothermal Energy Enhancer, a new machine learning (ML) tool that can reduce risks and costs for geothermal energy operations and enhance safety for neighboring communities.
Developed by researchers Abhash Kumar and Guoxiang Liu of NETL, Pengju Xing of the University of Utah, and interns Jeffrey Nguyen and Saif Qawasmeh with the National Nuclear Security Administration, the award-winning FlowDash tool demonstrated superior predictive accuracy, creativity and exploration indications. The innovation offers a custom two-level ML workflow to indicate the potential presence of geothermal energy. The tool not only maps the geometry of underground fracture networks but also identifies potential high fluid flow paths along subsurface fractures based on the expertise and technologies adapted from oil and gas for geothermal applications.
“The win at GEODE Datathon represents a significant recognition, as it underscores NETL’s leadership in leveraging advanced analytics and ML to solve critical energy challenges,” said Liu. “In just three weeks, we successfully and thoroughly prepared for this event due to the impressive commitment and hard work of each team member. GEODE Datathon was a great opportunity to exercise our oil and gas sector technologies in the geothermal sector and we are honored to have earned this recognition from such a competitive field of great ideas and technologies.”
FlowDash, which accurately identifies potential geologic “sweet spots” for discovering geothermal resources, stood out in a field of participants competing for geoscience and engineering-focused awards. FlowDash’s innovative, scalable ML-enabled workflow efficiently analyzed a large earthquake catalog provided by GEODE and multiple publicly available datasets to provide robust, low-risk and cost-effective insights that could inform decision-making by geothermal operators and other key stakeholder groups.
Researchers can further examine FlowDash-generated qualitative permeability maps to infer heat flow potentials and pathways which plays a crucial role for geothermal resource extraction and optimization. During the competition, the FlowDash developers illustrated the broader applicability of ML for subsurface characterization, particularly in data-rich but structurally complex geological settings. The team’s first-place recognition at the Datathon serves as a strong validation of the robustness and scientific value of this approach for geo-energy applications.
Now, the team plans to integrate additional hydraulic fracturing and seismic data and collaborate with university and industry partners to further refine the FlowDash methodology while expanding current geothermal applications and uncovering new possibilities for analyzing carbon dioxide storage sites, optimizing oil and gas operations and more.
The U.S. Department of Energy (DOE)-funded GEODE Consortium is a first-of-its-kind collaboration led by Project InnerSpace and the Society of Petroleum Engineers. The consortium aims to address four key facets for accelerating geothermal energy development and production: technology transfer, commercial viability, barriers to expansion and workforce adoption. GEODE brings together subject matter experts from the oil and gas and geothermal industries to research, develop and deploy next-generation geothermal technologies.
The GEODE Datathon included 43 teams divided into an engineering group of 22 teams and a geoscience group of 21 teams. More than 140 total participants represented 27 universities, 25 private companies and 16 U.S. States.
NETL is a DOE national laboratory dedicated to advancing the nation’s energy future by creating innovative solutions that strengthen the security, affordability and reliability of energy systems and natural resources. With laboratories in Albany, Oregon; Morgantown, West Virginia; and Pittsburgh, Pennsylvania, NETL creates advanced energy technologies that support DOE’s mission while fostering collaborations that will lead to a resilient and abundant energy future for the nation.