Machine Learning

Machine Learning
Advanced algorithms could better identify areas for exploratory geothermal drilling as part of machine learning initiatives.

Machine learning can assist in finding and developing new geothermal resources. If applied successfully, it can lead to higher success rates in exploratory drilling, greater efficiency in plant operations, and ultimately lower costs for geothermal energy.

What is Machine Learning?

Just as humans learn, so can machines. The machine “learns” as the algorithm running the program is taught to predict, classify, and uncover key insights while data mining—think computers learning to beat checkers masters after playing against them multiple times. Machine learning is the use of advanced algorithms to identify patterns and make inferences from data.

Machine learning and artificial intelligence can provide impactful insights on large, complicated datasets, including those used to analyze geothermal energy.

Machine Learning and Geothermal

Starting in 2018, the Office of Geothermal (OG), formerly named the Geothermal Technologies Office (GTO), funded early-stage research and development applications in machine learning to develop technology improvements in exploration and operational improvements for geothermal resources.

The rapidly advancing field of machine learning offers substantial opportunities for technology advancement and cost reduction throughout the geothermal project lifecycle, from resource exploration to power plant operations.

OG’s goals for the machine learning projects:

  • Provide impactful insights using datasets and encourage diverse machine learning techniques to drive geothermal exploration.
  • Assist in identifying and predicting the locations of new geothermal resources, leading to higher success rates in exploratory drilling.
  • Achieve greater efficiency in plant operations and lower costs for geothermal energy operators.

More than $9 million was invested across phase 1 and phase 2 of OG’s machine learning initiative, focusing on two areas:

  • Phase 1 ($5.5 million): Machine Learning for Geothermal Exploration: OG funded projects that advance geothermal exploration through the application of machine learning techniques to geological, geophysical, geochemical, borehole, and other relevant datasets. Of particular interest were projects that aimed to identify drilling targets for future work.
    • Colorado School of Mines (Golden, CO)
    • Lawrence Livermore National Laboratory (Livermore, CA)
    • Los Alamos National Laboratory (Los Alamos, NM)
    • National Renewable Energy Laboratory (Golden, CO)
    • Pennsylvania State University (University Park, PA)
    • University of Arizona (Tucson, AZ)
    • University of Houston (Houston, TX)
    • University of Nevada (Reno, NV)
    • University of Southern California (Los Angeles, CA)
    • Upflow Limited (Taupo, New Zealand)
  • Phase 2 ($3.5 million) Advanced Analytics for Efficiency and Automation in Geothermal Operations: OG also funded projects that applied advanced analytics to power plant and other operator datasets, with the goal of improving operations and resource management.
    • Colorado School of Mines (Golden, CO)
    • Los Alamos National Laboratory (Los Alamos, NM)
    • Pennsylvania State University (University Park, PA)
    • University of Houston (Houston, TX)

Learn more about recent efforts in hydrothermal resources and other OG priorities.