GeoThermalCloud: A Physics-informed AI/ML Framework for Geothermal Resource Exploration, Development, and Monitoring

GeoThermalCloud.jl is a repository containing data and codes required to demonstrate applications of machine learning methods for geothermal exploration, development, and monitoring.

GeoThermalCloud.jl includes:

• site data
• simulation scripts
• jupyter notebooks
• intermediate results
• code outputs
• summary figures
• readme markdown files
• Phase-I and Phase-II reports
• peer-review presentation to DOE-GTO

GeoThermalCloud.jl showcases the machine learning analyses performed for the following geothermal sites:

• Brady: geothermal exploration of the Brady geothermal site, Nevada
• SWNM: geothermal exploration of the Southwest New Mexico (SWNM) region
• GreatBasin: geothermal exploration of the Great Basin region

Reports, research papers, and presentations summarizing these machine-learning analyses are also available and will be posted soon.

GeoThermalCloud Machine Learning analyses are performed using Julia.

To install the most recent version of Julia, follow the instructions at https://julialang.org/downloads/

To install all required modules, execute in the Julia REPL:

import Pkg
Pkg.add("GeoThermalCloud")

GeoThermalCloud machine learning analyses can be executed as follows:

import Pkg
Pkg.add("GeoThermalCloud")
import GeoThermalCloud

GeoThermalCloud.SWNM() # performs analyses of the Sounthwest New Mexico region
GeoThermalCloud.GreatBasin() # performs analyses of the Great Basin region
GeoThermalCloud.Brady() # performs analyses of the Brady site, Nevada

GeoThermalCloud machine learning analyses can be also executed as Jupyter notebooks as well

GeoThermalCloud.notebooks() # open Jupyter notebook to acccess all GeoThermalCloud notebooks
GeoThermalCloud.SWNM(notebook=true) # opens Jupyter notebook for analyses of the Sounthwest New Mexico region
GeoThermalCloud.GreatBasin(notebook=true) # opens Jupyter notebook for analyses of the Great Basin region
GeoThermalCloud.Brady(notebook=true) # opens Jupyter notebook for analyses of the Brady site, Nevada

GeoThermalCloud analyses are performed using the SmartTensors machine learning framework.

SmartTensors provides tools for Unsupervised and Physics-Informed Machine Learning.

More information about SmartTensors can be found at smarttensors.github.io and tensors.lanl.gov.

SmartTensors includes a series of modules. Key modules are:

• NMFk: Nonnegative Matrix Factorization + k-means clustering
• NTFk: Nonnegative Tensor Factorization + k-means clustering

• Vesselinov, V.V., Mudunuru, M.K. Ahmmed, B., Karra, S., and O’Malley, D., (accepted): Machine Learning to Discover, Characterize, and Produce Geothermal Energy, CRS Press, Boca Raton, FL.

• Rau, E., Ahmmed, B., Vesselinov, V.V, Mudunuru, M.K., and Karra, S. (in review): Geothermal play development using machine learning, geophysics, and reservoir simulation, Renewable Energy.
• Mudunuru, M.K., Ahmmed, B., Rau, E., Vesselinov, V.V., and Karra, S. (2023): Machine Learning for Geothermal Resource Exploration in the Tularosa Basin, New Mexico. Energies, 16(7), 3098
• Mudunuru, M.K., Vesselinov, V.V. and Ahmmed, B., 2022. GeoThermalCloud: Machine Learning for Geothermal Resource Exploration. Journal of Machine Learning for Modeling and Computing.
• Ahmmed, B. and Vesselinov, V.V., 2022. Machine learning and shallow groundwater chemistry to identify geothermal prospects in the Great Basin, USA. Renewable Energy, 197, pp.1034-1048.
• Vesselinov, V.V., Ahmmed, B., Mudunuru, M.K., Pepin, J.D., Burns, E.R., Siler, D.L., Karra, S. and Middleton, R.S., 2022. Discovering hidden geothermal signatures using non-negative matrix factorization with customized k-means clustering. Geothermics, 106, p.102576.
• Siler, D.L., Pepin, J.D., Vesselinov, V.V., Mudunuru, M.K., and Ahmmed, B. (2021): Machine learning to identify geologic factors associated with production in geothermal fields: A case-study using 3D geologic data, Brady geothermal field, Nevada, Geothermal Energy.

• Mudunuru, M.K., Ahmmed, B., and Frash, L.: GeoThermalCloud for EGS -- An Open-source, User-friendly, Scalable AI Workflow for Modeling Enhanced Geothermal Systems, Geothermal Rising Conference, Reno, NV, October 1-5, 2023.
• Mudunuru, M.K., Ahmmed, B., and Frash, L.: Deep Learning for Modeling Enhanced Geothermal Systems, 48th Annual Stanford Geothermal Workshop, Stanford, CA, February 6-8, 2023.
• Frash, L. and Ahmmed, B.: A FORGE Datathon Case Study to Optimize Well Spacing and Flow Rate for Power Generation, 48th Annual Stanford Geothermal Workshop, Stanford, CA, February 6-8, 2023.
• Frash, L., Carey, J.W., Ahmmed, B., and others: A Proposal for Safe and Profitable Enhanced Geothermal Systems in Hot Dry Rock, 48th Annual Stanford Geothermal Workshop}, Stanford, CA, February 6-8, 2023.
• Ahmmed, B., Vesselinov, V.V., Mudunuru, M.K., and Frash, L.: A Progress Report on GeoThermalCloud Framework: An Open-source Machine Learning Based Tool for Discovery, Exploration, and Development of Hidden Geothermal Resources, 48th Annual Stanford Geothermal Workshop, Stanford, CA, February 6-8, 2023.
• Ahmmed, B., Vesselinov, V.V., Rau, E., and Mudunuru, M.K., and Karra, S.: Machine Learning and a Process Model to Better Characterize Hidden Geothermal Resources, GRC Transactions, v. 46, Reno, NV, August 28-31, 2022.
• Vesselinov, V.V., Ahmmed, B., Frash, L., and Mudunuru, M.K.: GeoThermalCloud: Machine Learning for Discovery, Exploration, and Development of Hidden Geothermal Resources, 47th Annual Stanford Geothermal Workshop, Stanford, CA, February 7-9, 2022.
• Vesselinov, V.V., Frash, L., Ahmmed, B., and Mudunuru, M.K.: Machine Learning to Characterize the State of Stress and its Influence on Geothermal Production, Geothermal Rising Conference, San Diego, CA, October 3-6, 2021.
• Ahmmed, B., Vesselinov, V.V.: Prospectivity Analyses of the Utah FORGE Site using Unsupervised Machine Learning, Geothermal Rising Conference, San Diego, CA, October 3-6, 2021.
• Ahmmed, B., Vesselinov, V.V., Mudunuru, M.K., Middleton, R., and Karra, S.: Geochemical characteristics of Low-, Medium-, and Hot-temperature Geothermal Resources of the Great Basin, USA, World Geothermal Congress, Reykjavik, Iceland, May 21-26, 2021.
• Vesselinov, V.V., Ahmmed, B., Mudunuru, M.K., Karra, S., and Middleton, R.: Hidden Geothermal Signatures of the Southwest New Mexico, World Geothermal Congress, Reykjavik, Iceland, May 21-26, 2021.
• Mudunuru, M.K., Ahmmed, B., Vesselinov, V.V., Burns, E., Livingston, D.R., Karra, S., Middleton, R.S.: Machine Learning for Geothermal Resource Analysis and Exploration, XXIII International Conference on Computational Methods in Water Resources (CMWR), Stanford, CA, December 13-15, 2020, no. 81.
• Mudunuru, M.K., Ahmmed, B., Karra S., Vesselinov, V.V., Livingston D.R., and Middleton R.S.: Site-scale and Regional-scale Modeling for Geothermal Resource Analysis and Exploration, 45th Annual Stanford Geothermal Workshop, Stanford, CA, February 10-12, 2020.
• Vesselinov, V.V., Mudunuru, M.K., Ahmmed, B., Karra, S. and Middleton, R.S.: Discovering Signatures of Hidden Geothermal Resources Based on Unsupervised Learning, 45th Annual Stanford Geothermal Workshop, Stanford, CA, February 10-12, 2020.

• Siler, D., Pepin, J., Vesselinov, V.V., Ahmmed, B., and Mudunuru, M.K.: A tale of two unsupervised machine learning techniques: What PCA and NMFk tell us about the geologic controls of hydrothermal processes, American Geophysical Union, New Orleans, LA,, December 13–17, 2021.
• Siler, D., Pepin, J., Vesselinov, V.V., Ahmmed, B., and Mudunuru, M.K.: A tale of two unsupervised machine learning techniques: What PCA and NMFk tell us about the geologic controls of hydrothermal processes, Geothermal Rising Conference, San Diego, CA, October 3-6, 2021.
• Ahmmed, B. Vesselinov, V. and Mudunuru, M.K., Integration of Data, Numerical Inversion, and Unsupervised Machine Learning to Identify Hidden Geothermal Resources in Southwest New Mexico, American Geophysical Union Fall Conference, San Francisco, CA, December 1-17, 2020.
• Ahmmed, B., Vesselinov, V.V., and Mudunuru, M.K., Machine learning to characterize regional geothermal reservoirs in the western USA, Abstract T185-358249, Geological Society of America, October 26-29, 2020.
• Ahmmed, B., Lautze, N., Vesselinov, V.V., Dores, D., and Mudunuru, M.K., Unsupervised Machine Learn- ing to Extract Dominant Geothermal Attributes in Hawaii Island Play Fairway Data, Geothermal Resources Council, Reno, NV, October 18-23, 2020.
• Vesselinov, V.V., Ahmmed, B., and Mudunuru, M.K., Unsupervised Machine Learning to discover attributes that characterize low, moderate, and high-temperature geothermal resources, Geothermal Resources Council, Reno, NV, October 18-23, 2020.
• Ahmmed, B., Vesselinov, V., and Mudunuru, M.K., Non-negative Matrix Factorization to Discover Dominant Attributes in Utah FORGE Data, Geothermal Resources Council, Reno, NV, October 18-23, 2020.
• Ahmmed, B., Vesselinov, V.V., and Mudunuru, M.K., Unsupervised machine learning to discover dominant attributes of mineral precipitation due to CO2 sequestration, LA-UR-20-20989, 3rd Machine Learning in Solid Earth Science Conference, Santa Fe, NM, March 16-20, 2020.