GeoThermalCloud.jl is a repository containing all the data and codes required to demonstrate applications of machine learning methods for geothermal exploration.
GeoThermalCloud.jl includes:
- site data
- simulation scripts
- jupyter notebooks
- intermediate results
- code outputs
- summary figures
- readme markdown files
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.
Machine Learning analyses are performed using Julia.
To install the most recent version of Julia, follow the instructions at https://julialang.org/downloads/
Machine Learning 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
To install required SmartTensors modules, execute in the Julia REPL:
import Pkg
Pkg.add("NMFk")
Pkg.add("NTFk")
Pkg.add("DelimitedFiles")
Pkg.add("JLD")
Pkg.add("JLD2")
Pkg.add("Gadfly")
Pkg.add("Cairo")
Pkg.add("Fontconfig")
Pkg.add("Mads")