Alert
Uploading external datasets is currently disabled. Tiered licensing support will be provided.
Characterization and managmenet of geologic reserrvoirs using AI/ML methods

GeoML

: AI/ML for characterizing and managing geologic reservoirs

Dataset:
Upload dataset. Currently, the feature is disabled
Reset
Select ML method for data analyses
ML method (characterization):
Select ML method for reservoir predictions
ML method (prediction):
Select the number of signatures (clusters) to extract
Number of features:
Select the number of interations
Number of iterations:
Execute ML analyses of the selected reservoir dataset
Data mapsML Analytics mapsML-generated feature mapsML-generated utilization mapsData tablesData and ML-generated figuresSignature matricesML modelsInformation about our AI/ML methods and toolsGeo exploration and utilizationLand Acknowledgement
Data map
Data attribute to plot
Map color theme and design
Map of the data locations associated with the ML extracted features
Map color theme and design
Map of the ML extracted features
Feature to plot
Map color theme and design
Map of alternative reseroir utilizations
Geothermal and fluid disposal utilization currently supported; more to come!
Map color theme and design
Distance between injection/extraction wells
Pressure increase at a radial distance around the injection/extraction will be less than or equal to specified target
Radial distance around the injection/extraction wells at which pressure will be less than or equal to the specified target
Data Table:
Data Plot:
Scatter plot of the reserovir attributes; dots are colored based on the extracted signatures
Signature Plot:
BiPlot of the extracted signatures in relation to the data locations
Signature BiPlot associated with Locations:
BiPlot of the extracted signatures in relation to the resorvoir attributes
Signature BiPlot associated with Attributes:
ML generated matrix representing the relationship between reserovir attributes and extracted signatures
Attributes / Signatures Matrix:
Sort reseroir attributes based on the signature association
Permeability field generated internally by the code. In the future, users would be able permeability fields and reservoir properties represenative of their sites.

Permeability [log10(m2)]:

Execute ML analysis predicted injection/extraction rates and pressures given a permeability field

Pressure [m]:

Generate a new random permeability field
Execute ML analysis predicting injection/extraction rates and pressures given a permeability field
ML predicted extraction rate

Predicted extraction rate: m3/s

ML predicted pressure at a distance away from the wells

Predicted pressure: m

Distance between injection/extraction wells
Pressure increase at a radial distance around the injection/extraction will be less than or equal to specified target
Radial distance around the injection/extraction wells at which pressure will be less than or equal to the specified target

AI/ML Technology

GeoML relies on our cloud-computing Artificial Intelligence and Machine Learning (AI/ML) methods and tools.

These methods and tools are provided by our open-source AI/ML codes SmartTensors and SmartML.

SmartTensors and SmartML are available on GitHub

SmartTensors/SmartML capabilities:

  • unsupervised (self-supervised), supervised, and physics-informed ML methods and tools.
  • pre- and post-processing methods and tools.
  • wide range of visualization options (maps, charts, plots).
  • robustness and interactivity.
  • diagnostics of the ML results, including sensitivity, uncertainty, risk and decision analyses.

GeoML: AI/ML software for analysis and interpretation of geochemical and geophysical data

GeoML capabilities:

  • providing methods and tools to better understand and utilize geologic subsurface for energy production and storage.
  • addressing geologic complexities, technology advancements, and energy needs.
  • utilizing artificial intelligence and machine learning (AI/ML) tools.
  • facilitating the interactions between the industry, regulators, stakeholders, and end-users.
  • merging data, computational methods, knowledge, expertise, and experience.
  • providing fast processing, analysis and dissemination of geologic information related to subsurface energy storage and production.
  • helping us understand and predict complex subsurface processes associated with energy extraction and storage in the subsurface.

Land Acknowledgement

EnviTrace LLC acknowledges that the place now called Santa Fe where our office is located sits on unceded ancestral Tewa Land and is still recognized as O'gha Po'oge, meaning White Shell Water Place.

We recognize that this land is just one piece of a larger, boundless terrain for Indigenous peoples that include the Nambe Pueblo, the Tewa, and the Jicarilla Apache.

O'gha Po'oge was once a thriving Pueblo village, and their descendants include the modern-day Tewa people who still live in Santa Fe and the local Pueblos of Nambe, Pojoaque, San Ildefonso, Ohkay Owingeh, Santa Clara and Tesuque.

We also recognize the violence, displacement, migration, and colonization that haunt this place.

We understand we are now stewards of this land, responsible for the care of water, air, and each other.

In addition, we pledge to donate a percentage of our profit to a local non-profit organization supporting the education of native students.

EnviTrace LLC:

Making Sense of Environmental & Earth-Science Data