Machine learning powers U. Houston apps

Researchers release online calculators for oil viscosity and CO2 sequestration.

Researchers at the University of Houston have developed three software tools ‘to make energy industry processes more efficient’. The online calculators, developed by Mohamed Soliman, Birol Dindoruk and Utkarsh Sinha, are available to industry professionals free of charge.

The UH Viscosity Calculator calculates oil viscosity at different temperatures using a ‘full-range’ machine learning-augmented method that covers viscosities from a fraction of a centipoise (cp), to a million cp.

The UH Carbon Dioxide MMP Calculator computes the minimum miscibility pressure of carbon dioxide gas streams as used in EOR/gas injection processes. Again, an ML-augmented method was used to ‘overcome the limitations of traditional approaches’. CO2 MMP is said to be particularly important for carbon dioxide sequestration projects.

The UH Hydrocarbon MMP Calculator is likewise used in natural gas re-injection, using a ‘light gradient boost’ ML model to ‘achieve the desired pressure without needing expensive compressors or risking reservoir damage’. The UH research team is now working on a tool for measuring the carbon dioxide solubility in realistic brines with mixed salts, ‘an important topic for the evaluation of carbon dioxide sequestration in deep saline aquifers’.

The apps were built with variously Shinyapps or the Streamlit open source framework. The online calculators themselves are free to use but the code is not open source. More from U Houston.

Click here to comment on this article

Click here to view this article in context on a desktop

© Oil IT Journal - all rights reserved.