denisplakic / Thermal-Recovery-of-Bitumen-from-West-Coast-Tar-Sands

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Thermal Recovery of Bitumen Project

Project Information

The project studies and projects a realistic representation of the fluctuations of pressure and temperature within a defined system if thermal recovery operations of hot steam injection and bitumen extraction are carried out. Through calibration, using numerical methods, and uncertainty analysis, a model has been developed which predicts the potential fluctuations of pressure and temperature under different circumstances.

Files in the repository include:

  • main.py: this file when ran will produce all the plots relevant to the study.
  • project_functions.py: this file contains benchmarking and unit tests which are necessary to validating the accuracy of our numerical integration function (solve_ode).
  • model_calibration.py: this file contains the functions for calibrating the model.
  • model_calibration_initial.py: this file contains the functions used for calibrating the initial model.
  • model_prediction.py: this file contains the functions for predicting many future scenarios of the model.
  • model_posterior.py: this file contains the functions for the uncertainty analysis of our model.
  • data: this file contains the pilot study's recorded data.
  • figures: this file contains all the figures generated from the study.

Project Use

This project may be used to make recommendation to the applicant in their resource consent application.

Installation

If you wish to edit, the repository may be cloned and the main.py file may be run using any editor such as Visual Studio Code.

Framework Used

The code has been built using Visual Studio Code, under the Python Programming Language.

Contact

sjeo598@aucklanduni.ac.nz
ckah285@aucklanduni.ac.nz
dpla864@aucklanduni.ac.nz

Contribute

We welcome all pull requests. If you wish to make major changes, please contact us to discuss what you would like to change.

Please ensure tests are updated appropriately.

Credits

Project Members: Angela Jeong, Chelsea Kah, Denis Plakic.
Thank you to Dr. David Dempsey and his team for their contributions.