Note: The Data used in here is public but owned by U.S. Department of Energy, Office of Fossil Energy

The goal of the study below is to use machine learning algorathems in order to train different set of wireline logs to predict DT the sonic log.

pandas matplotlib numpy scipy seaborn

The realm of petrophysics is still underdevelopment , studying statsitcal predictive methods of different logs is of importance . in this case, predicting sonic log can help in cost savings specially in hetrogenious reservoirs . also it can be used to enhance Geo-Models a 1% difference can mean millions of barrels of oil.

1. Can we Predict DT from different wireline logs?
2. Can we use predicted DT to predict porosity ?
3. Can Predicted DT be used in Seismic corrections?

My post in LinkedIn summarizes the main findings of the expirement.

• Applied Machine Learning Approach to Predict Sonic Log from Different Wireline Logs

The datasets used in this analysis are publicly relased by by U.S. Department of Energy, Office of Fossil Energy.