What is it about?

We investigate the well-log interpretation error introduced by nonstationarity, and use statistical and machine learning methods to mitigate the errors. At the same time we reveal the statistical assumptions of classical well-log preprocessing methods and discuss the proper scenarios of applications.

Featured Image

Why is it important?

For the first time we analyzed the statistical assumptions of statistical well-log preprocessing methods, lay the theoretical foundation for future study on statistical well-log preprocessing. The new methods to mitigate nonstationary error will help improve the well-log interpretation accuracy.


It is great to write this paper to understand the limitation of well-log preprocessing methods and develop new methods to resolve them.

Wen Pan
University of Texas at Austin

Read the Original

This page is a summary of: Improving multiwell petrophysical interpretation from well logs via machine learning and statistical models, Geophysics, March 2023, Society of Exploration Geophysicists, DOI: 10.1190/geo2022-0151.1.
You can read the full text:



The following have contributed to this page