publication

robot looking at a well log

New adventures in assisted well log interpretation

It’s been a while since my last update on my work on assisted well log interpretation through machine learning. The good news is that I have been quite busy on the topic in the meantime. In fact, I have been collaborating with Equinor, which is Norway’s biggest oil and gas company as well as an industry partner in my research group CIUS. This collaboration led to a tool for assisted well log interpretation that is now in active use by Equinor’s cased hole logging group. More on that below!

Automatic interpretation of well logs

At CIUS, we’ve recently been working on automatic interpretation of well logs through machine learning. In July, the Journal of Petroleum Science and Engineering published this work as an article, which I wrote together with Ioan Alexandru Merciu (Equinor), Lasse Løvstakken (NTNU/CIUS) and Svein-Erik Måsøy (NTNU/CIUS). Our article is open access, so you can always go read the full thing wherever you are. But if you feel like reading a shorter summary instead of a 17-page article, this blog post is for you.

SSPA header

The Proceedings of the 43rd Scandinavian Symposium on Physical Acoustics are out!

As of today, the Proceedings of the 43rd Scandinavian Symposium on Physical Acoustics is out! I was the editor for these proceedings, just like I was for the 42nd and 39th symposia. This year’s proceedings consists of 7 submissions: 5 full papers and 2 extended abstracts. Altogether, the submissions represent 82 pages of sciency goodness.

Header for the article "Getting started with acoustic well log data using the dlisio Python library on the Volve Data Village dataset"

Getting started with well log data

Last year, I wrote a post on DLIS files, one of the most common file formats for well log data. In it, I covered a few different approaches to extract data from such files. It seems like many people struggle with this, because that post quickly became my most popular one. Well over a thousand views later, it’s time to follow it up.

I have been working with acoustic well log data since 2018. During that time, I have learned a lot about how to work with such data, and I have been wanting to share my knowledge. To do so, I teamed up with Equinor’s Erlend Hårstad and Jørgen Kvalsvik, developers of the dlisio library. Since January, we have been working on a tutorial on how to use dlisio to work with well log data. As I am an acoustician, the tutorial naturally focuses on acoustic tools. However, much of what we show is general, valid for data from any tool.

We first presented this work at the 43rd Scandinavian Symposium on Physical Acoustics in the end of January. Just last week, we published the article that we wrote for the symposium’s proceedings. Along with it, we published a companion Jupyter Notebook, which contains the code underlying the article and some further details. As of June 2020, you can even run it on Binder, so that you can play with it online without having to download anything.