A simple approach to predict a type of fluid from post-stack seismic data using ML/AI algorithms.

PSS-Geo presented its part of the research project, the Fluid type prediction using ML/AI algorithms. Three hours of live-streamed video can be found on youtube: https://www.youtube.com/watch?v=YRr2ZM5bPzE

It was an intense week for GeoPython enthusiasts! Organised by Fluxgate Technologies, Nigeria, the event collected many participants. PSS-Geo is thankful for the possibility to share our solutions.

We presented a simple approach to predict a type of fluid from post-stack seismic data using ML/AI algorithms. We provided the codes and polished data in cvs format for quick execution. 7 wells + 3D seismic and inverted data of 16000km2 of one formation (152ms) was flattened to 0-time to easy run on any "home" PC.

Even though we showed a basic approach, we gave a direction on how it can be improved to run more efficiently. The video side chart is also with great tips!

So, if you "need a case solved" for your CV/portfolio, I believe in going thru our codes and making some modification base on tips or your ideas can be an easy achievement. If you need so, please send us a link to your code (modified our codes) to the comment here, we will issue you a prove letter/certificate.

Please contact us to download the task and codes. Task description: 37-42 minutes.



Featured Posts