Self-supervised similarity models based on well-logging data

TL;DR

This paper introduces a self-supervised deep learning approach using variational autoencoders to create universal data representations from well-logging data, enabling effective transfer learning across different oil fields with minimal additional data.

Contribution

It proposes a novel self-supervised method for well-logging data that produces universal representations, reducing the need for large labeled datasets in oil and gas applications.

Findings

Variational autoencoders provide the most reliable models.

Universal representations enable transfer learning with small target datasets.

The approach improves classification and clustering tasks in well-logging data.

Abstract

Adopting data-based approaches leads to model improvement in numerous Oil&Gas logging data processing problems. These improvements become even more sound due to new capabilities provided by deep learning. However, usage of deep learning is limited to areas where researchers possess large amounts of high-quality data. We present an approach that provides universal data representations suitable for solutions to different problems for different oil fields with little additional data. Our approach relies on the self-supervised methodology for sequential logging data for intervals from well, so it also doesn't require labelled data from the start. For validation purposes of the received representations, we consider classification and clusterization problems. We as well consider the transfer learning scenario. We found out that using the variational autoencoder leads to the most reliable and accurate models. approach We also found that a researcher only needs a tiny separate data set for the target oil field to solve a specific problem on top of universal representations.