Full text loading...
-
Channel Detection Using Unsupervised Learning Techniques
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 80th EAGE Conference and Exhibition 2018, Jun 2018, Volume 2018, p.1 - 5
Abstract
Channel, an important geological facies for exploration and development of oilfields, has a narrow appearance. It means that the detection of this facies in the huge volume of seismic data and using numerous introduced seismic attributes is one of the most challenging tasks for interpreters. To address this difficulty, several computer-assisted learning techniques have been introduced. In recent years, the interpreters paid more attention to unsupervised learning techniques such as k-means, self-organizing maps (SOM), principal component analysis (PCA), and independent component analysis (ICA), because these learning techniques do not need to the knowledge of the interpreters about the studied area. In this study, to detect channel facies of one of the southwest oilfields of Iran, the mentioned algorithms have been used, and the results show that the studied area has two main channel branches those can be detected by all applied algorithms. Additionally, two narrow branches of the channel and some other geological feature are detected using ICA and PCA.