Full text loading...
-
Neural networks technologies in oil and gas well logging
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 17th International Conference on Geoinformatics - Theoretical and Applied Aspects, May 2018, Volume 2018, p.1 - 6
Abstract
The artificial neural network (ANNs) is an effective tool for making empirically grounded decisions in cases where either the mathematical model of the research object or the process theory is absent or underdeveloped. Every year there are increases the number of investigations on the use of ANNs technology in the oil and gas well logging. Application of ANNs technology in the well logging of complexly constructed, thin-layered sections is a very promising direction for increasing the efficiency of qualitative and quantitative interpretation of geophysical data. The main directions of the implementation of ANNs in the well logging are: 1. Creation the synthetic curve of the geophysical parameter, if it is absent for any reason among other geophysical data. 2. Evaluation of capacitive-filtration characteristics of rocks in a well section with the using of the created ANNs. 3. Detection the rocks-reservoir and determination of their saturation in the sections of wells on the basis of original algorithms using ANNs. 4. Improvement of the resolution of the curve of the geophysical parameter on the basis of the created model of ANNs.
The conservatism of traditional approaches to the interpretation of geophysical data restrains the wider use of neural networks in practice.