Porosity and permeability are two important parameters to be considered in evaluating the characteristics of an oil field. The permeability is the key parameter in describing a hydrocarbon reservoir. In fact, knowing the exact values of permeability is an effective, efficient and important tool for engineers in the oil production process and management of a field. Over the years, these two petrophysical parameters have been derived using core and well tested data; however, the results obtained using the two methods are accurate, but not sufficient to describe the full field, because these methods are not applicable in many cases. Recently, an artificial neural network technique was introduced, and has been found very useful in various areas of sciences and engineering especially in petroleum engineering. In this study, we have combined regression analysis techniques and neural networks methods to estimate permeability from well-logging data obtained in part of an Iran's oil fields. Comparing the results from this study with conventional techniques, indicate that these methods have fairly and reasonably predicted the permeability. But it was observed that the artificial neural network has better results than the other method.
Key words: Reservoir rock, permeability, petrophysical data, logging, neural networks,regression analysis.
Copyright © 2023 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0