Title: A reservoir identification method based on rough set and support vector machine

Authors: Sun Han; Zhang Huan; Guo Haixiang; Cheng Jinhua

Addresses: School of Economics and Management, China University of Geosciences, 388 lu molu St., Wuhan 430074, Hubei, China ' School of Economics and Management, China University of Geosciences, 388 lu molu St., Wuhan 430074, Hubei, China ' School of Economics and Management, China University of Geosciences, 388 lu molu St., Wuhan 430074, Hubei, China ' School of Economics and Management, China University of Geosciences, 388 lu molu St., Wuhan 430074, Hubei, China

Abstract: There are many logging parameters that affect identification of oil zones in the course of discrimination. A mass of redundant information exists. The identified precision and speed are impacted. Conventional methods cannot identify oil zones effectively. The method is put forward that information is optimised by rough set combining with support vector machine (SVM), and applied to identify the zones. On the basis of analysing on rough set theory and SVM method, the SVM identifying process of oil zones based on rough set is put forward. The practical research on three wells in an oilfield having testing data has been done. The results with the artificial neural network (ANN) method were compared, and show that the method is effective and feasible, the identified precision rate is 96% over ANN.

Keywords: rough sets; attributes reduction; SVM; support vector machines; reservoir identification; oil zones; oil wells; oilfields; artificial neural networks; ANNs; oil reservoirs.

DOI: 10.1504/IJCAT.2014.066726

International Journal of Computer Applications in Technology, 2014 Vol.50 No.3/4, pp.196 - 199

Published online: 07 Feb 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article