Title: Estimation of earth stresses and fracture properties using numerical method and soft computing
Authors: Shike Zhang; Xiaoqi Niu; Shunde Yin; Zhongliang Ru
Addresses: School of Civil Engineering and Architecture, Anyang Normal University, Anyang, Henan 455000, China ' School of Civil Engineering and Architecture, Anyang Normal University, Anyang, Henan 455000, China ' Department of Chemical and Petroleum Engineering, University of Wyoming, Laramie, WY 82071, USA ' School of Civil Engineering, Henan Polytechnic University, Jiaozuo, Henan 454000, China
Abstract: Recently, artificial neural networks (ANNs) have been introduced to solve petroleum geomechanics characterisation problems dealing with the inherent nonlinear relationship in well log data. This paper presents a computational intelligence algorithm of artificial neural network (ANN) and genetic algorithm (GA), which is studied and applied in petroleum geomechanics to estimate the information of earth stresses and fracture properties based on wellbore displacement information when drilling well. In this computational intelligence algorithm, an ANN is used to represent the nonlinear relationship between the petroleum geomechanics parameters and the wellbore deformation. GA is used to identify the set of unknown earth stresses and natural fracture properties at wellbore scale in a large space based on the objective function. Our experiments illustrate that the proposed computational intelligence method is able to generate reliable results.
Keywords: computational intelligence; artificial neural networks; ANNs; genetic algorithms; geomechanical parameters; earth stresses; fracture properties; soft computing; petroleum geomechanics; oil industry; wellbore displacement; oil well drilling; wellbore deformation.
International Journal of Computational Science and Engineering, 2016 Vol.13 No.4, pp.390 - 399
Received: 30 Jun 2014
Accepted: 26 Aug 2014
Published online: 04 Nov 2016 *