1-D DC Resistivity Modeling and Interpretation in Anisotropic Media Using Particle Swarm Optimization


PEKŞEN E., YAS T., KIYAK A.

PURE AND APPLIED GEOPHYSICS, vol.171, no.9, pp.2371-2389, 2014 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 171 Issue: 9
  • Publication Date: 2014
  • Doi Number: 10.1007/s00024-014-0802-2
  • Journal Name: PURE AND APPLIED GEOPHYSICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.2371-2389
  • Keywords: Electrical anisotropy, Particle swarm optimization, 1-D DC interpretation, HANKEL-TRANSFORMS
  • Kocaeli University Affiliated: Yes

Abstract

We examine the one-dimensional direct current method in anisotropic earth formation. We derive an analytic expression of a simple, two-layered anisotropic earth model. Further, we also consider a horizontally layered anisotropic earth response with respect to the digital filter method, which yields a quasi-analytic solution over anisotropic media. These analytic and quasi-analytic solutions are useful tests for numerical codes. A two-dimensional finite difference earth model in anisotropic media is presented in order to generate a synthetic data set for a simple one-dimensional earth. Further, we propose a particle swarm optimization method for estimating the model parameters of a layered anisotropic earth model such as horizontal and vertical resistivities, and thickness. The particle swarm optimization is a naturally inspired meta-heuristic algorithm. The proposed method finds model parameters quite successfully based on synthetic and field data. However, adding 5 % Gaussian noise to the synthetic data increases the ambiguity of the value of the model parameters. For this reason, the results should be controlled by a number of statistical tests. In this study, we use probability density function within 95 % confidence interval, parameter variation of each iteration and frequency distribution of the model parameters to reduce the ambiguity. The result is promising and the proposed method can be used for evaluating one-dimensional direct current data in anisotropic media.