• DocumentCode
    1357660
  • Title

    Three-dimensional inverse scattering applied to cross-well induction sensors

  • Author

    Abubakar, Aria ; Van Den Berg, Peter M.

  • Author_Institution
    Lab. of Electromagn. Res., Delft Univ. of Technol., Netherlands
  • Volume
    38
  • Issue
    4
  • fYear
    2000
  • fDate
    7/1/2000 12:00:00 AM
  • Firstpage
    1669
  • Lastpage
    1681
  • Abstract
    Cross-well induction logging as known in the oil industry is a method for determining the electrical conductivity distribution between boreholes from the low-frequency electromagnetic field measurements in the boreholes. The authors discuss the reconstructions of the three-dimensional (3D) conductivity distribution using the contrast source inversion (CSI) method. In order to improve the reconstruction results, the concept of the extended Born approximation has been used to arrive at a preconditioning operator. Results of a number of numerical examples show that by using this preconditioning operator, a large conductivity contrast of the unknown objects (factor of 100) can be reconstructed up to an acceptable degree of accuracy. Moreover, in each iteration, the computational effort to generate the preconditioning operator is negligible
  • Keywords
    geophysical prospecting; geophysical techniques; inverse problems; terrestrial electricity; borehole method; contrast source inversion method; cross-well induction sensor; electrical conductivity; exploration; extended Born approximation; geoelectric method; geophysical measurement technique; induction logging; inverse problem; low-frequency electromagnetic field; preconditioning operator; prospecting; terrestrial electricity; tree-dimensional method; Approximation methods; Conductivity measurement; Electric variables measurement; Electromagnetic measurements; Geologic measurements; Green function; Image reconstruction; Inverse problems; Iterative methods; Scattering;
  • fLanguage
    English
  • Journal_Title
    Geoscience and Remote Sensing, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0196-2892
  • Type

    jour

  • DOI
    10.1109/36.851966
  • Filename
    851966