Challenges in current simultaneous well log correlation techniques

Wesley Banfield and Jonathan Edwards and Guillaume Caumon. ( 2016 )
in: 2016 RING Meeting, ASGA

Abstract

Well correlations are traditionally done by hand but the job becomes more and more complex and time consuming as the number of wells and markers increase. As computing power has increased, machines are being used more and more to carry out the task bringing with it other advantages, notably the reproducibility of the correlations. Different algorithms exist to correlate multiple linear signals. In this paper we build on Wheeler [2014a] simultaneous well log correlation technique. We stress that the inconsistencies between pairwise correlations are primarily due to the existence of unconformities, and that the least-squares minimization proposed by Wheeler [2014a] may locally lead to the collapse of some well intervals. We also suggest that geological rules can be more appropriate than random values to address incomplete well sections in well correlations. These results highlight the need for further research to better address the complex problem of simultaneous well correlations.

Download / Links

BibTeX Reference

@INPROCEEDINGS{,
    author = { Banfield, Wesley and Edwards, Jonathan and Caumon, Guillaume },
     title = { Challenges in current simultaneous well log correlation techniques },
 booktitle = { 2016 RING Meeting },
      year = { 2016 },
 publisher = { ASGA },
  abstract = { Well correlations are traditionally done by hand but the job becomes more and more complex and
time consuming as the number of wells and markers increase. As computing power has increased,
machines are being used more and more to carry out the task bringing with it other advantages,
notably the reproducibility of the correlations. Different algorithms exist to correlate multiple linear
signals. In this paper we build on Wheeler [2014a] simultaneous well log correlation technique. We
stress that the inconsistencies between pairwise correlations are primarily due to the existence of
unconformities, and that the least-squares minimization proposed by Wheeler [2014a] may locally
lead to the collapse of some well intervals. We also suggest that geological rules can be more
appropriate than random values to address incomplete well sections in well correlations. These
results highlight the need for further research to better address the complex problem of simultaneous
well correlations. }
}