9. Uncertainty Quantification and Oil Reservoir Modelling

  1. Mike Christie1,
  2. Andrew Cliffe2,
  3. Philip Dawid3 and
  4. Stephen Senn4
  1. Mike Christie

Published Online: 17 OCT 2011

DOI: 10.1002/9781119951445.ch9

Simplicity, Complexity and Modelling

Simplicity, Complexity and Modelling

How to Cite

Christie, M. (2011) Uncertainty Quantification and Oil Reservoir Modelling, in Simplicity, Complexity and Modelling (eds M. Christie, A. Cliffe, P. Dawid and S. Senn), John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9781119951445.ch9

Editor Information

  1. 1

    Institute of Petroleum Engineering, Heriot Watt University, Edinburgh, UK

  2. 2

    School of Mathematical Sciences, University of Nottingham, Nottingham NG7 2RD, UK

  3. 3

    Centre for Mathematical Sciences, University of Cambridge, Cambridge CB3 0WB, UK

  4. 4

    School of Mathematics and Statistics, University of Glasgow, Glasgow, G12 8QW, UK

Author Information

  1. Institute of Petroleum Engineering, Heriot Watt University, Edinburgh, UK

Publication History

  1. Published Online: 17 OCT 2011
  2. Published Print: 16 DEC 2011

ISBN Information

Print ISBN: 9780470740026

Online ISBN: 9781119951445

SEARCH

Keywords:

  • Bayesian framework;
  • Markov chain Monte Carlo (MCMC) technique;
  • oil reservoir models

Summary

This chapter describes three sampling algorithms: the neighbourhood algorithm, genetic algorithms (GAs), and swarm intelligence samplers such as particle swarm optimization. It shows a comparison of Bayesian credible intervals calculated from the database of nearly 160 000 function evaluations compared with credible intervals computed from 7000 stochastic samples generated by neighbourhood algorithm (NA). Similar results are obtained with GAs. The chapter shows the application of a number of techniques to the task of calibration or ‘history matching’ oil reservoir models. It shows that the calibration problem has successfully been tackled by a variety of methods, including swarm intelligence techniques and Markov chain Monte Carlo (MCMC) methods. The chapter shows that deployment of the latest algorithms allows the oil industry to gain traction on a challenging set of problems.

Controlled Vocabulary Terms

Bayesian probability; Markov chain Monte Carlo estimation