Reservoir modelling: reducing uncertainty (polynomial chaos expansion)

January 2015December 2016
Results showed that PCE performs well on test models and is able to identify the level of activity of each cell of the model.

Reservoir modelling: reducing uncertainty with polynomial chaos expansion (PCE)

This research assessed the use of a promising engineering technique (polynomial chaos expansion—PCE) to significantly reduce the number of dynamic flow simulations required to assess uncertainty associated with static and dynamic reservoir modelling primarily in coal seam gas models.

Reservoir modelling is an essential activity in the assessment of development strategies for both conventional and unconventional petroleum resources. Given the limited subsurface information that is available at early stages in the field life, industry usually determines the bounds on reservoir behaviour by investigating different scenarios and different realisations.

Results showed that PCE performs well on test models and is able to identify the level of activity of each cell of the model. The improved functionality, and application to history matching sensitivity analysis, suggests value for industry based research and development.

PROJECT OUTPUTS

  • Project status: Complete
  • Project title: Uncertainty modelling with polynomial chaos expansion
  • Project leader: Diane Donovan
  • Research group: The University of Queensland Centre for Coal Seam Gas & The University of Queensland School of Earth and Environmental Sciences
  • Timeframe: January 2015 - December 2016
  • Project funders: APLNG, Arrow Energy, Santos, QGC, University of Queensland