Use this URL to cite or link to this record in EThOS:
Title: An investigation into the links between upscaling and history-matching
Author: Monfared, Hashem
ISNI:       0000 0001 3414 7658
Awarding Body: Heriot-Watt University
Current Institution: Heriot-Watt University
Date of Award: 2007
Availability of Full Text:
Access from EThOS:
Access from Institution:
Upscaling procedures attempt to account for subgrid heterogeneity in such a way that coarse grid simulations produce flow scenarios similar to those that one would obtain by running simulations directly on fine grid geological models. The conventional single-phase upscaling approach leads to averaging of low and high permeability streaks. As a result, the underlying physics of th~ reservoir is ignored and the permeability variability decreases. Consequently, further adjustment to absolute permeability is required in the history matching stage. The essential issue is whether the ultimate permeability distribution of the history-matched model bears any semblance or relationship to that of the upscaled model. This dissertation investigates the link between upscaling and history matching. First, we introduced the Effective Permeability Ratio concept (EPR) to formulate the errors arising from upscaling. Later, by employing geostatistics and assisted history matching techniques, coarse history matched model was generated by adjusting absolute.permeability fields. The comparison of resulted coarse model with upscaled mo~el proved that the permeability variability, which plays a major role in the flow response of reservoir models, could be preserved using the proposed workflow. Furthermore, the capability of suggested workflow in generating multiple history matched models enabled us to investigate the uncertainty in prediction performance using the Bayesian framework. In the cases studied, the proposed workflow produced a comparable result to the truth case suggesting that, the geological knowledge at the fine scale can be preserved appropriately on the coarse scale and the uncertainty in the field prediction can be quantified.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available