Transference of Reservoir Uncertainty in Multi SAGD Well Pairs

Abstract

Extensive computational time is required to predict SAGD production performance by thermal reservoir simulation. The common practice is to evaluate the performance of a single well pair and combine the results to pads and projects. Production forecasts for multiple SAGD well pairs simultaneously are computational expensive. At the same time, there is a need to make reservoir management decision in presence of reservoir uncertainty. Accounting for joint uncertainty across the reservoir requires simultaneous management of multiple realizations. Thermal simulators cannot be applied to the transference of such uncertainty to SAGD performance of many well pairs. A proxy model based on the Butler's SAGD theory has been developed to overcome this challenge. The proxy is able to account for reservoir heterogeneity. Modifying factors are included to fit the proxy results to simulation outcomes. The methodology for the transference of the reservoir uncertainty includes: generation of the reservoir uncertainty model using geostatistical tools; ranking the geologic realizations using the Cumulative Steam Oil Ratio (CSOR) calculated by the unfitted proxy. The ranking process is done in a reservoir volume representative of a single well pair: reservoir simulation for the P10 to P90 realizations, fitting the proxy model to simulation results, and assessment of the uncertainty of the SAGD performance by Monte Carlo Simulation (MCS). This methodology was applied to a synthetic example of a pad of eight SAGD well pairs, using 3D geometry and based on hard data from an Athabasca bitumen deposit. The results show the uncertainty in the dynamic performance of SAGD production variables for the entire pad. The results permit decision making under the uncertainty. It is shown that approximate models accounting for uncertainty are better than precise models that cannot be used in stochastic calculations. IntroductionAfter the success of the AOSTRA's Underground Test Facility (UTF) project [1-3], the Steam Assisted Gravity Drainage (SAGD) has become the standard recovery technology of oil sands industry to exploit the vast in-situ bitumen resources in Canada. Numerous commercial applications and pilot projects have been implemented since the UTF. As March 2008, eighteen commercial applications are being operated or approved by ERCB in Alberta [4].