Multi-data reservoir history matching for enhanced reservoir forecasting and uncertainty quantification
KAUST DepartmentEarth Science and Engineering Program
Physical Sciences and Engineering (PSE) Division
Permanent link to this recordhttp://hdl.handle.net/10754/565998
MetadataShow full item record
AbstractReservoir simulations and history matching are critical for fine-tuning reservoir production strategies, improving understanding of the subsurface formation, and forecasting remaining reserves. Production data have long been incorporated for adjusting reservoir parameters. However, the sparse spatial sampling of this data set has posed a significant challenge for efficiently reducing uncertainty of reservoir parameters. Seismic, electromagnetic, gravity and InSAR techniques have found widespread applications in enhancing exploration for oil and gas and monitoring reservoirs. These data have however been interpreted and analyzed mostly separately, rarely exploiting the synergy effects that could result from combining them. We present a multi-data ensemble Kalman filter-based history matching framework for the simultaneous incorporation of various reservoir data such as seismic, electromagnetics, gravimetry and InSAR for best possible characterization of the reservoir formation. We apply an ensemble-based sensitivity method to evaluate the impact of each observation on the estimated reservoir parameters. Numerical experiments for different test cases demonstrate considerable matching enhancements when integrating all data sets in the history matching process. Results from the sensitivity analysis further suggest that electromagnetic data exhibit the strongest impact on the matching enhancements due to their strong differentiation between water fronts and hydrocarbons in the test cases.
SponsorsThe work presented in this paper has been supported in part by the project entitled Simulation of Subsurface Geochemical Transport and Carbon Sequestration, funded by the GRP-AEA Program at King Abdullah University of Science and Technology (KAUST).