| Popis: |
Integrated field development studies were performed to increase oil recovery from the Marrat reservoir in the Umm Gudair field, a large, low permeability, complex, naturally fractured and highly faulted carbonate reservoir. The studies involved rebuilding the static model, creating and history matching a new dynamic model and using it to examine redevelopment scenarios. These included well interventions and workovers under primary depletion, secondary waterflood and, following a screening exercise, low salinity flooding (LSF). A new structural interpretation of 3D seismic data provided a revised static geological model and yielded insight into the number, geometry and origin of the many faults intersecting the reservoir. Rock types defined from core analysis were distributed in the static geological model using trends from Bayesian lithofacies classification based on pre-stack inversion of seismic data. Porosity and permeability were modelled by rock type. Saturation-height functions for each rock type were developed from mercury injection capillary pressure (MICP) data; and the reservoir free water level was varied so that these functions honoured the log-based water saturation interpretation. The dynamic model input description was based on available and interpreted data for the assumed oil wet reservoir. The history matching was aided by sophisticated application of decline curve analysis (DCA) and used an Opportunity Index approach to optimise well placement. The history matching led to a simplified and effective solution for characterising the locally naturally fractured reservoir nature. The effect of high permeabilities associated with increased fracture density was accommodated by introducing facies-based and distance from fault-related permeability modifiers, while maintaining geological rigour. The dynamic model was used to examine a range of field redevelopment scenarios. This showed that LSF could enhance field recovery and achieve a three-fold increase in estimated ultimate recovery, in conjunction with other improved reservoir management strategies. The results provided support for specialised laboratory and dynamic modelling investigations as a precursor to LSF pilot trials. A low cost source of LSF injectant was identified which could contribute to lowering the overall carbon footprint. |
