Toward robust history matching of a giant carbonate gas-condensate reservoir

Carbonate reservoirs are highly heterogeneous and require considerable effort to characterize and model. This is because of complex depositional and diagenetic processes involved in their formation. This paper presents and discusses a systematic methodology that leads to a sound full-field dynamic model such that its behavior is closed to the field historical records without much interfering the model. The methodology was applied to a giant carbonate gas-condensate reservoir characterized by several heterogeneous layers. Sedimentology and sequence stratigraphy studies were performed to capture the heterogeneity and selection of geological zones and subzones. Prior to construction of the field simulation model, single-well modeling was conducted and tuned to well test data to target several objectives. These included evaluation of formation capacity (kh) and skin factor/non-Darcy effect as deducted from a match to drill stem test (DST) results. The effect of skin factor (S) and non-Darcy coefficient (D) on flow performance in the framework of sensitivity analysis was followed. Boost factors were obtained to upscale the core averaged kh to that of well test and used in the full-field dynamic model. The dynamic model was upscaled using production logging tool/DST data and incorporating the concept of hydraulic flow zone unit. The objective of tracking model coherency throughout the upscaling process was followed. The initial gas in place closely matched that of the static model. With the approach used in the current work, the data matching process took minimum effort. In addition, the model although of a relatively considerable size required short computing time for all forecast scenarios. (c) 2015 Curtin University of Technology and John Wiley & Sons, Ltd.