Calibration of DFN Model with Well Test Data – A Case Study of the North Kuwait Jurassic Complex

Abstract

Abstract The North Kuwait Jurassic Complex (NKJC) consists of six green fields with four identified reservoirs in the naturally fractured Jurassic carbonate formation. An integrated model has been constructed by incorporating seismic, geological, petrophysical, and engineering data to estimate the hydrocarbon pore volumes and to provide multiscenario production forecasts under various assumptions. A discrete fracture network (DFN) model was also incorporated using image logs, core data, and volume curvature attributes. The DFN model is further upscaled into a dual-porosity dynamic simulation model. Fracture characterization in the dynamic model requires translating the DFN attributes into three-directional permeability, aperture, and matrix-fracture transfer functions. Because of the inherent uncertainty of the fracture properties (fracture length, aperture, permeability, intensity, and connectivity) and the uncertainty involved in translating the "forest" of fractures into a dual-porosity model, calibration of the resulting model with well test and production data is essential to ensuring a dynamic model with acceptable predictive capabilities. There is a higher degree of uncertainty in the fracture geometry than in the matrix properties in the NKJC. This led to the decision to modify the fracture properties rather than the matrix properties for the purpose of calibrating/history matching the dynamic model. The following data was used for the purpose of calibrating the fracture network model: Short-duration production tests from 31 wellsLong-term production data (48 months) from a testing facility (9 wells)Sixteen pressure transient datasets (drawdown + buildup) from 12 wellsEight months of production history from 20 wells