Nowadays the oil and gas industry is increasingly facing the difficulties associated with fields characterized by the presence of natural fractures. These difficulties are due to uncertainties in geological model especially in the fracture distributions. To successfully address these challenges, it is paramount to improve the characterization and simulation of those reservoirs. This can only be performed through the fully integration of all available data for generating stochastic DFN model and then using streamline simulation approach for fast history matching.

In this paper, using interpretation of image logs closely was related best productive zones with open fractures and then their correlation with other parameters (including lithology and petrophysical logs) were determined through reservoir. Three main fracture direction set were defined by fold-related fractures. A discrete fracture network (DFN) was generated using inferred parameters from outcrops, FMI logs and cores. Then, permeability of DFNs developed was calibrated with available well test permeability. After that, a streamline simulation due to high computation speed, high accuracy and good visualization was used to repeated history matching multiple equi-probable of a dual porosity model in gas condensate reservoir. So, with running streamline simulation, three realizations (High, Medium and Low) ranked based on the objective function values. In final, comprehensive history matching will be done as optimally for the three-selected realizations.

The overall goal in this study was to develop representative fluid flow simulation model for reservoir management as gas cycling scheme in enhanced condensate recovery. This method due to using actual fracture parameters has great application in the high resolution fractured reservoir modeling for model ranking, screening and optimum dynamic model calibration for reduction of the history matching effort without fabricated by reservoir engineer.

You can access this article if you purchase or spend a download.