Scaling parameters for characterizing gravity drainage in naturally fractured reservoir



Journal Title

Journal ISSN

Volume Title



Numerical simulation of naturally fractured reservoirs undergoing immiscible gas injection requires specific information about fracture and matrix properties including laboratory determination of capillary pressure and relative permeability for each fluid phase. It also requires PVT analysis of fluid phases. Additionally, phase segregation due to gravity, capillarity, and gas diffusion must be considered. Numerical models for naturally fractured reservoirs are generally divided into two types. The first is the double porosity single permeability (dual porosity) model. The second is the double porosity double permeability (dual permeability) model. The difference between the two models is basically that the second type considers matrix block-to-block flow while the first does not. The present study is focused on the dual porosity model. Numerical models require a transfer function calculation between matrix and fracture. Therefore proper determination of mass transfer from matrix to fracture plays an important role in generating good simulation results. In a gas injection project, the difference in density between gas and liquid phases makes it important to consider gravity segregation and capillary forces that holds liquid in the matrix rock. The goal of this project is to determine methods of scaling dimensionless variables to simplify the analysis and thus identify the main parameters controlling the gravity drainage process in naturally fractured reservoirs matrix blocks. This work has application in optimization, history matching, and stochastic simulation through its promise to reduce the amount of computer time required. The primary tasks are a) analysis of gravity segregation with gas injection in a single matrix block, b) determination of dimensionless scaling groups, c) analysis and test of common dual porosity transfer functions, and d) application using a commercial dual porosity model.