The process of modeling ultra low permeability and desorption-controlled shale gas reservoirs has always been challenging. Desorption is an important issue in recovery of many shale plays. Although a large amount of gas in place comes from adsorbed gas, ultra tight matrix and high bottom hole pressure may not allow this gas be produced.
In this paper, an integrated workflow is described, which demonstrates a quantitative platform for shale gas production optimization through capturing the essential characteristics of shale gas reservoirs. Because organic matter has a greater sorption affinity for CO2 than methane, a comprehensive feasibility study has been performed to evaluate the applicability and significance of CO2 injection (with simultaneous production of methane) on expedition of desorption process and also CO2 sequestration in shale.
Modeling of complex fracture networks is a very important step in simulation of shale reservoirs. Discrete fracture network (DFN) models can be used to generate and validate multiple realizations for a quantitative measure of uncertainty. A 3-D discrete fracture network using typical shale properties was generated stochastically, based on different realizations (homogenous and heterogeneous fracture properties). The complex DFNs were upscaled in order to simulate fluid flow through the system. In addition, hydraulic fractures were introduced to the model. Sensitivity analysis has been performed on key matrix and fracture properties for both natural and hydraulic fractures.
Numerical simulation was performed using appropriate gridding (logarithmic local grid refinement around the hydraulic fractures) and, in order to capture long transient gas flow from matrix to fracture, multi porosity model with matrix sub grids has been employed. This step was followed by a history matching process and the fracture and reservoir properties were varied within a range that appears consistent with actual typical shale gas well performance. One model, which gave a better approximation of the actual production profile, was selected for CO2 injection study.
In order to understand the economic effect of matrix swelling on permeability and injectivity in the presence of CO2, different injection strategies (patterns, injection time interval and rate) were defined in order to recommend the best scenario for maximum and economic recovery.