Abstract
Currently, two unresolved challenges encountered in unconventional reservoir simulation are: (1) accurately modeling complex interaction between hydraulic fracture, natural fracture and tight matrix; (2) efficiently couple geomechanics with fluid flow in fractured reservoirs. The objective of this study is to couple geomechanics with two hybrid fracture models built on distinct methodologies of combining Embedded Discrete Fracture Model (EDFM) and Multiple INteracting Continua (MINC), so that a better model for simulation can be selected and the geomechanical impact on unconventional reservoir development can be observed and evaluated.
A geomechanics coupled EDFM has been developed to investigate how stresses and strains could affect fluid flow in fractures and matrix. Fluid flow and geomechanics were fully coupled in an Integrated Finite Difference (IFD) approach in this model. A new extension of this model to couple geomechanics for combined EDFM and MINC is developed. With the implementation of hybrid EDFM and MINC model, this coupled simulator can be leveraged to describe the complex fluid flow interaction between hydraulic fracture, natural fracture and tight matrix, under the influence of geomechanics. In this study, two hybrid models are compared with each other, and also with analytical solution and refined grid model: one is the embedded discrete fractures intersected with classical nested MINC blocks and the other is the MINC partition conforming the discrete fractures geometry and following the iso-pressure profile. These cases are compared in terms of oil production rate, cumulative production and pressure decline distribution.
The comparison results show that the second hybrid model is capable of better capturing the transient and long-term behavior, especially when geomechanics is taken into account. Several scenarios are investigated and interpreted: (1) changing the stimulated reservoir volume where both hydraulic and natural fractures are activated; (2) changing the number of continua and grid block size in two hybrid models and seek the optimal combination to accurately capture the transient flow behavior. The comparison of cases with and without geomechanics coupling can be utilized to observe its effects on cumulative productions and short- and long-term production rates. This coupled model is built on a parallel computing framework so it is also desired to examine the scalability of the model for large scale problems with complex fracture networks.
