Application of the DDFN (Discrete and Deformable Fracture Network) Method Within an Integrated Unconventional Reservoir Study Workflow

Abstract

Abstract Objectives/Scope The inherent complexity of unconventional resources, within the ever growing economic development of these, gave rise to many work-flows, in which both natural and hydraulic fractures are accounted for through the use of DFN (Discrete Fracture Network) models. Assessing the specific role of fractures in a multiphase flow context while inferring their mechanical behavior as well as their interaction leads to a better understanding of production characteristics from shale and tight reservoir. Methods, Procedures, Process A realistic reservoir case was considered for this study. A classical characterization methodology was used, integrating different scales, from seismic to core analysis. This characterization step, along with geomechanical considerations, such as brittleness leads to a statistical description of two fracture sets (natural and hydraulic), building a continuous DFN. This makes up our fracture model, geo-stochastically controlled. A simplified coupling method, for which two distinct mechanical laws (plastic-elastic) are applied, is used to describe the hydraulic fracturing process. Hydraulic fluid injection is simulated using an approximate fluid model (PAD + ‘Proppant’). Accounting for the pressure dependent fracture compressibility involves the inclusion of the dynamic behavior of fractures into the DDFN through the use of analytic and empirical fracture deformation models. Results and Observations The history-match of the BHP recorded during the stimulation and overall microseismic cloud, honors hydraulic fracturing characteristics such as injection rates and fluid properties, hence allowing the validation of both the characterization and geomechanical hypothesis formulated. This calibration was carried out on each hydraulically fractured stage, followed by an integration to the reservoir fluid flow simulator. Novel/Additive Information This paper describes a new method, called DDFN (Discrete and Deformable Fracture Network), applied at a large reservoir scale. The reservoir discretization method is computationally efficient, making it appropriate for any optimization of the hydraulic fracturing process. An additional characteristic of the DDFN approach is the by-passing of the up-scaling step, since the DDFN is included at the reservoir simulation scale directly, in the form of an unstructured grid, thus providing a more realistic representation of the overall fracture geometry. Examples of such simulation results performed using realistic data are shown. Discussion of the worth and limitations of the method is done. Planned Conclusion The application of this method to all stages makes up a realistic method, which could be used at reasonable speeds within any reservoir study. The main advantage of such a method is that it can be adapted to any characterization method. By nature modular, it could be linked to any workflow which provides a continuous fracture network made up of the interaction between natural and hydraulically induced fractures