A general poro-elastic model for pad-scale fracturing of horizontal wells
MetadataShow full item record
Economic production of oil and gas from tight rocks requires horizontal well drilling with multiple hydraulic fractures along the length of the horizontal wells. Multiple horizontal wells are drilled and fractured close to each other to increase the recovery of oil and gas from a single location or pad. Interference between fractures in a horizontal well pad is commonly observed in the field. There is no clear understanding of the impact of various operational and reservoir parameters on the observed interference. This inter-well interference can occur through the creation of complex fracture networks and/or poro-elastic stress changes. In this research, the development of a poro-elastic numerical simulator was undertaken to evaluate hydraulic fracturing practices in pad-scale scenarios. The primary motivation was to assess the impact of various operational parameters such as fracture spacing, well spacing and fracture sequencing on the geometry of the created fractures. Two approaches were used to understand the problem at hand. In the first approach, static fractures were simulated in 3-D and the impact of their stress shadow on subsequent fractures was studied. It was observed that fracture spacing, injection volume, and time between successive fractures were the most important parameters that could be used to optimize the creation of fractures in a well. Formation properties such as Young’s modulus and horizontal stress contrast modified the magnitude and spatial extent of the stress shadow and the extent of stress reorientation. It was shown that stage spacing, well spacing and fracture sequencing together with fracture designs (volume of sand pumped and fluids used) can be adjusted to obtain non-intersecting, transverse fractures that efficiently drain the reservoir. A hypothesis, time dependent closure of induced unpropped fractures, was presented to explain why zipper fracturing often outperforms conventional sequential fracturing. The hypothesis was tested and confirmed with a field data set made available to us by Shell from the Eagle Ford shale. In the second approach, a novel finite volume based 3-D, geomechanical, field-scale numerical simulator was developed to simulate propagation of multiple fractures simultaneously in a poro-elastic reservoir. This provided a more realistic model of the pad-scale fracturing process. The ability of the model to perform realistic pad-scale simulations was demonstrated for a variety of field situations such as multi-cluster multi-stage fracturing, infill-well fracturing, re-fracturing, mini-frac analysis and fracture network simulations. The inclusion of poro-elastic effects and reservoir heterogeneity in the model allowed us to examine the effects of reservoir depletion on fracture geometry in refraced and infill wells.