The use of multiple clusters per stage can reduce the number of stages needed to place a given number of fractures in a horizontal wellbore. However, there is concern that some of the perforation clusters may not propagate fractures at all as the fractures compete for frac fluid and fracture width. In this paper, the influence of various hydraulic fracturing design parameters and in-situ conditions on the resulting fracture dimensions and propagation patterns is studied. A 3D geomechanical model was built using ABAQUS to simulate the propagation of multiple competing fractures in a horizontal well. The reservoir was modeled as a porous elastic medium using pore pressure stress elements, and a plane of pore pressure cohesive elements was inserted at each perforation cluster to model fracture propagation. Also, flow distribution among perforation clusters was simulated using a parallel resistors model.

Using this model, a number of simulations were run with a combination of perforation cluster spacing, number of perforation clusters, fracture height, frac fluid viscosity, pumping rate, Young's modulus of the target formation, and formation heterogeneity. The results demonstrated that increasing perforation cluster spacing or decreasing the number of perforation clusters reduced stress interference between the fractures, and more fractures propagated. A high frac fluid viscosity increased fracture widths and led to more stress interference. Higher pumping rates led to longer and wider fractures, and to an increase in the number of propagated fractures. Thicker target formations and taller fractures required wider perforation cluster spacing. Additionally, a higher Young's modulus of the target formation led to more fracture interference, suggesting wider cluster spacing. Lastly, the introduction of any degree of heterogeneity significantly altered the fracture propagation patterns noted above.

Based on the results of these simulations, the following recommendations are made. First, there exists an optimum number of perforation clusters per fracture stage. This optimum depends primarily on the fracture design, the stage spacing and the mechanical properties of the formation. Second, slick water fracs have a better chance (compared to more viscous fluids) of propagating uniform multiple fractures when closely spaced multiple perforation clusters are used. Third, higher pumping rate should be adopted to ensure that a fracture propagates from all perforation clusters. Fourth, even small scale heterogeneities should be considered, as they can significantly alter the fracture propagation pattern. Lastly, when the target formation is relatively thick or has a high Young's modulus, a wider fracture spacing, a higher pumping rate, and a lower viscosity frac fluid should be used.

You can access this article if you purchase or spend a download.