Energy Science & Engineering (Oct 2020)

A semi‐analytical model for capturing dynamic behavior of hydraulic fractures during flowback period in tight oil reservoir

  • Pin Jia,
  • Ming Ma,
  • Linsong Cheng,
  • Christopher R. Clarkson

DOI
https://doi.org/10.1002/ese3.769
Journal volume & issue
Vol. 8, no. 10
pp. 3415 – 3440

Abstract

Read online

Abstract Hydraulic fracturing has been successfully employed for unconventional oil and gas recovery for decades. During flowback, the closure of the fracture may exhibit with the pressure drop of fracturing fluid dewatering. However, fracture closure always is ignored or treated as stress‐dependent fracture properties in previous flowback models. This paper presented a dynamic fracture model, which can comprehensively capture the dynamic behavior of hydraulic fractures during the flowback. A nonlinear relationship between fracture aperture and contact stress acting on the fracture surfaces is adopted to simulate fracture closure. The fracture aperture calculated by the displacement discontinuity method (DDM) is used to characterize the fracture pore volume and fracture conductivity, which will be dynamically updated in the flow model. Then, the pressure and saturation of each phase, along with the displacement on the fracture surface, are calculated by solving flow equations and geomechanics equations with iterative coupling approach. The new semi‐analytical model is validated by comparing it with a fully coupled stress‐porosity pressure numerical simulation model setup by ABAQUS® and CMG. Then, the dynamic behaviors of hydraulic fractures are investigated in detail by several cases. Results show that fracture closure is an important reason for the decline in production during the flowback and early production. And it is more important to enhance the properties of the stimulated reservoir volume (SRV) than to only create a fracture with high conductivity. Lastly, the key parameters (the fracture effective length and fracture conductivity under variable contact stress) can be interpreted by history‐matching the field flowback data.

Keywords