Abstract
We present a 3D numerical model for hydraulic fracturing and damage of low permeable rock in an anisotropic stress field. The model computes the intermittent damage propagation, microseismic event-locations, microseismic event-distribution, damaged rock volume and injection pressure. The model builds on concepts from invasion percolation theory, where cells in a regular grid are connected by transmissibilities, also called bonds. An intact bond breaks when the fluid pressure exceeds the least compressive stress and random uniformly distributed bond strength. Analytical expressions are obtained for the fractions of the weakest bonds in the x-, y- and z-directions as a function of the stress anisotropy and the rock strength anisotropy. A strategy is suggested for a fast solution of the pressure equation when the fluid pressure is restricted to the damaged rock volume. An expression for the volume of damaged rock is obtained in the cases where it has a high permeability. The model is demonstrated on a published case from the Barnett Shale and it reproduces the observed main features, such as the spatial and temporal distribution of the events, the magnitude–frequency distribution and the injection pressure. The microseismic event-distribution and the b-value depend on the permeability of the damaged rock volume. The b-value increases with decreasing permeability from a little < 0.6 to a value above 2 for the maximum possible permeabilities. The damaged rock volume is non-compact and similar to a percolation cluster for “high” damaged rock permeabilities, and it becomes increasingly compact with decreasing permeabilities. The resulting loop-less fracture network is found to have similar characteristics for different damaged rock permeabilities.
Highlights
Hydraulic fracturing is the basis for US shale gas and shale oil production
We present a 3D numerical model for hydraulic fracturing and the damage of low permeable rocks, which aims at producing realistic results in terms of injection pressure and microseismicity
Both the 2D model (Wangen 2017) and the suggested 3D model are in turn based on a 2D invasion percolation model for hydraulic fracturing developed by Norris et al (2014, 2015a, 2016), which is able to produce realistic microseismic event-distributions
Summary
Hydraulic fracturing is the basis for US shale gas and shale oil production. The year 2016 had an average production of 4.25 million barrels daily of crude oil from tight shales (EIA 2017). The pressure build-up produces fracturing of the tight shales and hydrocarbon expulsion through a pervasive fracture network. The naturally generated fracture network created by hydrocarbon generation seems to be important for hydraulic fracturing to efficiently enhance petroleum production. It appears that hydraulic fracturing is less effective in shales where hydrocarbon generation and expulsion are active processes and where the fracture network is open. The injection of large volumes of water under high pressure reopens the sealed fracture network and mobilizes the hydrocarbons that are trapped in the source rock (Turcotte et al 2014). We present a 3D numerical model for hydraulic fracturing and the damage of low permeable rocks, which aims at producing realistic results in terms of injection pressure and microseismicity. How the frequency–magnitude distribution depends on the permeability of the damaged rock is studied before some properties of the loop-less fracture network are discussed
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call