Abstract

Summary Polymer transport and fluid rheology were implemented in a fully implicit hydraulic-fracturing and reservoir simulator. For flow in the matrix, a fluid-rheology function was used with shear thickening at high shear rates and shear thinning at medium and low shear rates. For flow in fractures, a shear-thinning constitutive law with a zero shear-rate plateau was used. The average viscosity in each fracture element was calculated by assuming smooth and parallel fracture walls and numerically solving for the velocity, shear rate, and viscosity distribution across the aperture. The simulator was used to investigate the effect on injectivity of shear thickening at high shear rates near the wellbore. For comparison, simulations were performed by use of the full shear-thickening and shear-thinning rheology function and by use of a shear-thinning-only rheology function. In the former case, the shear thickening caused rapid buildup of fluid pressure and the creation of a hydraulic fracture at the wellbore. Once the fracture formed, shear thickening no longer occurred because there was lower Darcy velocity in the matrix caused by lower concentration of flow at the wellbore. As a result, after the formation of the hydraulic fracture, injectivity in the simulation with shear thickening and thinning was nearly identical to that in the simulation with only shear thinning. The simulations were repeated with the constraint that a hydraulic fracture was not permitted to form. In this case, the simulation with shear thickening had a significantly lower injectivity than the simulation with shear thinning only. This result shows that formation of an induced fracture is a plausible explanation for unexpectedly high injectivity during polymer injection because it prevents shear thickening caused by high flow rate because of concentration of flow at the wellbore. Simulations were performed to investigate the effect of polymer rheology on the pressure transients occurring after shut-in of an injection well. Shear thickening affected the shut-in transient only at very-early time. Shear thinning affected the entire duration of the transient, causing an increase in effective fluid viscosity as the Darcy velocity gradually decreased over time. Despite fluid-rheology effects, linear flow was clearly identifiable after shut-in in the simulations with hydraulic fractures. This result shows that hydraulic fractures around polymer-injection wells can be diagnosed in field data from the linear-flow regime in the shut-in transient, regardless of fluid-rheology effects.

Full Text
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