The interaction between the wellbore and pressure-induced fractures

Abstract

A method based on modeling a crack by a continuous distribution of dislocations has been used to study the interaction between the wellbore and pressure-induced fractures. This method allows the stress intensity factor and opening of the fractures to be calculated as a function of fracture length, in-situ state of stress, fracture pressure and rock properties. It also allows one to determine the wellbore pressure during quasi-static propagation. Two configurations have been considered in this report; a fracture in which acts a pressure equal to the pressure in the hole (hydraulic fracturing using a perfectly penetrating fluid) and a stress free fracture (i.e. sleeve fracturing or hydraulic fracturing using a non penetrating fluid). These two cases give respectively the lower bound and upper bound for the pressure during a micro-hydraulic fracture experiment.

The pressure behaviour has been found to be strongly dependent on fracture toughness and σ₂ for short cracks, but also on the mode of loading (i.e. pressurized crack vs. stress free crack). The influence of fracture toughness and σ₂ on the opening has been found to increase with fracture length when the fluid pressurizes the crack and to be relatively small during sleeve fracturing. These various behaviours could be used to determine the value of σ₂, σ₃, G/(1-ν) and fracture toughness if the crack opening at the wellbore is measured during a pressure test: the model is efficient enough to be used in an inversion procedure.

The model has also shown that the use of Kirsch's solution to predict breakdown pressure is only valid if the fluid does not penetrate the fracture.