(713e) A Reduced Order Model for Optimizing Hydraulic Fracture Stimulation of Horizontal Wells

One approach to optimizing hydraulic fracturing stimulations of horizontal oil and gas wells is to maximize created fracture surface area and/or maximize the uniformity of stimulation across each fracturing stage. Here we present a rapidly-computing model to capture the impact of fluid flow and fluid leakoff on surface area and/or uniformity optimization while enabling optimization requiring hundreds or thousands of model evaluations. The key innovation for the rapidly-computing model is a novel concept we call the “composite viscosity”, which is essentially to lump all dissipative processes into an equivalent viscosity that allows the simulator to estimate the length, width and pressure based on a reduced order model that globally conserves mass, momentum, and energy. The results of these simulations show a tradeoff. On one hand, low viscosity leads to larger fracture surface area because high viscosity leads to larger fracture widths and the potential for more suppression of fractures by the stresses induced by their neighbors. However, low viscosity also leads to decreased fracture surface area due to higher fluid losses to the formation. This tradeoff leads to the existence of a fluid viscosity, specific to the properties of a given reservoir rock, that optimizes fracture surface area and/or uniformity.