Near real-time return-on-fracturing-investment optimization for fracturing shale and tight reservoirs

摘要

Near real-time methodologies for maximizing return-on-fracturing-investment for shale fracturing. An example system can calculate, based on sonic data and density data, mechanical properties and closure stress of a portion of shale rocks for fracture modeling. The system can generate one or more rock mechanical models based on the mechanical properties and closure stress of the portion of shale rocks, and perform one or more fracture modeling simulations based on one or more treatment parameter values. Based on the one or more fracture modeling simulations, the system can generate a neural network model which predicts a fracture productivity indicator of an effective propped area (EPA) and/or an effective propped length (EPL), and calculate a return-on-fracturing-investment (ROFI) based on the EPA or EPL predicted by the neural network model.