Uncertainty analysis for hydraulic fracturing modeling in porous media containing oil and gas

Abstract

The hydraulic fracturing process (HFP) includes irreversible deformation, fluid-rock interactions, and fluid flow with phase transitions. Performance of a HFP may be quantitatively assessed through prescribed measures: length and height of the macro-fractures, incremental increase or decrease in porosity around main fractures, incremental increase in elastic deformation and temperature, fluid pressure distribution in the pore space etc. Given that formation and fluid properties are characterized poorly, performance predictions of HFP process are likely to be uncertain. Development and assessment of HFP strategy necessitates uncertainty quantification of formation and fluid properties. We perform Global Sensitivity Analysis to quantify and rank contributions to the uncertainty in HFP performance from individual input parameters of the model. All the processes (i.e., irreversible deformation, fracturing, micro-damaging, heat transfer) in the proposed mathematical model are strongly coupled. The explicit normalized-corrected meshless method (SPH method) is used to solve the resulting governing equations. The flexibility of the proposed numerical technique allows running efficiently using a great number of micro-and macro-fractures. The results are presented, discussed and future studies are outlined.