Proppant consolidation and hysteresis in hydraulic fractures: Insights from the application of a strain transfer model for interpreting fracture width changes from distributed fiber optic strain sensing

Abstract

Monitoring unconventional reservoirs using fiber optics distributed strain sensing (DSS) has become crucial for assessing the efficiency of stimulation operations. DSS improves our understanding of far-field fracture geometries induced by hydraulic fracturing. Intermediate layers between the reservoir and fiber core affect the quality of the strain measurement. This study aims to improve the interpretation of strain-based measurements from hydraulic fracture stimulation by proposing an analytical model. We adopted an existing strain transfer model used in the study of concrete structures to address variations in elastic properties and interfacial slip within multilayer systems. The strain transfer model is integrated with a mechanical model incorporating proppant deformation and fracture width hysteresis during production and pressure build-up. This approach accounts for strain transfer from fractured reservoir rock to optical fibers in a multilayer well completion system. We conducted a sensitivity analysis to investigate the influence of the mechanical behavior of proppant packs on fracture width and DSS measurements. Two different soil samples were considered for this sensitivity. After the first unloading cycle, a permanent fracture width reduction of approximately 20 % is observed. The rate of fracture width reduction decreased with successive cycles. The developed model was validated using Rayleigh frequency shift distributed strain sensing (RFS-DSS) field data from the Hydraulic Fracture Test Site 2 (HFTS2) project in the Permian-Delaware Basin. The modeled strain response captures the hysteresis behavior along the unloading and reloading paths of the proppant pack during the production and pressure buildup stages. We propose that the observed extensional strain rates are not solely due to changes in fracture aperture but also strain transfer between the intermediate layers of the well completion system. The observed semi-log plot behavior of peak strain change from the field data aligns with our model's prediction. This study presents a novel application of a strain transfer model to improve RFS-DSS strain-based measurements in unconventional stimulation operations. The developed strain transfer model significantly enhances the understanding of near-wellbore hydraulic fracture characteristics and the interrelationship between stimulation and production in unconventional reservoirs.