Effective RTA-Based Calibration for Numerical Hydraulic Fracture and Reservoir Flow Simulations in the Turner Formation

Abstract

AbstractInitial hydraulic fracture simulation calibration in unconventional reservoirs is commonly done based on data acquired during Fracture Calibration Tests (FCT), and in a postmortem fashion from treatment data acquired in the field. After the well has been put on production, another independent piece of information becomes available through Rate Transient Analysis (RTA) which can be used to constrain and finetune the simulated hydraulic fracture geometries.Production data used for RTA and observed fracture treatment data from a well in the Turner formation have been utilized to effectively calibrate an integrated numerical modeling study of a representative well in the Powder River Basin Province and landed in the Turner formation. The RTA provides two independent parameters. The first being the Ac√k from the linear flow regime, from which the total effective hydraulic fracture surface area, Ac is derived. The second parameter estimated from boundary dominated flow is the Stimulated Reservoir Volume (SRV). These two parameters are then used to constrain the hydraulic fracture simulations (with Ac) and the numerical flow model (with SRV).A two-folds feedback loop based on the available RTA interpretation allowed fast conjoint calibration and history matching of the hydraulic fracturing treatments and the long-term production performance.Numerical simulation results provided a final production history match achieved by constraining the effective drainage height (guided by geological observations), supported by the hydraulic fracture simulations and the RTA-derived information.Differences from RTA and numerical results are attributed to common RTA assumptions such as a homogeneous reservoir, single phase fluid, and constant fluid properties. The Turner formation in this area is relatively homogeneous, with the Carlile formation, a strong stress barrier, above and a highly laminated underlying Frontier formation. This led to the conclusion, that hydraulic fracture propagation downward into the Frontier was limited by the highly laminated rock and well contained within the Turner.Additional differences between the RTA and numerical results are attributed to the true complexity of the final hydraulic fracture geometry. RTA is an analytical solution for a set volume with geometric hydraulic fractures whereas these hydraulic fracture simulations are numerical and create varying geometries. Taking this concept a step farther, actual hydraulic fracture propagation will interact with natural fractures, localized stresses, and each other, creating an even more complex and hydraulic fracture geometry.Presented here is a novel application where the coupling of RTA and numerical simulation workflows has been successfully applied.A two-folds feedback loop based on RTA-derived parameters allows fast calibration and history matching of the hydraulic fracturing treatments and the long-term production performance.The independent hydraulic fracture surface areas and SRV are independently consumed, and results highlight impact of non-trivial geological features, such as stress barriers, laminations and lithological variability.