We present a general method for estimating fracture geometries from offset pressure signals that is scalable to large, complex, multi-well pad operations. The method leverages validated theory and implementation from the domain of mathematical optimization to efficiently and effectively compute macro-scale fracture dimensions for many realizations, based on randomized initial conditions. These realizations establish uncertainty bounds for a global solution, which is determined by way of fine-scale refinement of a candidate solution drawn from the macro-scale results.
Hydraulic fracturing continues to be instrumental in the development of unconventional reservoirs throughout the world. Large, complex, multi-well pad operations have led to evermore cost-effective extraction of resources with minimal surface footprint. Multi-bench landing of wells has further increased productivity by enabling operators to significantly increase the number of viable wells per section. Underpinning this development are advanced diagnostics that help completions engineers understand fracture geometries, the growth rate of fractures, and the impact of diversion and changes to completions schedules during treatment.
Among these advanced diagnostics is offset surface pressure monitoring. First pioneered by Statoil (Kampfer & Dawson, 2016), and subsequently validated by Statoil (Dawson & Kampfer, 2016) and others (Haustveit, et al., 2017), it is now common practice for operators to collect pressure data from isolated “monitor” stages in offset wells, enabling the observation of poroelastic responses induced by nearby fracturing that correspond to changes in the effective stress encompassing one or more monitor fractures. Observations are then compared to fully-coupled poromechanical 3D Finite Element Analysis (FEA), to identify specific fracture geometries giving rise to these responses.
The numerical FEA used to generate modeled pressure responses with which the comparison to field observations takes place is an example of a “digital twin.” The digital twin must account for the pad-specific well trajectories, stage locations, and timing of the treatment to yield results representative of the rock and fracture mechanics that drive offset well pressure observations.
Number of Pages
Looking for more?
Some of the OnePetro partner societies have developed subject- specific wikis that may help.