As unconventional development of North American resources continues to mature, fewer and fewer locations for ‘unbounded’ wells exist. Rather, new child wells must be placed within pre-existing patterns of parent wells. The optimization of infill well placement within drilling spacing units (DSUs) containing parent wells is hindered by the large degree of uncertainty around child well performance. Infill child wells have been observed with up to a 40% degradation from their unbounded parents in the Midland Basin. We present an AI enabled workflow to predict infill well performance and determine optimal child placement and design. The workflow has been applied to a five square mile development in the Midland Basin. The workflow uses a flexible approach that accounts for various attributes which could influence child well performance such as offset distance from parent and cumulative parent production. Engineers using this workflow are able run full-DSU design sensitivities on a variety of well counts and relative placements to optimize the economics of child infill wells while reducing the uncertainty in their production forecasts.
Hydraulic fractures propagate perpendicular to the minimum horizontal principal stress, Shmin; a fact demonstrated by Hubert and Willis (1957) and re-affirmed recently with field experiments in the Eagle Ford (Raterman et al., 2017) and the Permian at HFTS 1 and 2 (Gale et al., 2018: 2021). In addition, the magnitude of the least principal stress defines the pressure required for the propagation of hydraulic fractures (Hubert and Willis, 1957). In the sedimentary basins of interest to developers of unconventional reservoirs of North America, this least principal stress is also the least principal horizontal stress Shmin (Lund Snee and Zoback, 2022). Lund Snee and Zoback (2022) also discuss the fact that the orientation of Shmin does not vary over the depth range targeted for horizontal development. Therefore, it is the variations in the magnitude of Shmin within, above, and below a given hydraulic fracture stage that cause hydraulic fractures to propagate vertically or to stay in zone.
Previous publications describe a geomechanically-based, machine-learning methodology to create a Frac Fingerprint (Zoback, Ruths, et. al., 2022), estimating drainage volumes around horizontal wells. These Frac Fingerprints are directly informed by variations in the magnitude of Shmin and can be inferred around any well landed within a vertical stress profile. By generating Frac Fingerprints around both existing parent wells and undrilled child wells within a DSU, optimal spacing arrangements can be identified. The Frac Fingerprint for each well becomes a feature in a production model along with other geologic and engineering variables.
