| CPC E21B 7/04 (2013.01) [E21B 44/00 (2013.01); E21B 2200/20 (2020.05)] | 20 Claims |
|
1. A method comprising:
receiving input for a multiwell pad;
selecting, based at least in part on a portion of the input, a template for the multiwell pad utilizing a trained machine learning model trained utilizing pad data, wherein the template comprises initial well trajectories associated with one or more anti-collision assurances or one or more anti-collision strategies;
generating, based at least in part on the template, well trajectories for the multiwell pad, wherein each of the well trajectories extends from a surface location of the multiwell pad to one or more reservoir target locations and wherein each of the well trajectories comprises at least one curve generated using one or more dogleg severity (DLS) values and a plurality of geometric control points that are not required to lie on one or more of the well trajectories, wherein the well trajectories are generated via a pad sketcher component configured to operate in a first operational mode and a second operational mode, the pad sketcher component comprising:
a preprocessor component configured to assign target points and one or more optimizations when the pad sketcher component operates in the first operational mode; and
a generator component coupled to the preprocessor component and configured to utilize the target points and one or more optimizations to generate an output as specifications for the generated trajectories of the multiwell pad when the pad sketcher component operates in the first operational mode and generate the output as specifications for the generated trajectories of the multiwell pad via bypassing the preprocessor component when the pad sketcher component operates in the second operational mode; and
outputting the specifications for the generated trajectories of the multiwell pad.
|