Back to results
US 12,608,520 B2Granted
Intelligent bit design
AssigneeHalliburton Energy Services, Inc.
InventorsWilliam Brian Atkins, Radompon Sungkorn, Michael Stephen Pierce
Life of this patent
Estimating the term…
Abstract
A method includes determining a bit design based on a design criterion, determining a cutter design to be mounted on the bit design at a bit design position based on cutter information, wherein the cutter information is based on a historical cutter performance metric of one or more cutters having the cutter design. The method also includes generating a drill bit model based on the bit design and the cutter design and determining that the drill bit model meets the design criterion.
Claims18
A method comprising:
- determining a bit design based on a design criterion;
- determining a cutter design to be mounted on the bit design at a bit design position based on cutter information, wherein the cutter information is based on a historical cutter performance metric of one or more cutters having the cutter design, wherein the cutter information further comprises a cutter design grading value, wherein the cutter design grading value is based on a set of cutter grading values, and wherein the set of cutter grading values are determined using a machine-learning framework comprising a neural network;
- generating a drill bit model based on the bit design and the cutter design;
- and
- determining that the drill bit model meets the design criterion.
2 more elements
A method comprising:
- acquiring a set of cutter training images and a set of training design grading values, wherein each of the set of training design grading values is associated with one or more of the set of cutter training images;
- and
- generating a trained machine-learning framework from a machine-learning framework using the set of cutter training images as inputs and the set of training design grading values as target values in an objective function during training, wherein the trained machine-learning framework provide a cutter design based on a historical cutter performance metric and a cutter design grading value, wherein the cutter design grading value is in the set of training design grading values.
An apparatus comprising:
- a processor;
- and
- a machine-readable medium having program code executable by the processor to cause the apparatus to,
- determine a bit design based on a design criterion,
- determine a cutter design to be mounted on the bit design at a bit design position based on cutter information, wherein the cutter information is based on a historical cutter performance metric of one or more cutters having the cutter design, wherein the cutter information further comprises a cutter design grading value, wherein the cutter design grading value is based on a set of cutter grading values, and wherein the set of cutter grading values are determined using a machine-learning framework comprising a neural network,
- generate a drill bit model based on the bit design and the cutter design, and determine that the drill bit model meets the design criterion.
3 more elements
Dependencies are inferred from the claim text.