US 11,747,502 B2
Automated offset well analysis
Cheolkyun Jeong, Katy, TX (US); Francisco Jose Gomez, Abingdon (GB); Maurice Ringer, London (GB); Paul Bolchover, Beijing (CN); and Paul Muller, Beijing (CN)
Assigned to Schlumberger Technology Corporation, Sugar Land, TX (US)
Filed by Schlumberger Technology Corporation, Sugar Land, TX (US)
Filed on Oct. 8, 2021, as Appl. No. 17/450,419.
Application 17/450,419 is a continuation of application No. 16/407,186, filed on May 9, 2019, granted, now 11,143,775.
Prior Publication US 2022/0026596 A1, Jan. 27, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G01V 1/50 (2006.01); G06N 5/046 (2023.01); G06N 20/00 (2019.01)
CPC G01V 1/50 (2013.01) [G06N 5/046 (2013.01); G06N 20/00 (2019.01); G01V 2200/16 (2013.01); G01V 2210/63 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method, comprising:
receiving offset well data collected while drilling one or more offset wells, the offset well data comprising data collected during drilling one or more wells that are offset from a well being planned;
generating a machine-learning model configured to predict drilling risks in the well being planned using drilling measurements or inferences based on the offset well data;
receiving drilling parameters for one or more drilling system components that are being evaluated for use in drilling the well being planned;
determining that the drilling parameters that are being evaluated are within engineering specifications for the one or more drilling system components;
using the machine-learning model, generating a projected drilling risk profile for the well being planned, the projected drilling risk profile quantifying projected drilling risks as a value representing a likelihood of one or more drilling risks being realized while drilling the well being planned using the drilling parameters that are being evaluated;
generating a visual representation of the projected drilling risk, the visual representation comprising the value representing the likelihood of one or more drilling risks being realized while drilling the well; while drilling the well using the drilling parameters:
receiving real-time drilling measurements;
inputting the real-time drilling measurements into the machine-learning model; receiving, from the machine-learning model, an updated drilling risk profile quantifying the drilling risks; and
displaying the updated drilling risk profile to one or more users and a comparison of the updated drilling risk with the projected drilling risk.