US 11,781,416 B2
Determination of elastic properties of a geological formation using machine learning applied to data acquired while drilling
Andrey Bakulin, Dhahran (SA); Robert Smith, Dhahran (SA); and Stanislav Glubokovskikh, Maylands (AU)
Assigned to Saudi Arabian Oil Company, Dhahran (SA)
Appl. No. 16/612,645
Filed by Saudi Arabian Oil Company, Dhahran (SA)
PCT Filed Oct. 16, 2019, PCT No. PCT/RU2019/000738
§ 371(c)(1), (2) Date Nov. 11, 2019,
PCT Pub. No. WO2021/075994, PCT Pub. Date Apr. 22, 2021.
Prior Publication US 2021/0140298 A1, May 13, 2021
Int. Cl. E21B 44/00 (2006.01); E21B 49/00 (2006.01); G06N 20/00 (2019.01); G01V 1/46 (2006.01); G01V 1/50 (2006.01); G06F 18/24 (2023.01); G06F 18/213 (2023.01)
CPC E21B 44/00 (2013.01) [E21B 49/003 (2013.01); G01V 1/46 (2013.01); G01V 1/50 (2013.01); G06F 18/213 (2023.01); G06F 18/24 (2023.01); G06N 20/00 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
extracting, by a computer system, a first feature vector from data acquired during drilling, the data acquired during drilling comprising drilling parameters representing surface and downhole parameters, wherein the first feature vector indicative of a drilling environment classification comprises a drilling dynamics of a drilling process, the drilling dynamics related to elastic properties of rock being drilled;
determining, by a machine learning classification algorithm of the computer system, the drilling environment classification based on the first feature vector, wherein the determined of the drilling environment classification is used to select an appropriate machine learning regression algorithm to predict the elastic properties;
selecting, by the computer system, the machine learning regression algorithm that satisfies a threshold accuracy from a plurality of machine learning regression algorithms based on the drilling environment classification;
extracting, by the computer system, a second feature vector from the data acquired during drilling based on the determined drilling environment classification and the selected machine learning regression algorithm, the second feature vector indicative of the elastic properties of a geological formation;
determining, by the selected machine learning regression algorithm, the elastic properties of the geological formation based on the second feature vector; and
generating, on a display device of the computer system, a graphical representation of the elastic properties of the geological formation to enable drilling optimization and to refine a system of drilling classes.