US 12,276,769 B2
Pulsed neutron informed machine learning techniques to determine porosity and lithology
Mayir Mamtimin, Spring, TX (US)
Assigned to Halliburton Energy Services, Inc., Houston, TX (US)
Filed by Halliburton Energy Services, Inc., Houston, TX (US)
Filed on Nov. 10, 2022, as Appl. No. 17/984,674.
Claims priority of provisional application 63/326,020, filed on Mar. 31, 2022.
Prior Publication US 2023/0314652 A1, Oct. 5, 2023
Int. Cl. G01V 5/06 (2006.01); G01V 5/04 (2006.01); G01V 5/10 (2006.01); G06N 3/063 (2023.01); G06N 3/08 (2023.01)
CPC G01V 5/102 (2013.01) [G01V 5/045 (2013.01); G06N 3/063 (2013.01); G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method comprising:
disposing a pulsed neutron logging (PNL) tool into a borehole that is disposed in a formation;
emitting at least one neutron from a neutron source on the PNL tool into the formation;
capturing one or more gamma rays with a gamma ray detector expelled from formation in response to the at least one neutron from the neutron source to form a plurality of PNL measurements in a log;
acquiring a plurality of historical PNL measurements;
reducing the number of plurality of historical PNL measurements to form a reduced set of PNL measurements;
training a targeted machine learning (ML) model based on the reduced set of PNL measurements and plurality of PNL measurements; and
identifying at least one formation property with the targeted ML model.