CPC E21B 41/00 (2013.01) [E21B 43/25 (2013.01); E21B 47/10 (2013.01); E21B 47/138 (2020.05); G01V 99/005 (2013.01); G06F 18/214 (2023.01); G06F 18/24 (2023.01); G06N 20/00 (2019.01); E21B 2200/20 (2020.05)] | 20 Claims |
1. A method implemented by one or more computing devices, the method comprising:
accessing historical data from a plurality of databases, the historical data including well production data of a plurality of wells, well completions data, flow meters data, and well rate tests data;
accessing historical perforation data and historical reservoir properties data from a simulation model;
training, using a plurality of input values, a machine learning model for predicting oil flow values at the perforated intervals of a plurality of target wells, wherein the plurality of input values comprise fluid flow values and rock quality index values associated with perforated intervals of the plurality of wells that are based at least in part on the historical perforation data and the historical reservoir properties data from the simulation model, and wherein the fluid flow values and rock quality values are linked to the well production data included in the historical data of the plurality of wells;
predicting, using the trained machine learning model, the oil flow values at the perforated intervals of the plurality of target wells; and
generating a synthetic production log that includes the oil flow values at the perforated intervals of the plurality of target wells.
|