In the final part of this three-part series, we extend our learning of Part 2 to the multivariate model and train a single model to predict three outcomes: oil, gas, and water.
This article focuses on the introduction of one of the flow-network-based models called GPSNet that has growing popularity in the literature and shows promising results during our proof-of-concept applications.
In Part 2 of this three-part series, we dive into a practical example using the production data of Equinor’s Volve field data set.
Page 1 of 28