Method and system for predicting flow rate data using machine learning

A system, comprising:

• a first well control system coupled to a first well;
• a second well control system coupled to a second well;
• a third well control system coupled to a third well;
• and
• a reservoir simulator comprising a computer processor, wherein the reservoir simulator is coupled to the first well control system, the second well control system, and the third well control system, the reservoir simulator being configured to:
• obtain first acquired pressure data for the first well and the second well;
• obtain first acquired pressure gradient data for the first well and the second well, wherein the first acquired pressure gradient data corresponds to a first pressure difference based on vertical depth at the first well and a second pressure difference based on vertical depth at the second well;
• obtain first acquired temperature data regarding the first well and the second well;
• determine first predicted pressure data for the third well using a first machine-learning model and the first acquired pressure data;
• determine first predicted pressure gradient data for the third well using a second machine-learning model and the first acquired pressure gradient data;
• and
• determine predicted flow rate data for the first well using a third machine-learning model, the first predicted pressure data, the first predicted pressure gradient data, and the first acquired temperature data.

A non-transitory computer readable medium storing instructions executable by a computer processor, the instructions when executed by the computer processor are configured to perform a method comprising:

• obtaining first acquired pressure data for a plurality of wells in a geological region of interest;
• obtaining first acquired pressure gradient data for the plurality of wells, wherein the first acquired pressure gradient data corresponds to a pressure difference based on vertical depth at one or more wells among the plurality of wells;
• obtaining first acquired temperature data regarding the plurality of wells;
• determining first predicted pressure data for a first well in the geological region of interest using a first machine-learning model and the first acquired pressure data;
• determining first predicted pressure gradient data for the first well using a second machine-learning model and the first acquired pressure gradient data;
• determining predicted flow rate data for the first well using a third machine-learning model, the first predicted pressure data, the first predicted pressure gradient data, and the first acquired temperature data;
• and
• transmitting a command to a control system based on the predicted flow rate data.