US Patent Application 17804413. MACHINE-LEARNED BASED REAL-TIME VIRTUAL GAS METERING simplified abstract

From WikiPatents
Jump to navigation Jump to search

MACHINE-LEARNED BASED REAL-TIME VIRTUAL GAS METERING

Organization Name

SAUDI ARABIAN OIL COMPANY

Inventor(s)

Ammal F. Al-anazi of Dammam (SA)

James O. Arukhe of Dhahran (SA)

Fatai A. Anifowose of Al-Khobar (SA)

MACHINE-LEARNED BASED REAL-TIME VIRTUAL GAS METERING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17804413 titled 'MACHINE-LEARNED BASED REAL-TIME VIRTUAL GAS METERING

Simplified Explanation

The patent application describes a method for predicting gas flow rates in wells using machine learning techniques.

  • The method involves receiving modeling data, which includes field instrument data and associated gas flow rate data for a group of wells.
  • The modeling data is split into a train set, a validation set, and a test set, and then pre-processed.
  • A machine-learned model and architecture are selected, and the model is trained using the pre-processed data to predict gas flow rates.
  • The trained model is then used to predict gas flow rates using real-time field instrument data from another group of wells.
  • The real-time field instrument data is pre-processed in a similar manner to the modeling data.
  • The method aims to provide accurate predictions of gas flow rates in real-time, which can be valuable for optimizing well operations and production.


Original Abstract Submitted

A method including receiving modeling data, wherein the modeling data includes field instrument data and associated gas flow rate data for a first plurality of wells. The method further includes splitting the modeling data into a train set, a validation set, and a test set and pre-processing the modeling data. The method further includes selecting a machine-learned model and architecture and training the machine-learned model to predict gas flow rates from the pre-processed modeling data using the training set. The method further includes using the machine-learned model to predict gas flow rates using real-time field instrument data from a second plurality of wells, wherein the real-time field instrument data has been pre-processed similarly to the modeling data.