Breakthrough in Artificial Intelligence: Machine Learning Model Improves Operational Efficiency in Oil and Gas Industry

Strategic Insights -

Researchers from Universidad Industrial de Santander in Colombia have developed a robust machine learning model based on artificial neural networks to classify six flow patterns in oil-water two-phase flow within horizontal pipelines. This innovation provides significant value for improving pipeline design, optimizing flow assurance strategies, enhancing corrosion control, and supporting real-time operational decision-making in multiphase transport systems in the oil and gas industry.

Key Takeaways:

  • The machine learning model achieved an overall accuracy of 95.4% with training, validation, and testing accuracies of 97.1%, 92.8%, and 90.3%, respectively.
  • The model demonstrated rapid convergence with a training time of only 2 seconds, making it a reliable and computationally efficient tool for flow pattern recognition.
  • The research assembled a database comprising 1,846 experimental data points from the literature, encompassing various operating conditions, including fluid properties, superficial velocities, and pipe diameters.
  • After evaluating 104 network configurations, the optimal model was selected, demonstrating the effectiveness of the approach.
  • The study provides significant value for the oil and gas industry by improving pipeline design, optimizing flow assurance strategies, enhancing corrosion control, and supporting real-time operational decision-making.
  • The researchers, Daniel Yesid Uribe-Tarazona and Carlos Mauricio Ruiz-Diaz, are affiliated with the School of Mechanical Engineering at Universidad Industrial de Santander, Bucaramanga, Colombia.
  • The research was supported by Universidad Industrial De Santander and published in Frontiers in Mechanical Engineering.

Statistics:

  • 1,846 experimental data points were assembled for the machine learning model.
  • 104 network configurations were evaluated to select the optimal model.
  • The machine learning model achieved an accuracy of 95.4% with a training time of 2 seconds.
  • The training, validation, and testing accuracies were 97.1%, 92.8%, and 90.3%, respectively.
  • The cross-entropy error was 0.024.

Sources:

  • NewsRx LLC, Universidad Industrial de Santander Researchers Describe Recent Advances in Machine Learning [Machine learning technique for the identification of two-phase (oil-water) flow patterns through pipelines]. Energy Weekly News. July 25, 2025; p 726.
  • Frontiers in Mechanical Engineering, Machine learning technique for the identification of two-phase (oil-water) flow patterns through pipelines. Frontiers in Mechanical Engineering, 2025,11.
  • Universidad Industrial de Santander, School of Mechanical Engineering, Bucaramanga, Colombia.
  • Daniel Yesid Uribe-Tarazona and Carlos Mauricio Ruiz-Diaz, Universidad Industrial de Santander, Bucaramanga, Colombia.