Trending Content

Hybrid Machine Learning Virtual Flow Metering

Add to Cart
Course Credit: 0.15 CEU, 1.5 PDH

Well flow metering is one of the most important measurements for oil and gas production operations. It not only provides a means of fiscal flow allocation, but also helps in predicting the overall reservoir depletion and reservoir life. Well flow metering using test separators is a dependable method; however, this process may require production deferment and is sometimes not practical for certain wells due to flow assurance issues. Multiphase flow meters (MPFM), typically installed subsea on well heads, can measure multiphase flow rates on a continuous basis, but have their own problems — expensive, requires regular calibration, limited useful life and high cost of maintenance or replacement. Virtual Flow Metering (VFM) using mechanistic model-based steady state or transient multiphase flow simulators is now a mature technology installed on facilities around the world. This software-based solution requires typical measurements already installed in wells and has clear advantages over MPFM by being less costly, easier to maintain and readily deployable. Model-based VFMs however require good estimate of GOR and water-cut, which are typically inputs to the model and must be estimated if not known.

Digital transformation initiatives in oil and gas industry have sparked interest in hybrid solutions, where physics-based models are used in tandem with machine learning (ML) techniques and provide clear benefits over pure model-based or pure historical data-driven solutions. In a hybrid VFM solution, machine learning is done using synthetic data generated from physics-based models validated against historical data if available. Any gap in quantity or quality of training data can be easily filled with physics-based models since they can simulate a wide variety of conditions and scenarios, even if they have not happened in the real-world yet, e.g., water cut or GOR changes or choke blockages. In this webinar, we will present case study of hybrid ML VFM deployed by Kongsberg Digital on production facilities.

SPE Webinars are FREE to members courtesy of the

Post Tags

 1 chapter
  1 downloadable resource

Course Chapters

  • 1Hybrid Machine Learning Virtual Flow Metering - Chapter 1
    Media Type: Video


Earn credits by completing this course0.15 CEU credit1.5 PDH credits


Dr. Neeraj Zambare
Kapil Mukati