Virtual Flow Meter Design Combining Pressure Sensing and Machine Learning Techniques

Over the past few years, many oil companies have invested in the research of making autonomous and digital models. This research can help in the development of society and improvement of accuracy and production. As the world is in need of petrol and its subordinates, the industry and the education sector are deployed in finding the digital smart solutions. A virtual flow meter (VFM) is one such solution, providing an affordable way to continuously predict flow rates in petroleum production systems. Traditional hardware metering devices, which can be prohibitively expensive, are replaced by data-driven models that utilize readily available field measurement data, such as pressure readings. These properties serve as the training data for the model. Based on the pressure sensor readings, the developed model can accurately compute the flow rates. In this approach, the model integrates both first principles and data-driven methodologies to precisely estimate water flow rates.