ML Technique Provides Hybrid Framework for Multiphase-Flow Measurement

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 228033, “Advancing Multiphase Flow Measurement: Machine Learning vs. Traditional MPFM in Flow-Loop Testing,” by Henry Qi Jin and Willow Liu, SPE, Medeng. The paper has not been peer-reviewed. _ This study demonstrates the transformative potential of physics-informed neural networks as a hybrid framework for multiphase flow measurement. A physics-informed neural network (PINN) was developed to solve the complex Navier-Stokes equations for multiphase flow measurement, including flow rates of individual phases in the multiphase mixture and phase ratios. The results were compared with a mathematics-based multiphase-flow-measurement approach and validated through a controlled loop test using oil, water, and gas. The PINN reduced reliance on predefined correlations and offered greater adaptability to changing fluid properties. The repeatability of results confirmed the model’s reliability, positioning PINNs as a promising alternative for complex multiphase-flow-measurement applications. Result To evaluate the performance of the proposed PINN for multiphase-flow-rate estimation, its predictions were compared with those from a mathematical-model-based multiphase flowmeter (MPFM) and the ground-truth values obtained from the high-temperature flow loop. The comparisons were structured into three categories to better understand the behavior of the PINN under varying flow conditions: variation in water cut, variation in gas volume fraction (GVF), and variation in total mass flow rate. For each category, multiple test cases were conducted to ensure consistency and validate the repeatability of the PINN’s performance.