Flow Regime Detection Using Gamma-Ray-Based Multiphase Flowmeter: A Machine Learning Approach

Flow Regime Detection Using Gamma-Ray-Based Multiphase Flowmeter: A Machine Learning Approach

Jian Hua Zhu Ritesh Munjal Ani Sivaram Santhiyapillai Rajeevan Paul Jing Tian Guillaume Jolivet

Singapore Well Testing Center, Schlumberger, Singapore

Institute of Systems Science, National University of Singapore, Singapore

Available online: 
| Citation



The presence of intermittent flow regime such as slug flow could cause issues to oil-gas well pipe-line-riser structures due to large fluctuations in pressure, leading to the production rate reduction and damage in the pipe structure. Monitoring multiphase flow regimes in production pipe systems is thus important. There are nowadays increasing use of multiphase flowmeter (MPFM) for well production flowrate metering. The associated phase fraction and flowrate measurement sensors in MPFMs could be potentially employed for multiphase-flow regime detection with no additional component required. In this study, a machine learning model is proposed to infer multiphase-flow regime from the measurements of a vertically installed gamma-ray and Venturi-based MPFM. Flow loop tests have been carried out at Singapore Well Testing Center with flows of various flow regimes observed at the horizontal inlet pipe section, such as dispersed bubbly, stratified, intermittent (slug) and annular flow regimes. The flow regime has been determined by visualization from a side glass in the flow loop pipe section and from real-time images reconstructed by an electrical-capacitance tomography system. The MPFM real-time measurements and derived or calculated data (such as Venturi differential pressure and gamma-ray mixture density) are then used as machine learning training data, with the flow regimes to be the training target. Various machine learning methods have been experimented, such as convolutional neural network (CNN), long short-term memory (LSTM) and CNN-LSTM. It has been found that LSTM method with regularization, balancing and logarithmic normalization of the calculated parameters can achieve the highest accuracy on flow regime prediction (99.6%). This study is the first attempt to predict flow regime at horizontal entrance section upstream of an MPFM with measurements made at a vertical Venturi throat section. The study also proves that flow regime could be accurately predicted by a gamma-ray and Venturi-based MPFM.


flow regime detection, gamma-ray, machine learning, multiphase flow, multiphase flowmeter.


[1] Parsi, M., Vieira, R., Sajeev, S.K., Mclaury, B.S. & Shirazi, S.A., Experimental study oferosion in vertical slug/churn flow. Corrosion, 2015.

[2] Taitel, Y. & Dukler, A.E., A model for predicting flow regime transitions in horizontaland near horizontal gas-liquid flow. AIChE Journal, 22, pp. 47–55, 1976. https://doi.org/10.1002/aic.690220105

[3] Mandhane, J.M., Gregory, G.A. & Aziz, K., A flow pattern map for gas-liquid flow inhorizontal pipes. International Journal of Multiphase Flow, 1(4), pp. 537–553, 1974.https://doi.org/10.1016/0301-9322(74)90006-8

[4] Omebere-lyari,N.K. & Azzopardi, B.J., A study of flow patterns for gas/liquid flow insmall diameter tubes. Chemical Engineering Research and Design, 85(2), pp. 180–192,2007. https://doi.org/10.1205/cherd05059

[5] Lin, S. & Kew, P.A., Pressure fluctuation and flow regimes of air-water flow ina small tube. Experimental Heat Transfer, 14(2), pp. 135–144, 2001. https://doi.org/10.1080/08916150121275

[6] Xu, L.J. & Xu, L.A., Gas/liquid two-phase flow regime identification by ultrasonictomography. Flow Measurement and Instrumentation, 8, pp. 145–155, 1997. https://doi.org/10.1016/s0955-5986(98)00002-8

[7] Mukherjee, T., Das, G. & Ray, S., Sensor-based flow pattern detection–gas-liquid-liquidupflow through a vertical pipe. AIChE Journal, 60, pp. 3362–3375, 2014. https://doi.org/10.1002/aic.14488

[8] Oon, C.S., Ateeq, M., Shaw, A., Wylie, S., AI-Shamma’a, A. & Kazi, S.N., Detection ofthe gas-liquid two-phase flow regimes using non-intrusive microwave cylindrical cavitysensor. Journal of Electromagnetic Waves and Applications, 30(17), pp. 2241–2255,2016. https://doi.org/10.1080/09205071.2016.1244019

[9] Naser, M.A., Elshafei, M. & Sarkhi, A.A., Artificial neural network application formultiphase flow patterns detection: a new approach. Journal of Petroleum Science andEngineering, 145, pp. 548–564, 2016. https://doi.org/10.1016/j.petrol.2016.06.029

[10] Rosa, E.S., Salgado, R.M., Ohishi, T. & Mastelari, N., Performance comparison of artificialneural networks and expert systems applied to flow pattern identification in verticalascendant gas and liquid flows. International Journal of Multiphase Flow, 36(9), pp.738–754, 2010. https://doi.org/10.1016/j.ijmultiphaseflow.2010.05.001

[11] Wu, H., Zhou, F. & Wu, Y., Intelligent identification system of flow regime of oil-gaswatermultiphase flow. International Journal of Multiphase Flow, 27(3), pp. 459–475,2001. https://doi.org/10.1016/s0301-9322(00)00022-7

[12] Hernandez, L., Julia, J.E., Chiva, S., Paranjape, S. & Ishii, M., Fast classificationof two-phase flow regimes based on conductivity signals and artificial neural networks.Measurement Science and Technology, 17, pp. 1511–1521, 2006. https://doi.org/10.1088/0957-0233/17/6/032