Prediction of water holdup in vertical and inclined oil–water two-phase flow using artificial neural network

• Artificial neural network (ANN) is used to predict water holdup of oil–water two-phase flow in a vertical and an inclined pipe.
• The input parameters are superficial velocity of water and oil and the inclination angles of the pipe.
• 468 experimental data points were taken from literature to develop the ANN model.
• ANN predictions have very good agreement with the experimental water holdup data.
• Comparison of the prediction results of the proposed ANN model with Mukherjee et al. (1981) correlation is also presented.

This paper presents the application of artificial neural network (ANN) in prediction of water holdup of oil–water two-phase flow in a vertical and an inclined pipe (90°, 75°, 60°, and 45° from horizontal) without knowing the type of flow pattern. For this purpose, superficial velocity of water and oil and the inclination angles of the pipe were used as input parameters, while water holdup values of two-phase flow were used as output parameters in training and testing of the multi-layer, feed-forward, back-propagation neural networks. Experimental data (468 data points) were taken from literature and used for developing of the ANN model. The obtained results showed that the network predictions have very good agreement with the experimental water holdup data. The accuracy between the neural network predictions and experimental data was achieved with low average absolute percent error (AAPE) and high coefficient of determination (R2) for both training data (AAPE = 2.34% and R2 = 0.999) and testing data (AAPE = 2.89% and R2 = 0.997) sets. In addition, a comparison of the prediction results of the proposed ANN model with Mukherjee et al. (1981) correlation (AAPE = 9.83% and R2 = 0.961) revealed that the correlation had more deviations.