Study on Artificial Neural Network for Predicting Gas-Liquid Two-Phase Pressure Drop in Pipeline-Riser System

The pressure drop for air-water two-phase flow in pipeline systems with S-shaped and vertical risers at various inclinations (−1°, −2°, −4°, −5° −7° from horizontal) was predicted using an artificial neural network (ANN). In the designing of ANN model, superficial velocity gas liquid as well inclination downcomer were used input variables, while values flows determined output. An a hidden layer containing 14 neurons developed based on trial-and-error method. A sigmoid function chosen transfer layer, linear output layer. Levenberg-Marquardt algorithm training model. total 415 experimental data points reported literature collected creation networks. statistical results showed that proposed is capable calculating dataset low average absolute percent error (AAPE) 3.35% high determination coefficient (R2) 0.995.