Two-Phase flow regimes identification using artificial neural network with nonlinear normalization

Multiphase flow measurement is a very challenging issue in process industry. There are several techniques to estimate multiphase flow parameters. However, these techniques need correct identification of the flow regimes first. Artificial Intelligence is one promising technique for identification of the flow regimes. In this paper we used Artificial Neural Network in identifying the flow regimes using multiphase flow parameters such as superficial velocity of liquid and gas, pressure drop, liquid hold up and Reynolds' number. We proposed a pre-processing stage to normalize large data range and to reduce overlapping between flow regimes. It was shown that using the natural logarithms of certain flow parameters as inputs to neural network improved the identification process.