This study presents a novel method for determining the density of liquid phase in annular regime of two phase flows by utilizing dual modality densitometry technique and artificial neural network (ANN) model of radial basis function (RBF). The detection system is comprised of one 137Cs radioactive source, one 1 inch sodium iodide (NaI) scintillation detector for registering transmitted photons and one 1 inch NaI detector for registering scattered photons. An experimental setup was designed in order to provide the required data for testing and training the ANN. Registered counts in two detectors were utilized as the inputs of ANN and density of liquid phase was used as the output of ANN. The results show that the proposed approach could be successfully applied for predicting the density of liquid phase in annular regime of gas-liquid two phase flows. The liquid phase density was predicted with root mean square error (RMSE) of less than 0.0155 without a prior knowledge of the void fraction percentage in annular regime.
