Intelligent densitometry of petroleum products in stratified regime of two phase flows using gamma ray and neural network

Radiation based instruments in the petroleum industry are usually utilized to measure the volume fraction and flow regime type of multiphase flows in stable conditions. But in unstable conditions (when the temperature and pressure could change in pipelines), in addition to mentioned parameters, online measuring the density of liquid phase is of great importance. In this work, a combination of dual modality densitometry technique and artificial neural network (ANN) was used in order to predict the density of liquid phase in the stratified regime of gas-liquid two phase flows. In the first step, a Monte Carlo simulation model was used to obtain the optimum position for the scattering detector in dual modality densitometry configuration. At the next step, an experimental setup was designed based on obtained optimum positions for detectors from simulation to generate the required data for training and testing the ANN. Applying this novel method, density of liquid phase was predicted with the mean relative error (MRE) of less than 0.1243% in the stratified regime of gas-liquid two phase flows for void fractions in the range of 10-70%.