System to detect the interface region and identify products transported by polyducts using artificial intelligence and gamma radiation through MCNP6 code

Abstract

This article presents a methodology for predicting the purity level of oil byproducts that are transported in polyducts, thus identifying the interface region. The detection system consists of a 662 keV gamma-ray source and two NaI(Tl) detectors to register the transmitted and scattered beams. An artificial neural network (ANN) was trained and evaluated for interpret the spectra recorded in both detectors. The Purity Level and the Purity Indicator of the fluids were assigned as outputs from the ANN. The models for a stratified flow regime are composed of fluids (gasoline, kerosene, oil, and glycerol) with several combinations of biphasic flows developed using MCNP6 code. The results were determined with the MAPE of 2.40% for all patterns in the prediction of the purity level of fluids. The proposed methodology has the potential to detect the interface region and identify the petroleum byproducts, with an accuracy of 99% as purity level.