ARTIFICIAL INTELLIGENCE AND PROMPT GAMMA NEUTRON ACTIVATION ANALYSIS FOR VOID FRACTION PREDICTION

Resumo

This report presents a simulation of a water-gas biphasic system in annular and stratified flow regimes using a neutron beam from a ²⁴¹Am-Be radiation source to assess the potential of prompt gamma neutron activation analysis combined with an artificial neural network (ANN) for flow measurements. The MCNP6 code was employed to create a measurement geometry featuring a neutron flat source with parallel emission and a spherical detector. Prompt gamma ray spectra served as input data for an ANN designed to predict void fraction and identify flow regimes. The findings indicated that the ANN estimated void fraction for both flow regimes with a Mean Absolute Percentage Error (MAPE) of less than 1.6% across all data and identified flow regimes with 100% accuracy.