Nuclear reactor monitoring with antineutrino flux measurement method

In this study, the design, simulation, and analysis of a reactor neutrino detector using gadolinium-doped plastic scintillators as the active detector material have been conducted. Initially, a computer program was developed to generate reactor antineutrinos and inverse beta decay (IBD) events. The detector design and simulation were performed using the Geant4 program. Signal events within the detector volume were generated using the outputs obtained from the developed program for the produced IBD events. Additionally, background events were simulated using CRY (Cosmic-ray Shower Generator) within the Geant4 framework. Subsequently, a cut-based analysis was performed for the discrimination of signal and background events. Furthermore, machine learning methods such as gradient-boosted decision trees and convolutional neural networks were utilized to improve the signal-background separation performance.