A fast and high-precision gamma-ray buildup factor prediction model based on ensemble machine learning

While the Geometric Progression (GP) fitting formula achieves near-unbiased accuracy in predicting buildup factors, its parameter determination process is computationally intensive and fails to meet the high real-time demands of radiation field calculations during nuclear accidents. Based on machine learning technology, this paper proposes a highly generalizable Ensemble Machine Learning model that integrates the advantages of multiple models through a stacking algorithm, significantly improving prediction accuracy and generalization performance. This provides an efficient and reliable solution for predicting buildup factors of new materials and greater penetration depths, thereby meeting the real-time and accuracy requirements of complex radiation field calculations.