Computation of gamma-ray exposure buildup factors up to 10 mfp using generalized feed-forward neural network

Abstract

This paper presents an approach based on generalized feed-forward neural network (GFFNN) to compute exposure buildup factors (B(D)) for point isotropic sources in infinite homogeneous media at energies varying from 0.03 MeV to 15 MeV and up to depths of 10 mean free paths (mfp). The results obtained by using the proposed model have been compared with the ANSI standard data, the calculations by use of EGS4 Monte Carlo code and Invariant Embedding (IE) Method for water, iron, lead and concrete. The comparisons have shown that the GFFNN model improved B(D) estimation with respect to the other methods, particularly for lead and concrete.