Bayesian evaluation of photofission product yields of Th isotopic chains

The fragment yields in photon-induced fission reactions of thorium (Th) isotopes are important in modern nuclear energy applications, as well as in the evolution of the nuclear structure in its isotopic chains. Bayesian neural networks (BNN) models have been constructed to describe fragment yields in photonuclear fission reactions of thorium isotopes ranging from 216Th to 232Th, including those of 232Th at various incident photon energies. The predicted results of the optimized BNN models show good agreement with the measured data in the reactions. The double-layer BNN models successfully illustrate the systematic transition from asymmetric to symmetric fission in thorium isotopes, including associated odd-even effects, energy dependence, and the leftward shift in mass yield distributions. The developed BNN models provide new tools for predicting fragment yields in thorium photonuclear fission reactions.