Study on Intermediate- and High-Energy Proton and Neutron Induced Fission Cross-Sections by Coupling the Liège Intranuclear Cascade Model with Bayesian Neural Networks

摘要: The intermediate- and high-energy proton- and neutron-induced fission cross-sections serve as essential nuclear data for the design and safe operation of the Accelerator-Driven Subcritical System (ADS). This work proposes a hybrid framework combining the Liège Intranuclear Cascade Model (INCL) with Bayesian Neural Networks (BNN) to calculate relevant fission cross-sections. Numerical results are further validated against the simulated data from the INCL-ABLA++ and INCL-GEMINI++ coupled models. The results demonstrate that the established INCL-BNN hybrid framework incorporates adequate physical constraints. It not only accurately reproduces the evolutionary trend of fission cross-sections, but also maintains excellent consistency with available experimental measurements. In energy ranges lacking experimental data, the INCL-BNN model exhibits a physically reasonable predictive trend.