Abstract
This work deals with the modeling of homogenized few-group cross sections by Artificial Neural Networks (ANN). A comprehensive sensitivity study on data normalization, network architectures and training hyper-parameters specifically for Deep and Shallow Feed Forward ANN is presented. The optimal models in terms of reduction in the library size and training time are compared to multi-linear interpolation on a Cartesian grid. The use case is provided by the OECD-NEA Burn-up Credit Criticality Benchmark [1]. The Pytorch [2] machine learning framework is used.
Highlights
Few-group cross sections come from transport calculations where the neutronic flux is computed using a detailed discretization in energy and space
The transport code APOLLO2.8 [3] was used to generate cross section data for the OECD-NEA Burn-up Credit Criticality Benchmark, whose material and geometrical specifications are fully available in [1]. It consists of a typical PWR fuel assembly composed of 17×17 UO2 fuel rods with 4% w/o 235U enrichment and with 25 guide tubes
Following literature recommendations we considered a Shallow Artificial Neural Networks (ANN) with number of neurons N = 20 and activation function f (x) = tanh(x) due to its good mapping capabilities, zero mean and smooth output [11]
Summary
Few-group cross sections come from transport calculations where the neutronic flux is computed using a detailed discretization in energy and space. The transport code APOLLO2.8 [3] was used to generate cross section data for the OECD-NEA Burn-up Credit Criticality Benchmark, whose material and geometrical specifications are fully available in [1]. It consists of a typical PWR fuel assembly composed of 17×17 UO2 fuel rods with 4% w/o 235U enrichment and with 25 guide tubes. We analyze the cross sections that capture the majority of the macroscopic cross section behavior (high ratio σirgCi/ i Ciσirg being Ci the concentration of the i-th isotope)
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
