Using recurrent neural networks for estimation of minor actinides’ transmutation in a high power density fusion reactor

摘要

In this paper, recurrent neural networks (RNNs) were presented for the computation of minor actinides’ transmutation with reactor’s operation period. The results of the RNNs implemented for the computation of the change in the atomic density of minor actinides (237Np, 241Am, 242Cm, 238Pu, 239Pu) and the results available in the literature obtained by using Scale 4.3 (Übeyli, 2004) were compared. The results brought out that the proposed RNNs could provide an accurate computation of the atomic densities of minor actinides of the hybrid reactor with respect to operation period of reactor.