Optimization under uncertainty for robust fuel cycle analyses

Abstract

Optimization of nuclear fuel cycles is essential for experts and policy makers for studying and analyzing the future of the nuclear energy. In the case of advanced electronuclear transition scenarios, multiple parameters with a complex dependence have to be fine-tuned in order to achieve a set of predefined objectives. However, in the presence of uncertainties the solutions obtained in this way may not be stable since small perturbations could break the delicate balance between different parts of the scenario. In this work, the optimization of an uncertainty European-based sustainable transition scenario has been studied. This scenario, which has been analyzed with the TR_EVOL nuclear fuel cycle simulator system, is aimed at reducing the transuranic inventory masses while keeping the fuel cycle costs. To that end, an extension of the DEMO evolutionary multiobjective algorithm has been implemented within TR_EVOL for allowing the inclusion of constraints and uncertainties with a methodology that can be used by any fuel cycle simulator. Results show the importance of coupling optimization and uncertainty analyses due to the suboptimal and unstable solutions that can be obtained if not considered jointly. In addition, the uncertainties shrink the decision space. It was found that in their presence the transuranic mass can be reduced and stabilized by a factor ranging between 65% and 71% with an increase of the cost of 16% and 18.5% after 300 years of operation by using advanced systems when compared with an open fuel cycle strategy.