Resilience of nuclear fuel cycle scenarios: Definition, method and application to a fleet with uncertain power decrease

Abstract

Nuclear fuel cycle scenario studies can be of benefit for decision-making in nuclear power industry development, thanks to their abilities to make projections of industrial strategies and to assess the associated impacts on the nuclear fuel cycle system. In scenario studies, according to imposed constraints, one proposes trajectories, which are defined as fully characterized evolutions of the studied nuclear fuel cycle system, to represent different prospective developments for the nuclear power industry. However, the scenario studies are usually subject to uncertainties because of the lack of knowledge or information about the future at the time of study. As a result, the real situation can differ from the expectation as the real-world time goes, and the trajectories proposed by the previous scenario studies can be disrupted without satisfying the imposed constraints anymore. The first method to overcome this problem is the resistance strategy, which consists in finding the trajectories that can bear any disruption caused by the uncertainties to meet the imposed constraints. But this resistance strategy may not always be sufficient for scenario studies as the uncertainties in scenario studies are generally so high that the associated impact of disruptions can be too strong to guarantee the existence of said resistant trajectories. As a complementary solution, a resilience strategy is proposed in this study. The goal of this paper is to propose a resilience assessment method for scenario problems. In a resilience study, the resilience of a trajectory is defined as a capability to be readjusted to avoid the constraint violations caused by disruptions. The readjustment is carried out by changing the predesigned levers. A scheme is developed based on the Kriging surrogate models and the state-of-the-art Stepwise Uncertainty Reduction (SUR) algorithm to evaluate the resilience of trajectories. The developed resilience assessment method is applied to a scenario problem with uncertain power decrease as a demonstration. The results show that this method is an efficient tool to deal with the failure of trajectories caused by disruption.