Nuclear reactor reload using Quantum Inspired Algorithm

摘要

Nuclear Reactor Reload Optimization Problem (NRROP), which focuses on the economics and safety of the Nuclear Power Plant (NPP), is a classical problem in Nuclear Engineering that has been studied for more than 40 years. For decades, the NRROP was carried out by specialists that used their knowledge and experience to build configurations of the reactor core aiming at fulfilling the safety regulations of the NPP. Since thel980s, however, researchers have proposed metaheuristics optimization to automate this process. Recently, new researches have shown that Quantum Inspired Evolutionary Algorithms are among the best alternatives to deal with optimization problems in Nuclear Engineering. In the present work, Quantum Evolutionary Algorithm (QEA) is used for optimizing the NRROP of a Brazilian "2-loop" Pressurized Water Reactor (PWR) Nuclear Power Plant, Angra 1. The main goal of this research is to show the performance of QEA to solve the NRROP compared with its classical counterpart, the Genetic Algorithm (CA). In addition, manual reload and other optimization methods are also used to demonstrate the feasibility of QEA to solve cycle 7 of Angra 1. The performance of QEA, such as search ability and fast convergence, is better than GA and compatible with other Quantum Inspired Evolutionary Algorithms presented in the literature.