A Study on the Application of Rule-based Refueling Optimization for an Existing CANDU Reactor

The fuel reloading problem has been a goal for fuel management in terms of both safety and economic aspects since the birth of all kinds of nuclear reactors. To determine the fuel reloading strategy and mechanism for each reactor that minimizes unit energy cost under the constraints imposed on various core parameters has been the most significant task in core management. CANDU reactor has a specific refueling mechanism so called on-power refueling, which is to refuel during operation. CANDU reactor refuels on average about 20 fuel bundles (8 fuel bundles per channel) daily to maintain the criticality of reactor during normal operation. Therefore in CANDU reactor, to determine refueling candidates is more important and difficult process. And that is why the system-wise optimal refueling process should be developed for more efficient core management. The objective of the study is to apply a technique, which can be used to determine refueling candidates in an optimum manner, for an existing CANDU reactor. In the study we used combination method, permutation method, and genetic algorithm for the purpose of optimizing refueling candidates, which are based on 'Floppy Rules' announced by Dennis Brissette