Comparison of three methodologies for optimization of the layout of in-core detectors with a specific number of detectors

Abstract

The core monitoring system (CMS) is one of the most important systems in safety of an operational reactor. The detectors arrangement has a significant effect on the evaluation of the reactor power distribution and thermal hydraulic parameters. In this paper, three optimization methodologies are explored under a specific number of detectors to achieve the appropriate detector arrangement. These methods consist of the knight moving law (KML) method, the random forest (RF) method and the correlation coefficient (CC) method. The information theory is used to evaluate the performance of these optimization methods. At last, the CC method can obtain the best detector arrangement with smallest mutual information. Meanwhile, with RF method it may be difficult to obtain the best solution, comparing with the other methods. Although the information entropy method provides a glimpse of the detector arrangement, a more comprehensive evaluation needs to be further refined in engineering practice, taking into account issues such as the influent of detector failure and detector random error, and sensitivity analysis under more reactor operating conditions.