Optimization of flux mapping in-core detector locations in AHWR using clustering approach

Abstract

Measurements from in-core Self Powered Neutron Detectors in the nuclear reactor serve as input to the Online Flux Mapping System which is used for effective monitoring and regulation of the nuclear reactor. It is important to optimize the positioning of in-core detectors given that placement at all desired locations is not pragmatic. In Advanced Heavy Water Reactor (AHWR), which uses thorium-based fuel and boiling light water as coolant, the neutronics and thermal hydraulics are strongly coupled. However, neutronically different regions of the core may be loosely coupled. The knowledge of the 3-D core neutron flux distribution is an important safety feature of online monitoring in AHWR. This paper proposes a data-driven approach based on in-core detector measurements to arrive at the number and optimal locations for in-core detectors in AHWR. K-means clustering algorithm has been applied to in-core detector measurements of AHWR to identify the linear relationship among the detectors and to group them into clusters. Detectors within the clusters are highly correlated with each other and one detector is sufficient to represent others in a cluster, resulting in optimization of the number and locations of in-core detectors. The proposed in-core detector layout has been tested for its effectiveness in flux reconstruction using a Flux Mapping Algorithm in test cases representing typical operational scenarios of AHWR.