Optimization of In-Core Detector Locations in AHWR Using Bond Energy Algorithm

Abstract

Abstract A solution to optimization of in-core detectors placement for Advanced Heavy Water Reactor (AHWR) has been attempted. AHWR houses in-core detector units with Self-Powered Neutron Detectors (SPND) distributed axially and their measurement serves as an input to Online Flux Mapping System (OFMS) to monitor the three-dimensional neutron flux distribution. There is a requirement of placing these in-core detectors at optimum locations to retrieve maximum information about the reactor while keeping their number to the minimum. This paper attempts to optimize SPND placement through the application of Bond Energy Algorithm (BEA), a clustering technique which groups the SPNDs based on correlation. This works on the concept of grouping strongly correlated SPNDs into blocks and choosing one SPND from each block as the optimal location. The higher the uncorrelation among optimal SPNDs, the higher the independent information retrieved about the actual configuration of the reactor. It can be inferred from this work that the number and location of SPNDs are highly dependent on the initial set of SPND locations and the correlation threshold. It can be seen that as the correlation threshold increases, the number of optimal locations increases. The obtained optimal locations have been validated for various operational reactor configurations using different Flux Mapping Algorithms (FMA).