Burnup measurement error analysis of HTR fuel spheres using ab-initio Monte-Carlo simulations

Abstract

Multi-pass pebble-bed HTRs need a fast and efficient burnup measurement system (BUMS) for irradiated spherical fuel elements. Typically, for every single pass through the system, we are restricted to 55 h cooling time and 10 s measurement time. These restrictions make a high precision gamma spectrum collection and analysis difficult in general. Using the ab-initio Monte-Carlo simulation approach, we determine the one-sigma relative statistical uncertainty to be ±6.5% at the final 200 MWth MODUL burnup of 9.2% FIMA when the traditional gamma measurement systems are used. This FIMA uncertainty range is originated from the limiting statistics of a very short measurement time and the simultaneous interference of neighboring peaks at so short cooling times. It is the first time that the analysis contains the consideration of a 140La caused Compton shoulder which influenced the background near the Cs-137 peak. Longer cooling times would reduce peak interference and longer measurement times would allow better statistics, as would higher count rates in more advanced gamma counting systems and a narrower width of the four photopeaks in the near 137Cs spectrum. Thereby, the relative statistical error could be brought down to 2–3%, which would be better for engineers. The always existing relative statistical error which can be accepted and recommended in the real measurement should be evaluated in many aspects comprehensively.