Speaker
Description
The transfer function of a zero-power thermal reactor was successfully measured by means of a pile noise experiment at low frequency. During a month-long experimental campaign, neutron fluctuations were acquired using excore fission chambers in a sub-critical state of the reactor. The analysis made use of the Cross-Power Spectral Density (CPSD) methodology. Firstly, the prompt neutrons kinetic parameters, such as $\alpha_p$, ${\beta_\text{eff}}$, and $\Lambda$, were obtained from a short measurement in a critical state. It required a power calibration that was carried out using in-core gold foil activation and a calibrated fission chamber. Secondly, these parameters were used to normalize the subcritical transfer function to extract the 8-group delayed neutrons abundances. A substantial effort was put into day-to-day reactivity monitoring in order to allow data averaging. A standard point kinetic model was used to fit that data by means of Bayesian inference and Markov Chain Monte-Carlo programs, namely CONRAD and Stan, respectively. A very good agreement was found between the experimental a$_i$ and the ones computed by TRIPOLI-4 Monte-Carlo transport code along with the JEFF3.1.1 nuclear data library. C/E on the 5th and 8th groups, which are the largest and smallest groups, are $+ 0.9 \pm 6.7$ % and $+4.6 \pm 56$ %, respectively. The uncertainties on prior abundances between 6 to 101 %, held mainly by nuclear data, were lowered between 4 to 56 %, depending on delayed groups. When ENDF/B.VII rev.0 or rev.1 nuclear data were employed, the results of this study suggest a correction of -13 % on the mean lifetime of delayed neutrons.
The CPSD is reported in the Figure below and compared to the 8-group point kinetic model fed with kinetic parameters computed with TRIPOLI-4 $\times$ JEFF3.1.1.