Data Assimilation Using Subcritical Measurement of Prompt Neutron Decay Constant

The prompt neutron decay constant α in a steady-state subcritical system can be directly measured using a reactor noise analysis method such as the Feynman-α method. To reduce the nuclear data–induced uncertainty of keff for a target system, this study investigates the applicability of data assimilation techniques, i.e., the bias factor method and the cross-section adjustment method, based on a subcritical measurement of α conducted at Kyoto University Critical Assembly (KUCA). The sensitivity coefficients of keff and α with respect to the nuclear data were efficiently estimated using a deterministic SN transport code with first-order perturbation theory. As a result, the a priori relative uncertainty of keff due to the 56-group SCALE covariance data can be reduced if there is strong correlation between the measured α and the target keff. The experimental value of α contributes to improving the nuclear data of total fission spectrum χ and total fission neutron number ν via strong correlations between χ and prompt χp and between ν and prompt νp, by utilizing the sensitivity coefficients of α with respect to prompt χp and νp.