Skip to Main content Skip to Navigation

New evaluation of delayed-neutron data and associated covariances

Abstract : In the context of nuclear reactor physics, one of the main issues is the control of the chain reaction. After a fission event has occurred, around three prompt neutrons are immediately emitted. Luckily, a small fraction of neutrons - less than 1% - appears with a certain delay with respect to the fission event. Delayed neutrons (DN) are an easy way out from the uncontrollable prompt-neutron driven chain reaction because they slow down the response of the reactor to a change in reactivity. Quantifying the fraction of such neutrons as well as their kinetic behavior is essential to design a safe reactor. In particular, in order to determine the safety margins, a rigorous estimation of uncertainties and covariances is necessary. The main consequence of a lack of accurate DN data is the conservatism in the design of a reactor. Nowadays, the discrepancy between the reactivity simulated using different international databases, rises up to 16% for LWRs, and the uncertainties associated with the currently recommended parameters are still too large for the needs of the industry. In addition to that, the covariance matrix associated with the abundances has never been produced. The DN parameters are generally estimated through a calculation or an integral measurement. This Ph.D. aims at producing a new set of DN parameters by performing both calculations and measurements, and by exploiting the two through the Bayesian inference, with the aim of reducing the uncertainties and creating the associated covariance matrices. The results for 235U, together with the calculated values for 238U, have been tested in three benchmarks, giving satisfying solutions with respect to the effective delayed neutron fraction. The main achievement of this work is the great improvement in the predicted reactivity, especially for negative periods, which is accompanied by an extremely good precision, never exceeding 2.2%.
Complete list of metadatas

Cited literature [99 references]  Display  Hide  Download
Contributor : Daniela Foligno <>
Submitted on : Thursday, February 6, 2020 - 2:40:15 PM
Last modification on : Friday, February 7, 2020 - 1:42:00 AM
Long-term archiving on: : Thursday, May 7, 2020 - 5:03:00 PM


Files produced by the author(s)


  • HAL Id : tel-02469257, version 1



Daniela Foligno. New evaluation of delayed-neutron data and associated covariances. Physics [physics]. CEA Cadarache, 13115 SAINT-PAUL-LEZ-DURANCE; Aix Marseille Université, CNRS, Centrale Marseille, ED 353 Sciences pour l'ingénieur, Mécanique, physique, micro et nanoélectronique, 2019. English. ⟨tel-02469257⟩



Record views


Files downloads