Adapting the feed-back model to dynamic core heterogeneity

The purpose of this work is to develop methodologies for adapting the feed-back model for multiparameter cross-section databases of codes DRAGON and DONJON to dynamic core heterogeneity. Such heterogeneity is exhibited in CANDU reactors during large-break loss-of-coolant accident. In next-generation CANDU designs cooled by light water, the traditional modelling paradigm, ignoring core heterogeneity, is not representative. It leads to very significant errors in estimation of reactivity coefficients and reactor behaviour. Modelling core heterogeneity (i.e. influence on the reactivity and other neutronic parameters for a given channel from the level of voiding in neighbouring channels) and correcting such errors is the subject of the present work.

► Adapting feed-back model for multiparameter cross-section databases to dynamic core heterogeneity.
► Ignoring such core heterogeneity leads to very significant errors in estimating reactivity coefficients, etc. in LBLOCA for CANDU cooled by light water, e.g. CANDU-NG.
► The modified feed-back model is also briefly validated.