Although nuclear reactors are designed to be operated at nominal full power
and steady-state conditions, they also have to withstand incident or accident
situations leading to non steady-state conditions. Such situations arise
typically from equipment malfunction, inappropriate operator action or other
perturbations to the system. Thus, predicting the nuclear reactor behaviour in
such situations is of prime importance. This requires models and methods that
are detailed and accurate enough to represent the dynamic behaviour of the plant
with an acceptable level of confidence and fidelity.
The modelling is challenging, since nuclear power reactors are very complex
systems. The complexity comes first from the size of the system and the variety
of physical and mathematical models that are needed including the strong
heterogeneity of the nuclear core and the strong interaction between the neutron
kinetics and the thermal-hydraulics, and second from the intervention of
external factors (equipment failure, operator action) and the plant automatic
control logic.
Concerning the size of the system, it is customary to combine modelling tools at
different scales: the macro scales (representing large parts of the system in a
coarse manner) and the meso scales (representing some parts of the system in a
relatively detailed and sophisticated manner). Such a combined meso/macro scale
approach has long been used for modelling the time- and space-dependence of the
neutron flux in the nuclear core, where very detailed transport calculations at
the fuel assembly level are first carried out to provide macroscopic
cross-section data to be used in a core simulator. Such a combined meso/macro
scale methodology does not exist yet for the thermal-hydraulic modelling of the
time- and space-dependence of the flow and state fields within the nuclear core.
Furthermore, most of the macroscopic thermal-hydraulic models are based on
diffusive numerical schemes that cannot be applied to fast transients or reactor
instabilities and on constitutive flow regime maps developed for steady-state
conditions.
The Research Area Core and plant dynamics will thus aim at improving the
existing simulation strategies and developing new ones for accurately
representing the dynamics of nuclear reactors.
The emphasis will be on:
the coupling between simulation tools representing different physical
phenomena (neutron kinetics/thermal-hydraulics)
the coupling between simulation tools modelling the same physical
phenomena but at different scales (CFD/system code, transport/diffusion
calculations)
increasing the fidelity of primary system simulation for transients and
accidents. This includes improved reliability, better accuracy and improved
understanding of biases and uncertainties
a better understanding of BWR instabilities, of their occurrence and
thus a better capability of predicting them
a better understanding of the interactions between the fluids in the
system and the physical boundaries of the system (Fluid-Structure
Interaction, FSI)
expanding the knowledge on the range of applicability of existing tools
and methods.
