Core-wide multi-physics simulation of fuel behaviour during large-break LOCA using NEXUS

Large-break loss-of-coolant accidents (LB-LOCAs) are among the most critical scenarios in nuclear power plant safety analysis due to their potential to significantly challenge the integrity of barriers confining radioactive material. This study demonstrates, for the first time, the application of the NEXUS framework to perform detailed, core-wide fuel assessments during a design basis accident. NEXUS integrates RELAP for thermal-hydraulics, PARCS for neutronics and ENIGMA for fuel performance to analyse the thermo-mechanical responses of fuel rods across the reactor core during accident conditions. The analysis, applied to a 1240 MWe/3400 MWth Pressurised Water Reactor (PWR) with Optimized ZIRLO cladding and conventional UO${}_2$ fuel, presents a computationally efficient methodology to capture core-wide fuel behaviour — an area where comprehensive approaches remain scarce in the literature, particularly for metrics beyond peak cladding temperature.

An LB-LOCA scenario, specifically a double-ended guillotine break of a cold leg, was modelled as a design basis accident to evaluate fuel integrity under limiting conditions. Four LB-LOCA cases were simulated to assess the impact of varying power profiles and fuel states on cladding rupture stress and plastic hoop strain increment. Enhanced features in the ENIGMA code, including high-temperature clad creep, oxidation, and failure models, as well as a dynamic phase change model, were employed to facilitate margin-to-failure calculations under LOCA conditions. Results highlight relatively benign outcomes at the beginning of cycle (BOC) compared to more challenging end-of-cycle (EOC) scenarios, primarily driven by variations in rod internal pressures.

Both best-estimate and power conservatism cases were studied to assess fuel performance. The successful demonstration of NEXUS’ capabilities opens avenues for employing best-estimate plus uncertainty (BEPU) methodologies in future evaluations. Given its computational efficiency, this study highlights the NEXUS framework’s ability to balance detailed fuel assessments with reasonable computational demands for design basis accidents.