Coupled code calculation of rod withdrawal at power accident

• Published in: Nuclear engineering and design 2013-08, Vol.261, p.285-305
• Main Authors: Grgic, Davor, Benfrk, Vesna, Sadek, Sinisa
• Format: Article
• Language: English
• ISSN: 0029-5493; 1872-759X
• DOI: 10.1016/j.nucengdes.2012.10.026

► Sensitivity calculations (withdrawal speed, initial power, secondary side influence) were performed for the rod withdrawal at power accident in PWR. ► Best estimate coupled RELAP5-PARCS code calculation was done, using COBRA code to model the core thermal-hydraulics. ► Specific modelling features included reactor vessel split model, explicit model of the RTD bypass and the overtemperature ΔT setpoint function. ► Average whole core values and the local hot spots were predicted. ► Local fuel centerline temperature and local DNBR were calculated using a COBRA-like model. ► Influence of the burnup on the fuel centerline temperature was studied. The rod withdrawal at power (RWAP) accident is analyzed for NPP Krško as part of activity related to possible resistance temperature detectors (RTDs) bypass removal. The RWAP accident can be departure from nucleate boiling (DNB) or overpower limiting accident depending on initial power level and rate and amount of reactivity addition. In this paper we have analyzed the response of the plant in current configuration to RWAP for different withdrawal speeds and different initial power levels. By demonstrating adequacy of current protection system we can, in the next step, quantify the influence of change in narrow range coolant temperature measurement to available safety margins. The overtemperature ΔT setpoint and its relation to local DNBR values are in center of attention. The coupled RELAP5–PARCS code was used as the calculation tool with the provision to extend the calculation to local pin-by-pin COBRA subchannel calculation for selected state points derived from main coupled code results. In the first part of the calculation methodology, point kinetics calculation is performed using standalone RELAP5 to reproduce USAR results, and in the second part, more demanding coupled code calculation is introduced.