Modelling of the DHR heat exchanger using the computer code AC2-ATHLET

Abstract

Recent interest for Small Modular Reactor (SMR) and Advanced Modular Reactors (AMR) ac- celerates research and engineering activities oriented to the design of new reactor concepts, fuel cycle closure and improvement in scientific tools. In this context, code assessment and valida- tion is one of the most relevant research lines in thermal-hydraulics, as well as in reactor physics modelling; promoting benchmarks against measurements and code-to-code comparisons at in- ternational level. This master thesis is realised in the framework of an internship in the start-up newcleo, which was recently founded with the purpose of developing compact and competitive Lead Cooled Fast Reactors (LFR). In the framework of an internship realised in newcleo for the fulfilment of the Master thesis, this study aims at providing a first contribution to the applicability of ATHLET code to lead cooled fast reactors (LFRs) for both primary and secondary systems and their major components. The focus of this work has been dedicated to three different topics. Firstly, the evaluation of the robustness of liquid metal - non-condensable gas interface modelling, with benchmarking activities on two different numerical solving schemes (RELAP-SCDAPSIM and ASYST, semi- implicit; and ATHLET, implicit). Secondly, the reactor’s primary and secondary loops have been modelled by means of ATHLET system code. The physical behaviour of different nodalisation options has been studied, including complex ATHLET modules, in order to achieve the most realistic conditions possible. This included the test and characterisation of ATHLET modules still undergoing validation, such as pseudo-3D modelling capabilities. Finally, the assessment of the heat transfer mechanism in a bayonet tube decay heat removal (DHR) has been carried out, including operation under different conditions. Starting from a base case DHR, optimisation of both boundary conditions and design was performed, resulting in a system capable of extracting decay heat in a safe and steady manner. It is important to note that a special effort on avoiding confidential data has been dedicated, including only open source data and calculated values derived from this data. Results obtained show the capability of the ATHLET code to cope with a liquid metal and non- condensable gas interface; while RELAP was shown to be missing some mitigation scheme to avoid numerical oscillations in physical magnitudes. This artificial oscillations may result in a loss of information about oscillations with a physical origin, if the latest is not strong enough. In addition, the interest of developing a simplified, but representative, test case of the Nuclear Steam Supply System (NSSS), for analysing code behaviour and for young engineers training purposes on a lead cooled reactor has been confirmed. A comparison of steady state variables generated with different models is included, with extra analysis regarding main parameters time evolution and steady state distribution. Finally, the validity of ATHLET was proven as a tool suitable for optimizing heat transfer mechanism and operating a bayonet tube Decay Heat Removal (DHR). Potential improvement have been proposed during the internship, mainly re- garding the loss of regenerative heat exchange between water and steam channels, outside the lead pool (cover gas region). In conclusion, the results provided confirm the interest towards ATHLET code to support Research and Development activities and training of young engineers and researchers.

Outgoing