The Primary Heat Transfer System (PHTS) of the water-cooled European (EU) DEMO consists in two independent cooling circuits, the breeding zone (BZ) PHTS and the first wall (FW) PHTS. The configuration under study foresees the presence of an Intermediate Heat Transfer System and an Energy Storage System (IHTS+ESS) in order to operate the turbine during both the pulse (2 hours) and the dwell time (10 minutes) at almost constant load, despite the EU DEMO pulsation of the plasma power. Within the framework of the EUROfusion WPBB research activity, a RELAP5/Mod3.3 system code model is developed to investigate the thermal-hydraulic (TH) performances of the primary cooling systems in case of occurrence of transient conditions belonging to the category of ¿Decrease in Coolant System Flow Rate¿. The nodalization includes the breeding blanket (i.e. breeding zone, first wall, manifolds and connecting pipes), the PHTS of FW and BZ, the Once Through Steam Generators connected with the BZ system and the Water/Molten Salt Heat EXchanger connected with the FW system. The nodalization was initially developed to simulate the nominal conditions with the pulse and dwell phases. Then, starting from the pulse phase, several accidental scenarios are selected for a preliminary evaluation of the TH behavior of the PHTS systems with the aim of the design improvement. The transient results are analyzed to assess the appropriateness of the current PHTS design and to identify needs of modifications or mitigation actions.
The accidental analysis plays a crucial role on determining reactor design principles and components. So far, only two types of analyses involving the BB component were accomplished:
- Detailed steady-state analyses of BB subsystems (single BU, single manifold, etc.), performed with CFD codes (as far as possible) or equivalent numerical approaches (FEM, SYS-TH, etc.);
- Transient analyses of the overall BB PHTS system, performed with fully integrated system codes, such as MELCOR.
The limit of the former type of analyses is that only a small section of the BB component can be analyzed at once and with huge resources in terms of computational time and power. For what concerns the fully-integrated system codes, such as MELCOR, they were developed throughout the last decades to simulate severe accidents involving fission/fusion rectors. These codes are provided with a significant set of additional models to represent chemical reactions, radioisotope transport, hydrogen production, etc. The presence of these algorithms increases the computational time and power required to the single calculation. The only way to contain these parameters is to simplify the input deck simulating the fission/fusion reactors. For the BB component, that has a quite complex design, the model simplification could be significant. The use of the best-estimate system code RELAP5/Mod3.3 is justified to close the gap between these two types of analyses. This code is prevalently a TH system code that allows to perform system level simulations with a higher degree of detail with respect to the fully-integrated codes. Moreover, calculations could be performed with a significant reduction of computational resources. From the TH point of view, the results obtained have the same level of accuracy, if not higher. In fact, the RELAP5 series codes are fully-validated for PWR TH operative conditions, that are the same of DEMO BB PHTS. Hence, this code is strongly recommended for the transient scenarios selected for this simulation activity. The RELAP5 input deck level of detail is lower than the one of CFD analyses. Although, it is sufficient to simulate the main TH phenomena occurring in the single BB subsystems (Bus, Manifolds, etc.). The CFD results could be used to calibrate the single components of RELAP5 model in order to increase the fidelity and reliability of the input. In conclusion, the innovation of this proposal consists in the use, for the first time in the field of fusion reactors, of a code that has the ambitious purpose of modelling in detail complex components such as the BB, perform system level accidental analyses and evaluate the need of design improvements and mitigating actions and strongly reduce the computational resources needed. Moreover, the new features implemented by DIAEE in the modified version, increases significantly the modelling capability of RELAP5/Mod3.3 with respect to fusion reactors. The liquid breeder flow path inside the BB and the HITEC molten salt circuit (IHTS system) can be modelled. New correlations are available to evaluate the heat transfer coefficient in special components, such as BZ OTSGs and FW HEXs.