Conference Agenda

The Japanese fusion reactor, JA-DEMO, is designed to generate electricity at approximately 300 MW. The amount of enthalpy stored in the reactor's coolant will be significantly larger than that of ITER's. Consequently, we must consider the potential risk of a Loss of Coolant Accident (LOCA) in the JA-DEMO reactor. In this research, we conducted a thermohydraulic LOCA analysis of the Japanese water-cooled DEMO reactor, JA-DEMO, with TRACE code. The U.S. NRC has developed a TRACE code for LOCA analysis of light water reactors. We analyzed three distinct LOCA scenarios: In-Vessel LOCA, Divertor LOCA, and Ex-Vessel LOCA. In the In-Vessel and Divertor LOCA scenarios, water enters the plasma chamber (PC) from the outer blanket and divertor, respectively. In the Ex-Vessel LOCA scenario, water enters the vault from a pipe in the primary cooling system. Initially, we conducted a conservative analysis assuming the maximum break area in each coolant pipe, primarily observing pressure transients in the plasma chamber and vault. Subsequently, we tuned several parameters like pipe break areas for parameter surveys, finding that the break area must be smaller than a threshold in In-Vessel LOCA to maintain the PC pressure below the design pressure of 0.5 MPa. Additionally, we explored optimal component geometries to minimize the impact of LOCA in JA-DEMO.

Magnetohydrodynamics (MHD) investigates the behaviour of conducting fluids interacting with magnetic fields, such as the liquid metals foreseen in the blanket of many fusion reactor designs. The intricated physics involved in MHD scenarios often results in significant computational costs. In this regard, Reduced Order Modelling (ROM) methods may represent a promising solution, as they can approximate complex systems with lower-dimensional yet still-accurate models especially in multi-query and real-time contexts. One of the most famous techniques is the Dynamic Mode Decomposition (DMD), a data-driven algorithm designed to learn the best linear model based on time series datasets. In this work a parametric version is applied, which treats DMD operators as snapshot data, mapping parameter values to modal coefficients. This framework allows for the efficient capture of transient dynamics across a range of parameters, improving computational efficiency and accuracy. This approach is applied to a MHD scenario involving compressible lead-lithium flowing in a channel subjected to different magnetic field intensities, which represent the varying parameter. The channel includes regions on the walls at different temperatures to investigate the effects of various magnetic configurations on the thermo-hydraulics of the liquid metal. This study represents an application of a promising ROM technique to an advanced thermohydraulic scenario, involving conductive fluids influenced by magnetic fields. The results show that the parametric DMD significantly reduces the computational burden while keeping a desired accuracy in predicting the complex MHD flows, highlighting its potential for broader applications in fusion technology and MHD systems.

Inductionless magneto-convection is directly simulated in a Rayleigh-Bénard configuration using the high-fidelity finite-difference solver ‘Xcompact3d’. The Rayleigh numbers considered are in the range whilst the Hartmann numbers (Ha) are in the range 0-1000. Two Prandtl numbers are considered; Pr=0.71 corresponds to air whilst Pr=0.025 corresponds to the liquid LiPb eutectic present in fusion breeder blanket systems

In the presence of no magnetic field the flow is turbulent, chaotic and unsteady. Applying a magnetic field leads to a dramatic reduction in turbulence levels, with steady laminar flow observed in the high Ha limit. Field orientation is a critical factor; wall-parallel fields lead to ‘quasi-2D’ turbulence where there is little variation in the flow along the field direction, whilst wall-normal fields lead to 3D structures that are significantly damped along all three of the spatial directions. In the wall-normal case a significant degradation of the heat transfer performance is observed with increasing Ha whilst the wall-parallel field has little influence on the overall heat transfer performance.

A physics focused analysis is then conducted with a focus on the coherent turbulent structures present in the system, and in particular how the strength of the applied magnetic field influences these structures. The analysis is conducted using conditional averaging to understand the separate roles of ejecting and impacting plumes through the perspective of turbulence budgets. Additionally, a spectral analysis is conducted to understand the most dominant structures in the flow and the roles of these structures in the recurring cycle of magneto-convection.

Achieving fusion reactor design goals require transient analyses to assess conditions during operational and accident scenarios. Analysis of cyclic behavior is essential for plant operation considerations and component lifetime predictions. While transient analysis requires consideration of start-up, shut-down and disruptions, pulsed operation also involves oscillating reactor power. The frequency of oscillations determines thermal inertia and stresses acting on the materials and the blanket. Engineering analysis must therefore include time-dependent loads. Simulating transients are traditionally computationally prohibitive, especially for fluid flow and heat transfer analysis. A major challenge and bottleneck for high fidelity pulsed simulation is turnaround time (currently months for a single ITER discharge). In this work we develop a flexible framework and utilize exascale computing to enable high-fidelity transient simulations.

Accurate modeling of the heat source in the reactor to ensure safe operations is done using tools (OpenMC, MCNP) developed through the FERMI project. While the power level of the fusion reactor determines the magnitude of the resulting neutronic heat deposition, the irradiated structural materials generate decay heat during and after the pulse. These are calculated using the aforementioned tools. Both pulsed and steady state concepts require active cooling during operation, maintenance, and shut-down. Transient thermal hydraulics analysis of the various components is performed to include the decay heat obtained from neutronics. These calculations require frequent data transfer of the volumetric heat deposition from neutronics to the conjugate heat transfer module. Sensitivity of the data exchange frequency will be studied to assess the optimum rate without loss of accuracy.

The Breeding Blanket (BB) is a fundamental component for fusion reactors, responsible power production, neutron shielding, and tritium generation for sustaining fusion. For DEMO, the Water Cooled Lithium Lead (WCLL) and Helium Cooled Pebble Bed (HCPB) designs are leading candidates. ITER plays a pivotal role in validating these BB concepts, using Test Blanket Modules (TBMs) to evaluate their functionality under reactor conditions, providing data on performance, efficiency, and safety.

To provide a validation for WCLL BB concept components, as well as characterization of mock-ups and portions of the BB on a relevant scale, Water Loop (WL) facility is currently under development at the ENEA R.C. Brasimone. The WL facility comprises three thermally coupled loops. The first loop emulates the DEMO Primary Heat Transfer System (PHTS) thermal-hydraulic conditions, operating with water ranging between 295-328°C at 15.5 MPa. This loop is featured with flanges enabling the non-simultaneous connection with different Test Sections (TSs), thereby enhancing its versatility. The TSs can be tested in different operative conditions including inside a Vacuum Chamber (VC), where they undergo irradiation by an electron beam gun aimed at replicating fusion reactor heat flux and simulating the tokamak environment or in connection with a PbLi loop, in order to simulate normal or accidental conditions. The secondary and tertiary loops are primarily tasked with dissipating this power, ultimately exchanging it with a cooling tower.

The present presents a comprehensive overview of the facility layout and requirements, and a RELAP5/Mod3.3 characterization of the facility main thermal-hydraulic parameters under operative conditions.

Tokamaks are inherently pulsed fusion reactors due to the transformer function of the central solenoid inducing plasma current. Non-inductive current methods try to provide steady plasma current, however the efficiency is currently an open issue and therefore their benefit could not compensate the detrimental cost when pursuing steady-state power operation with tokamaks.

An alternative is the so-called Indirect Coupling Design of the Balance of Plant System where the plasma pulsed operation is decoupled from the Power Conversion System of the Fusion Power Plant by using an Intermediate Heat Transfer System (IHTS) hosting an Energy Storage System (ESS). This is actually the reference option selected for the He-cooled EU-DEMO Design.

Presently the HELOKA-Upgrade Storage experimental project is in construction at the Karlsruhe Institute of Technology aiming at studying the behavior of such indirect concept in a mock-up facility. The functionality and operability of the IHTS during normal EU-DEMO operation will be investigated. The first phase of the project consists of a molten salt (MS) loop with an ESS coupled to a water-cooling system acting as heat sink, where the MS loop heat source is an electrical heater. Heat transfer measurements will be performed in a test section undergoing similar conditions as in the Helium-MS Heat Exchanger of EU-DEMO Design. In a later phase, the heat source will be the existing high-temperature Helium loop.

The present paper presents pre-test analysis performed with SIM-code to assess the thermal-hydraulic behavior of the MS and water loops supporting the experimental campaign.

The operation of the ITER reactor will represent a milestone in nuclear fusion research, serving as crucial step towards the realization of commercial fusion energy production by bridging the gap between current research efforts and future industrial-scale deployment. A key component of a fusion reactor is the Breeding Blanket (BB) that must generate tritium fuel, shield the vacuum vessel from high-energy neutrons and transfer the heat generated by the plasma to the power conversion system. One of the proposed BB concepts is the Water-Cooled Lithium Lead (WCLL) which is going to be tested under realistic fusion reactor conditions in ITER in the form of a Test Blanket Module (TBM).

In this framework, at the ENEA R.C. Brasimone the construction of W-HYDRA, an experimental infrastructure dedicated to the investigation of the water and lithium-lead technologies is ongoing. As part of W-HYDRA, Water Loop (WL) facility will investigate the WCLL technology and thus a 1:1 scale mock-up of the WCLL BB TBM will be hosted and experimentally studied. The design characteristics and performance of the TBM will be assessed to provide valuable experimental results in view of ITER operation.

The present paper is focused on the thermal-hydraulic numerical study of the TBM component within WL using RELAP5/Mod3.3. Specifically, the study investigates Normal Operation State conditions (i.e., pulse-dwell and dwell-pulse transients) and accidental scenarios (i.e., Loss Of Feedwater Accident, LOFA), aiming to provide preliminary insights into the operation of WL and investigate the control strategy for conducting the TBM experimental campaigns.