Conference Agenda

The unexplored potential of compact heat exchangers for use in light water small modular reactors offers a promising area for improving nuclear technology. Micro-channel heat exchangers provide high thermal efficiency and compact designs, making them suitable for integral designs. However, there is a significant gap in understanding their performance under liquid-boiling conditions, and no comprehensive database currently exists. This highlights the need for more research.

This study focuses on a pre-test analysis of a microchannel heat exchanger from the E-SMR database, developed within the ELSMOR project for a light water small modular reactor. Two models are used to simulate the performance of the heat exchanger: one with RELAP5 and the other with computational fluid dynamics (CFD) using STAR-CCM+. The comparison between the two codes addresses the limitations of thermal-hydraulic system codes like RELAP5 in accurately modeling microchannel heat exchangers, prompting the need for CFD to improve confidence in the simulations.
The pre-test analysis shows good agreement between the models. These will be validated against experimental data from a facility being built at Politecnico Milano to test compact heat exchangers.

The coupling of neutronic and thermal-hydraulic phenomena is important for the multiphysics modelling of the transient behaviour of nuclear reactors (e.g., design basis accidents). In this study, we analyse the behaviour of soluble-boron-free (SBF) water-cooled small modular reactors (SMRs) using coupled neutronic and thermal-hydraulic models. These models are developed using the Serpent Monte Carlo neutron transport and the Sub-channel CFD (SubChCFD) coarse-mesh CFD (computational fluid dynamics) codes. Typical industrial nuclear thermal performance software utilise nodal neutron kinetics and sub-channel nuclear thermal-hydraulic methods to simulate transient behaviour, but these methods oversimplify the geometry and 3D behaviour of coolant flow.

Although CFD models offer a high-fidelity alternative, it is computationally demanding to perform reactor transients. This is due to the fine computational meshes required and the modelling of complex turbulent and multiphase flows. Recently, coarse-mesh computational fluid dynamics (CM-CFD) models have been developed to mitigate this issue. These models utilise a sub-channel-based “filtering” mesh upon which empirical frictional and heat transfer thermal-hydraulic correlations are computed. In addition, the CM-CFD models also solve the Reynolds-Averaged Navier-Stokes (RANS) equations on a coarse computational mesh. However, unlike other CM-CFD approaches, SubChCFD also integrates and utilises experimental data.

This paper uses the coupled model to perform a full-core steady-state simulation of an SBF small modular reactor nuclear fuel assembly design developed by Alzaben et al.

In a pressurized water reactor (PWR), unwanted boron dilution transients can be caused by a large safety or regulating valve opening. When diluted coolant flows into the reactor pressure vessel, the cross-flow of two coolants of unequal boron concentration locally decreases in the core, possibly inducing a prompt change of reactor reactivity with a high impact on safety.

Concerns have focused on the behavior of pressurized water reactors (PWR) operating with soluble boron fluid in the reactor coolant. For the current analysis, a comprehensive understanding of factors influencing flow mixing patterns and boron diffusion in the reactor core is pursued through numerical investigations utilizing 3D computational fluid dynamics (CFD) simulations. These simulations play a crucial role in enhancing reactor operation safety.

A scaled PWR reactor pressure vessel, simulating a one-loop with different initial conditions of flow rates and Reynolds numbers for both the coolant and the safety injection flow into the cold leg. The simulation utilizes a customized solver implemented in OpenFOAM that considers the boron transport model using a transient flow algorithm coupled with a standard k-epsilon turbulence model. Given the magnitude of these simulations. The results provide a 3D mixing pattern under boron dilution transient and agree with experimental data regarding the optimal conditions for the best mixing and diffusion behavior of boron distribution entering the reactor core. This occurs at a specific ratio of injection to Reynolds number in the cold leg.

Small Modular Reactors (SMRs) are gaining importance in addressing energy challenges. To facilitate the wider adoption of these energy systems, it is essential to develop simulation tools that accurately represent the SMR phenomena. In this context, multi-physics and multi-scale analyses provide deeper understanding of SMR behavior under accident conditions. In this study, the US-NRC neutronic core simulator code PARCS, the KIT in-house thermal-hydraulic code TWOPORFLOW, and the US-NRC system thermal-hydraulic code TRACE were utilized. The TRACE/PARCS/TWOPORFLOW coupling code was developed following the ICoCo methodology, which involves exchanging data fields through mesh interpolation. An explicit temporal coupling is implemented, on one hand PARCS and TPF solve the reactor core using a domain-overlapping approach. On the other hand, TRACE solves the rest of the primary circuit and the selected auxiliary systems using a domain-decomposition approach. The NuScale plant has been analyzed using this multi-scale, multi-physics tool, showing good agreement with the reference data. Future work may explore a semi-implicit temporal coupling to enhance simulation stability.

Small modular reactors (SMR) have emerged as next-generation nuclear power systems, offering enhanced safety, efficiency, and economic advantages. Among their critical components, helical coil steam generators (HCSG) have been extensively studied for their effective heat exchange capabilities. However, primary-side cross flow within HCSG could induce vortices and turbulent structures between tubes, resulting in non-uniform heat transfer and flow instabilities that negatively impact overall system stability. Large eddy simulation (LES) based computational fluid dynamics (CFD) can accurately capture this complex behavior but requires fine meshes and short time steps, leading to high computational costs. In this study, a deep learning-based reduced-order modeling (ROM) strategy is proposed to maintain both accuracy and computational efficiency in analyzing local flow regions between HCSG tube layers. Proper Orthogonal Decomposition (POD), Dynamic Mode Decomposition (DMD), and a nonlinear autoencoder are employed to reduce data dimensionality, followed by a Long Short-Term Memory (LSTM) network for predicting flow evolution. These ROM frameworks (POD-LSTM, DMD-LSTM, and Autoencoder-LSTM) are compared to identify the most effective approach for significantly reducing simulation overhead while preserving CFD-level predictive accuracy. The results indicate that linear methods effectively capture dominant features such as large-scale vortex formation and dissipation, whereas the nonlinear autoencoder emphasizes random flow diffusion and chaotic behavior. Notably, the POD-LSTM model demonstrates superior performance in predicting flow field dynamics, achieving higher coefficients of determination (R^2) compared to the other models.

The LUT Heating Experimental Reactor (LUTHER) is a Small Modular Reactor (SMR) concept designed for safe district heating. The project has a heat capacity of 24 MWth, which is sufficient for heating small communities and businesses. To improve the efficiency of heat distribution, it is crucial to locate the plant in close proximity to consumers. As a result, a primary design criterion is to maintain the highest standards of safety.

Many severe accidents in nuclear reactors, such as rapid reactivity injection and core meltdown, are largely prevented by the reactor core's design. To transfer decay heat to the ultimate heat sink after potential accidents, a fully passive system has been developed to transfer heat from the core to the environment through boiling, free convection, condensation and wall conduction, using bedrock as an intermediate heat sink. The presence of non-condensable gases significantly influences heat transfer and steam condensation, making the calculations more complex and design more challenging.

In this paper, heat fluxes in the heat exchanger from containment to bedrock were calculated and visualized using the TRACE version 5 system code software. The effectiveness of employing bedrock as a heat sink was evaluated, and essential design parameters for the heat exchanger were established. These parameters include optimal pipe spacing, the appropriate pipe depth for maintaining a low surface temperature, the pipe length and inclination angle to facilitate efficient condensate flow.