The design of nuclear reactors is based on a balance between performance and respect for safety limits. To ensure that these safety limits are not exceeded, safety margins are set. The definition of these safety margins depends on the uncertainties associated with the simulation of the physical parameters that assess nuclear safety during the design phase. If these uncertainties are accurately assessed, the safety margins will be set at the best level, allowing high performance and economic benefits with guaranteed safety. There remains significant potential to enhance the uncertainty assessments in multiphysics simulations for Light-Water Small Modular Reactors (LW-SMRs), particularly in steady-state, depletion and transient/accident scenarios. LW-SMRs are of considerable international interest, as reflected by the various designs currently under development. New LW-SMR reactor designs are frequently categorized as Generation III+ and typically feature the following design characteristics: small cores (leading to a high rate of neutron leakage); heterogeneous cores (involving many neutron absorbers when cores are soluble boron free); use of passive safety systems to mitigate accidents; and, in some cases, natural circulation for cooling. These new features affect the choices of simulation options and require adapting the calculation schemes, which are conventionally implemented for large reactors.
To assess the predictive capabilities of modelling and simulation (M&S) packages for LW-SMRs, the Expert Group on Reactor Systems Multi-Physics (EGMUP) under the auspices of the Working Party on Scientific Issues and Uncertainty Analysis of Reactor Systems (WPRS) launched in 2025 the "Light-water cooled small modular reactor benchmark for uncertainty quantification and propagation in multiphysics simulations" (LW-SMR benchmark). This benchmark will compare the performance of models of different fidelity, including traditional improved (best-estimate) systems and novel high fidelity (best-effort) systems. The latter will serve as reference solutions and provide input to High-to-Low fidelity information schemes for improving traditional calculations. Also, compatibilities of different simulation packages will be considered. The uncertainty assessment will include model and numerical uncertainties along with other sources of uncertainties.
The goal of this benchmark is to improve methods to model and to assess simulation uncertainties for LW-SMRs by proposing progressive numerical benchmarking and uncertainty propagation exercises. The benchmark includes two different configurations:
The activity will consist of four phases for each of the two configurations including benchmark specifications and benchmark execution:
A stopping point (project gate) is planned at the end of Phase II in June 2027, in order to assess the possibility of launching the next phases, depending on participation and results of Phase II, as well as the number of expected participants for Phases III and IV.
To participate in the benchmark and to gain access to the working area, please e-mail: wprs@oecd-nea.org.
Details, including specifications and results templates can be found in the working area:
LW-SMR Benchmark (requires password | reminder)
