Development and Validation of a General-Purpose Neutronics-Thermal Hydraulic Coupling Code Based on preCICE and the OpenFOAM-preCICE Adapter

Abstract: [Background]: The safe operation of nuclear reactors relies on precise simulation of multi-physics couplings, including neutron transport, conjugate heat transfer, and fluid dynamics. Traditional customized coupling programs often suffer from low efficiency and insufficient accuracy, struggling to meet the demands of complex scenarios in advanced reactor designs, such as high-resolution digital twin simulations.[Purpose]:  This study aims to develop a general-purpose three-dimensional nuclear-thermal coupling code with high precision, leveraging the open-source multi-physics coupling library preCICE and its adapter OpenFOAM-adapter. The goal is to enable accurate prediction of safety parameters and in-depth analysis of multi-physics interactions within reactor cores. [Methods]: The proposed framework integrates a neutron transport module and a thermal-hydraulic module. The neutron physics module employs a self-developed neutron transport solver based on the finite volume method, validated against benchmark problems. The thermal-hydraulics module combines a 3D solid heat conduction model (laplacianFoam) and a buoyant turbulent flow model (buoyantPimpleFoam) for detailed analysis of temperature and velocity fields. Through secondary development of the preCICE adapter, the conjugate heat transfer (CHT) module is enhanced to support volume coupling and bidirectional data exchange between neutron and thermal-hydraulics solvers. A typical pressurized water reactor (PWR) single-rod benchmark is used for validation, involving grid independence analysis and comparison of three data mapping methods: nearest-neighbor, nearest-projection, and radial basis function (RBF) interpolation.[Results]: The program accurately outputs key parameters, including 3D power distribution, neutron flux density, velocity fields, and temperature fields. Quantitative verification shows a relative error of less than 0.1% for the coolant outlet temperature and 0.14% for the maximum cladding temperature, satisfying the precision requirements of reactor design codes. Grid independence analysis confirms that a medium-fidelity thermal-hydraulic grid combined with a coarse neutron grid balances accuracy and computational efficiency. Among data mapping methods, nearest-neighbor mapping provides acceptable precision with the lowest computational cost, while RBF interpolation, though more accurate, incurs higher computational overhead.[Conclusions]: The developed program breaks through the limitations of traditional customized coupling approaches, supporting heterogeneous grid configurations and large-scale parallel computing. It offers a robust tool for high-resolution nuclear-thermal coupling simulations, facilitating safety analysis, design optimization, and digital twin modeling of reactor cores.