High-fidelity Modeling of Full-scale Pressurized Water Reactor Flow Fields for Machine Learning Applications

2026-05-23

AI Summary

This study addresses the challenge of highly non-uniform coolant flow distribution at the inlet of pressurized water reactor cores, where high-fidelity data are scarce for machine learning–based modeling and sparse sensor optimization. A full-scale, high-fidelity CFD model of the lower plenum and core inlet region is developed, incorporating pump-induced swirl boundary conditions to perform transient simulations and generate a physically consistent three-dimensional flow field dataset. For the first time, this work integrates high-fidelity CFD with spatially aware deep learning architectures—such as 3D CNNs and ConvLSTM—to systematically evaluate their performance in partial field reconstruction and short-term prediction. Results demonstrate that spatiotemporal models like ConvLSTM significantly outperform purely sequential or operator learning approaches, accurately reconstructing assembly-level mass flow rates, with errors primarily localized in the highly turbulent lower regions, thereby validating the method’s effectiveness and physical consistency for complex reactor flow modeling.

Abstract

This work presents a high-fidelity computational fluid dynamics (CFD) and data-driven modeling framework for assembly-level flow characterization in a four-loop pressurized water reactor (PWR). A full lower-plenum and core-inlet domain was constructed using publicly available geometry and operating conditions, enabling transient simulations with pump-induced swirl boundary conditions. The results show that cold-leg swirl and lower-plenum transport generate strongly heterogeneous assembly-wise inlet flow distributions, particularly near the lower core region, while axial resistance and mixing progressively homogenize the flow at higher elevations. These physics-informed datasets were subsequently used to evaluate machine learning (ML) applications for partial field reconstruction and short-term autoregressive prediction. A 3D convolutional-based inpainting model successfully recon-structed missing assembly-level mass flow rates from partial observations, with errors concentrated in the highly turbulent base (bottom) layer and diminishing significantly in upper layers. Comparative analysis across multiple ML models demon-strates that spatially aware architectures, particularly ConvLSTM, significantly outperform sequence-based (LSTM) and operator-learning (DeepONet) approaches by effectively capturing coupled spatio-temporal dynamics. The study also high-lights key challenges, including the sensitivity of inlet flow predictions to turbulence and mesh resolution, as well as the absence of full-scale experimental validation data. Despite these limitations, the results remain consistent with expected physical behavior. Overall, this work establishes high-fidelity CFD as a critical foundation for developing data-driven surrogates, sparse sensing strategies, and future multiphysics coupling frameworks.

Problem

• Pressurized Water Reactor
• Inlet Flow Distribution
• High-fidelity CFD
• Machine Learning
• Full-scale Validation

Innovation

• high-fidelity CFD
• machine learning for nuclear reactors
• spatio-temporal flow reconstruction
• ConvLSTM
• pressurized water reactor