Image generated by AI
AI Insight
This study develops AI-based surrogate models using neural operators to enable real-time simulation of fluid dynamics in helical coil steam generators for small modular nuclear reactors. The researchers combined reduced-order models with deep operator networks (DeepONet) and Fourier neural operators (FNO), incorporating multi-scale techniques to accurately predict complex flow patterns including Kármán vortex streets. The multi-scale latent DeepONet successfully captured instantaneous periodic vortex dynamics, while FNO variants provided reliable time-averaged flow predictions and pressure drop estimates.
Why it matters
This work addresses a critical bottleneck in digital twin technology for nuclear reactor operation by replacing computationally expensive CFD simulations with fast AI surrogates that maintain high accuracy. The framework enables real-time monitoring and control of small modular reactors, enhancing their safety and operational efficiency while providing practical guidance for selecting appropriate neural operator architectures based on specific simulation requirements.
Understand the Science
Abstract: Real-time thermal-hydraulic simulation is essential for digital twin (DT) technology that supports the safe and efficient operation of small modular reactors (SMRs). Computational fluid dynamics (CFD) provides high-fidelity flow analysis, but its computational cost prevents direct use in DT applications. AI-based surrogate modeling has been actively investigated to address this limitation, yet neural operator–based surrogates for CFD-level transient analysis of SMR-specific geometries have not been reported. This study presents an integrated framework that combines a reduced-order model (ROM) with neural operators, applied to the helical coil steam generator (HCSG) of the System-integrated Modular Advanced Reactor (SMART). Two ROM strategies tailored to each CFD data type were compared, an MLP-based autoencoder (AE) for unstructured mesh data and a convolutional autoencoder (CAE) for structured mesh data, and each was coupled with the deep operator network (DeepONet) to construct the latent DeepONet (L-DeepONet). The Fourier neural operator (FNO) was additionally adopted for comparison. A multi-scale technique was incorporated into both frameworks to mitigate spectral bias and improve the prediction of K'{a}rm'{a}n vortex streets developing inside the HCSG. The multi-scale L-DeepONet captured the instantaneous periodic vortex dynamics in both velocity and pressure fields, while the FNO and its multi-scale variant predicted the time-averaged mean flow and provided reliable pressure drop estimates. These complementary characteristics provide a practical model-selection guideline that links each architecture to specific DT objectives based on CFD data type and the required level of flow resolution.
Source: Neural Operator-Based Surrogate Model for CFD:Helical Coil Steam Generator in Small Modular Reactor