Neural Operator-Based Surrogate Model for CFD:Helical Coil Steam Generator in Small Modular Reactor

2026-05-28

AI Summary

This study addresses the critical need for efficient, high-fidelity thermal-hydraulic simulation in digital twins of small modular reactors (SMRs), where conventional CFD is prohibitively expensive. Focusing on the helical-coil steam generator of the SMART reactor, the authors propose a multiscale surrogate modeling framework that integrates autoencoders with neural operators. For the first time, DeepONet and Fourier Neural Operator (FNO) are applied to CFD-grade transient analysis tailored to SMR-specific geometries. The resulting multiscale L-DeepONet accurately captures the instantaneous periodic dynamics of Karman vortex streets in velocity and pressure fields, while the multiscale FNO effectively predicts time-averaged flow fields and delivers reliable pressure drop estimates. Together, these models establish a complementary paradigm that enables flexible selection based on digital twin requirements.

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 Kármán 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.

Problem

• surrogate model
• computational fluid dynamics
• small modular reactor
• digital twin
• neural operator

Innovation

• neural operator
• surrogate modeling
• multi-scale learning
• helical coil steam generator
• digital twin