A Surrogate model for High Temperature Superconducting Magnets to Predict Current Distribution with Neural Network

2025-09-07

AI Summary

To address the prohibitively high computational cost of finite element method (FEM) simulations in multi-physics modeling of large-scale REBCO high-temperature superconducting (HTS) magnets—hindering rapid design iteration—this work proposes a fully connected residual neural network (FCRN)-based surrogate model for efficient and accurate spatiotemporal current density prediction in HTS coils. The architecture comprises 12 residual blocks with 256 neurons per layer, achieving high fidelity within the training domain and maintaining <10% error under 50% extrapolation. Compared to conventional FEM, the model accelerates prediction by several orders of magnitude; even accounting for training overhead, it delivers substantial net efficiency gains. This study represents the first successful application of FCRN to coupled multi-physics surrogate modeling of HTS magnets, establishing a novel paradigm for design optimization of meter-scale HTS magnet systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Large Multimodal Models (LMMs)
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Finite element method (FEM) is widely used in high-temperature superconducting (HTS) magnets, but its computational cost increases with magnet size and becomes time-consuming for meter-scale magnets, especially when multi-physics couplings are considered, which limits the fast design of large-scale REBCO magnet systems. In this work, a surrogate model based on a fully connected residual neural network (FCRN) is developed to predict the space-time current density distribution in REBCO solenoids. Training datasets were generated from FEM simulations with varying numbers of turns and pancakes. The results demonstrate that, for deeper networks, the FCRN architecture achieves better convergence than conventional fully connected network (FCN), with the configuration of 12 residual blocks and 256 neurons per layer providing the most favorable balance between training accuracy and generalization capability. Extrapolation studies show that the model can reliably predict magnetization losses for up to 50% beyond the training range, with maximum errors below 10%. The surrogate model achieves predictions several orders of magnitude faster than FEM and still remains advantageous when training costs are included. These results indicate that the proposed FCRN-based surrogate model provides both accuracy and efficiency, offering a promising tool for the rapid analysis of large-scale HTS magnets.

Problem

• Predicts current distribution in HTS magnets
• Reduces computational cost of FEM simulations
• Enables rapid design of large-scale REBCO systems

Innovation

• Surrogate model using residual neural network
• Predicts current distribution in magnets
• Faster than finite element method