Revisiting Heat Flux Analysis of Tungsten Monoblock Divertor on EAST using Physics-Informed Neural Network

2025-08-05

AI Summary

To address the challenge of real-time heat flux estimation for the tungsten monoblock divertor on the EAST tokamak—where conventional finite element methods incur prohibitive computational cost and fail to meet online simulation requirements—this work proposes a mesh-free thermal conduction modeling approach based on physics-informed neural networks (PINNs). The governing heat conduction equation is incorporated as a soft constraint into the loss function, jointly optimized with boundary/initial conditions and sparse experimental measurements. Validated under both uniform and non-uniform heating scenarios, the model achieves accuracy comparable to finite element simulations (relative error <3%) while accelerating computation by 40×, enabling millisecond-level real-time prediction of divertor heat flux density for the first time. We release a high-quality experimental dataset and fully reproducible training code, establishing an AI-augmented, physics-guided modeling paradigm for fusion engineering applications.

Technology Category

• Machine Learning: Hardware-aware ML
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

Estimating heat flux in the nuclear fusion device EAST is a critically important task. Traditional scientific computing methods typically model this process using the Finite Element Method (FEM). However, FEM relies on grid-based sampling for computation, which is computationally inefficient and hard to perform real-time simulations during actual experiments. Inspired by artificial intelligence-powered scientific computing, this paper proposes a novel Physics-Informed Neural Network (PINN) to address this challenge, significantly accelerating the heat conduction estimation process while maintaining high accuracy. Specifically, given inputs of different materials, we first feed spatial coordinates and time stamps into the neural network, and compute boundary loss, initial condition loss, and physical loss based on the heat conduction equation. Additionally, we sample a small number of data points in a data-driven manner to better fit the specific heat conduction scenario, further enhancing the model's predictive capability. We conduct experiments under both uniform and non-uniform heating conditions on the top surface. Experimental results show that the proposed thermal conduction physics-informed neural network achieves accuracy comparable to the finite element method, while achieving $ imes$40 times acceleration in computational efficiency. The dataset and source code will be released on https://github.com/Event-AHU/OpenFusion.

Problem

• Estimating heat flux in EAST fusion device efficiently
• Replacing grid-based FEM with Physics-Informed Neural Network
• Achieving high accuracy with 40x faster computation

Innovation

• Physics-Informed Neural Network for heat flux
• Combines boundary, initial, and physical losses
• 40x faster than Finite Element Method