用于快速垂直不稳定性增长率预测的物理注意力Transformer代理：从Alcator C-Mod到SPARC
Physics Attention Transformer Surrogate for Rapid Vertical Instability Growth Rate Prediction: Alcator C-Mod to SPARC
浏览论文内容
中文总结 AI 辅助
该研究提出物理注意力Transformer（PAT）模型，用于快速预测C-Mod和SPARC平衡态的垂直不稳定性增长率，其误差低于FNO2D、DeepONet等模型，可支撑聚变相关的控制研究。
中文摘要 AI 辅助
本研究针对C-Mod和SPARC平衡态中主导的n=0垂直不稳定性增长率的快速预测问题展开，这类平衡态下的非刚性自由边界响应模型运算速度过慢，无法满足控制周期的使用需求。我们采用在MEQ-FGE-L标签上训练的物理注意力Transformer（Physics Attention Transformer，简称PAT）模型，对标量增长率及对应的二维扰动环向电流密度进行预测。实验结果显示，在保留的C-Mod平衡态上，平均绝对误差为5.4 s⁻¹；在合成的SPARC算例上，平均绝对误差为12.7 s⁻¹，空间本征函数误差接近5%。我们还将PAT与基于算子的机器学习模型FNO2D和DeepONet进行对比，发现PAT的归一化增长率误差显著更低，空间重构效果更优。这些结果表明，PAT能够以控制相关的延迟复现MEQ-FGE-L的输出，可支撑后续关于增长率余量监测和 proximity感知形状控制的研究。
英文摘要
In this work, we investigate rapid prediction of the dominant $n{=}0$ vertical instability growth rate in C-Mod and SPARC equilibria, where nonrigid free boundary response models are too slow for control cycle use. Using a Physics Attention Transformer trained on MEQ-FGE-L labels, we predict both the scalar growth rate and the associated two dimensional perturbed toroidal current density. We find mean absolute errors of 5.4~s$^{-1}$ on held out C-Mod equilibria and 12.7~s$^{-1}$ on synthetic SPARC cases, with spatial eigenfunction errors near 5\%. We also compared PAT with operator based ML models : FNO2D and DeepONet, where we found PAT predicts a much lower normalised growth rate error and improved spatial reconstruction. These results indicate that PAT can reproduce MEQ-FGE-L outputs at control relevant latency and could support future studies of growth rate headroom monitoring and proximity aware shape control.
