首页|Improved training framework in a neural network model for disruption prediction and its application on EXL-50
Improved training framework in a neural network model for disruption prediction and its application on EXL-50
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
万方数据
A neural network model with a classical annotation method has been used on the EXL-50 tokamak to predict impending disruption.However,the results revealed issues of overfitting and overconfidence in predictions caused by inaccurate labeling.To mitigate these issues,an improved training framework has been proposed.In this approach,soft labels from previous training serve as teachers to supervise the further learning process;this has lead to a significant improvement in predictive model performance.Notably,this enhancement is primarily attributed to the coupling effect of the soft labels and correction mechanism.This improved training framework introduces an instance-specific label smoothing method,which reflects a more nuanced model assessment on the likelihood of a disruption.It presents a possible solution to effectively address the challenges associated with accurate labeling across different machines.
neural networkdisruptionsoft labelEXL-50 tokamak
蔡剑青、梁云峰、Alexander KNIEPS、齐东凯、王二辉、向皓明、廖亮、黄杰、阳杰、黄佳、刘建文、Philipp DREWS、徐帅、顾翔、高轶琛、罗宇、李直、the EXL-50 Team
展开 >
Institute of Plasma Physics,Chinese Academy of Sciences,Hefei 230031,People's Republic of China
Forschungszentrum Jülich GmbH,Institut für Energie-und Klimaforschung-Plasmaphysik,Partner of the Trilateral Cluster(TEC),Jülich 52425,Germany
ENN Science and Technology Development Co.,Ltd,Langfang 065001,People's Republic of China
NETL
NSTL
万方数据
