Use of machine learning for a helium line intensity ratio method in Magnum-PSI
Use of machine learning for a helium line intensity ratio method in Magnum-PSI
复制标题
在 Magnum-PSI 中使用机器学习进行氦线强度比方法
DOI：
10.1016/j.nme.2022.101281
复制
发表时间：
2022
影响因子：
2.6
通讯作者：
Ohno Noriyasu
中科院分区：
文献类型：
--
作者：
Kajita Shin;Iwai Sho;Tanaka Hirohiko;Nishijima Daisuke;Fujii Keisuke;van der Meiden Hennie;Ohno Noriyasu
氦 (He) 线强度的光学发射光谱 (OES) 已用于测量各种等离子体装置中的电子密度 n e 和温度 T e。在这项研究中，引入了具有五个隐藏层的神经网络来模拟线性等离子体设备 Magnum-PSI 中激光汤姆逊散射的 OES 数据与 n e/T e 之间的关系，并与多元回归分析进行比较。结果表明，在 2× 10 18< n e< 8× 1 0 20 m− 3 和 0. 1< T e< 4 eV 范围内，神经网络将预测值（ne 和 T e）的残差降低到多元回归分析的一半以下。我们检查了训练和验证数据的两种不同的数据分割方法，即考虑和不考虑排放单位。分割方法的比较表明，当积累足够的数据集时，即使对于新的放电数据，残差也会降低至~ 10%。
Optical emission spectroscopy (OES) of helium (He) line intensities has been used to measure the electron density, n e, and temperature, T e, in various plasma devices. In this study, a neural network with five hidden layers is introduced to model the relation between the OES data and n e/T e from laser Thomson scattering in the linear plasma device Magnum-PSI and compared to multiple regression analysis. It is shown that the neural network reduces the residual errors of prediction values (n e and T e) less than half those of the multiple regression analysis in the ranges of 2× 10 18< n e< 8× 1 0 20 m− 3 and 0. 1< T e< 4 eV. We checked two different data splitting methods for training and validation data, ie, with and without considering the unit of discharge. A comparison of the splitting methods suggests that the residual error will decrease to∼ 10% even for a new discharge data when accumulating a sufficient data set.