摘要
针对精铜中痕量银元素定量时特征X射线存在背景干扰的问题,采用非对称加权惩罚最小二乘算法(arPLS)对X射线荧光光谱进行基线校正,建立精铜中痕量银元素定量检测模型。通过改变算法平滑参数(λ)和收敛条件(ratio),研究痕量元素X射线光谱基线随参数变化规律,对60组光谱数据采用调整参数并图像监控的方式估计基线。所建立的定量模型不确定度系数R2=0.9862,均方根误差(RMSE)为0.0040 wt%。与不经背景校正算法相比,定量模型R2提高了0.0211,RMSE降低了0.0023 wt%,预测值和参考值平均相对误差缩减了2倍,预测不确定度由0.0028 wt%降低为0.0025 wt%。所建立的定量模型可应用于冶金在线检测、金属回收、合金强化等领域。
To address the background interference of characteristic X-rays in the quantification of trace silver elements in fine copper,the asymmetrically reweighted penalized least squares algorithm(arPLS)was used to perform baseline correction of X-ray fluorescence spectra and to establish a model for the quantification of trace silver element in fine copper.By varying the algorithm smoothing parameterλand convergence condition ratio,the baseline variation law of trace element X-ray spectra with parameters has been studied,and then the baseline has been estimated more accurately for 60 sets of spectral data by adjusting the parameters and image monitoring.The determination coefficient R2 of the established quantitative model is 0.9862 and the root mean square error(RMSE)is 0.0040 wt%.Compared with the algorithm without background correction,the R2 of the quantitative model has improved by 0.0211,the RMSE has decreased by 0.0023 wt%,the average relative error between the predicted and reference values has shrunk by a factor of 2,and the prediction uncertainty has decreased from 0.0028 wt%to 0.0025 wt%.The developed quantitative model will be applied to metallurgy online inspection,metal recycling,and alloy reinforcement.
作者
郝军
李福生
江晓宇
王清亚
杨本永
曹杰
HAO Jun;LI Fusheng;JIANG Xiaoyu;WANG Qingya;YANG Benyong;CAO Jie(Engineering Re-search Center of Nuclear Technology Application,Ministry of Education,East China University of Technology,Nanchang 330013,China;State Key Laboratory of Nuclear Resources and Environment,East China Universi-ty of Technology,Nanchang 330013,China;Yangtze Delta Region Institute(Huzhou),University of Elec-tronic Science and Technology of China,Huzhou 313001,China;Anhui Institute of Optics and Fine Mechan-ics,Chinese Academy of Sciences,Hefei 230031,China)
出处
《分析试验室》
EI
CAS
CSCD
北大核心
2022年第8期904-909,共6页
Chinese Journal of Analysis Laboratory
基金
国家自然科学基金(41875042)项目资助。
关键词
背景扣除算法
X射线荧光分析仪
工作曲线性能
非对称加权惩罚最小二乘算法
background correction algorithm
X-ray fluorescence analyzer
model performance
asymmetrically reweighted penalized least squares algorithm
