一种两阶段变量选择的LIBS定量分析方法
DOI:
作者:
作者单位:

作者简介:

通讯作者:

基金项目:

国家自然科学基金与中国工程物理学会联合基金项目(No.U1530109);国家自然科学基金项目(No.11972313)资助


A two-stage variable selection method for LIBS quantitative analysis
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
    摘要:

    使用机器学习方法结合激光诱导击穿光谱(Laser Induced Breakdown Spectroscopy,LIBS)进行定量分析,变量选择的结果直接影响最终的定标模型。现有的变量选择方法多存在需要先验知识、计算量庞大等问题,因此提出一种两阶段变量选择方法。第一阶段为排序阶段,以皮尔逊相关系数r为排序准则快速排除与目标元素的浓度无关的变量,保留的变量集合记为S1。第二阶段为搜索阶段,使用近似马尔科夫毯(Approximate Markov Blanket,AMB)排除S1中的冗余变量,保留的变量集合记为S2。为了测试该方法的有效性,将该方法得到的变量集合S2,与偏最小二乘法-变量重要性投影(Partial Least Squares-Variable Importance Projection,PLS-VIP)得到的变量集合S3进行比较。S2和S3分别结合3种机器学习方法建立土壤中锶元素的定量分析模型,结果显示,变量集合S2的3种定标模型决定系数R2均大于0.99,RE均小于5 %,RMSE均小于22 ppm,RSD均小于20 %,显著优于S3的定标模型。表明这种两阶段变量选择方法不仅能够高效的进行变量筛选,也在结合不同机器学习算法进行LIBS定量分析时具有一定普适性。

    Abstract:

    Combining laser induced breakdown spectroscopy(LIBS)with machine learning methods to do quantitative analysis,variable selection is the key.There are many problems in the existing variable selection methods,such as requirement for prior knowledge and large amount of computation,so a two-stage variable selection method is proposed.In the first stage,Pearson correlation coefficient,as a sort criterion,quickly excludes variables independent of the target elements concentration,the set of retained variables is recorded as S1.In the second stage,the Approximate Markov Blanket(AMB)excludes redundant variables in S1 as a search criterion,and the set of reserved variables is S2.To test the effectiveness of the method,the variable set S2 and variable set S3,which is got by the PLS-VIP,were used to establish quantitative analysis models of strontium in soil by combining three types of machine learning methods respectively,and then the quality of S2 can be tested by comparing the prediction ability of the two groups of models.The result shows that all the calibration models obtained by S2 with the R2 of more than 0.99,RE of lower than 5 %,RMSE of lower than 22 ppm and RSD of lower than 20 %,significantly better than the calibration models obtained by S3.As a result,this two-stage variable selection method can not only work effectively,but also owns certain universality when combining with different machine learning methods to do a better LIBS quantitative analysis.

    参考文献
    相似文献
    引证文献
引用本文

郭宇潇,史晋芳,王慧丽,邱荣,邓承付.一种两阶段变量选择的LIBS定量分析方法[J].激光与红外,2021,51(4):435~440
GUO Yu-xiao, SHI Jin-fang, WANG Hui-li, QIU Rong, DENG Cheng-fu. A two-stage variable selection method for LIBS quantitative analysis[J]. LASER & INFRARED,2021,51(4):435~440

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
历史
  • 收稿日期:
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2021-05-11