Application of Partial Least Squares Method to Quantitative Analysis of Laser-induced Breakdown Spectra
The basic principles of laser induced breakdown spectroscopy(LIBS)technique,principal component analysis(PCA)method and partial least squares(PLS)method were introduced.The principal component information was extracted from 36 dimensions near the characteristic spectral line of the Pb element,and the 36-dimensional wavelength data were compressed to 2 dimensions,and then the principal component scores of 20 pulses for each sample were used for partial least squares fitting,and after averaging the data,the fitting results were of high quali ty,with the value of the squared fitting coefficient increasing from 0.498 10 to 0.970 00;and the sum of squares of the residuals decreasing from 0.725 29 decreased to 1.363 66*10-4.The PCA method can effectively reduce the space of sample data with certain correlation,and can significantly improve the efficiency of data processing for data with large dimensions,and combined with the PLS method to fit the compressed principal elements,the experimen tal conclusion concludes that PLS is suitable for LIBS quantitative analysis.
laser-induced breakdown spectroscopyelemental Pbprincipal element analysispartial least squares method
NETL
NSTL
万方数据
