摘要
本文研究两种偏最小二乘法[经典编最小二乘法(CPLS)和基于核心矩阵的偏最小二乘法(KPLS)]同时测定三组分混合物。根据数学原理编制三个程序(SPGRAFA,SPGRPLS和SP-GRKPLS)执行这些计算。八个误差函数用以推断因子数目。因为核心矩阵维数小于原始数据矩阵,所以KPLS法适于计算具有较多光谱数和较少样品数的数据矩阵。实验结果显示对相互重叠的光谱用这两种方法均能获得令人满意且十分吻合的结果。
Two partial least squares methods,classical partial least squares(PLS)and partial least squares based on a kernel algorithm (KPLS),were studied for simultaneous determination of a three component mixture.Three programs called SPGRAFA,SPGRPLS and SPGRKPLS were designed to perform the calculations.Eight error functions were calculated for deducing the number of factors.Because the size of the kernal matrix was much smaller than the original data matrix,the KPLS applied to calculating the matrix with many wavelengths and fewer number of samples.Experimental results showed both methods to be successful even there was overlap of spectra and agreed well.
出处
《光谱学与光谱分析》
SCIE
EI
CAS
CSCD
北大核心
1997年第5期115-120,共6页
Spectroscopy and Spectral Analysis
基金
国家自然科学基金
关键词
多组分测定
偏最小二乘法
三组分
分光光度法
Simultaneous spectrophotometric determination, Multicomponent determination, Partial least squares