Premium
Application of partial least‐squares calibration methods in low‐resolution EDXRS
Author(s) -
Urbański Piotr,
Kowalska Ewa
Publication year - 1995
Publication title -
x‐ray spectrometry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.447
H-Index - 45
eISSN - 1097-4539
pISSN - 0049-8246
DOI - 10.1002/xrs.1300240209
Subject(s) - partial least squares regression , calibration , linearity , analytical chemistry (journal) , materials science , mineralogy , computer science , mathematics , chemistry , statistics , chromatography , engineering , electronic engineering
The partial least‐squares calibration method was studied on some examples of EDXRS applications: determination of strontium and iron in samples of rock, determination of tin in plated deposits, measurements of ash and sulphur in powdered samples of coal and coating thickness gauging. The spectra of the PLS predictors and the scatter plots of the load and score vectors are presented and an analysis of their interpretation is made. A non‐linearity problem which appeared in some cases was approached either by pretreatment of the entry data or modification of the latent variable. Application of this calibration technique in EXDRS gives two main advantages: (i) it allows systematic factors influencing the final results to be included in the calculated model and (ii) it can provide a new insight into some aspects of the measurement methodology since the data in the following steps of the calibration procedure can be studied and analysed.