Premium
Determination of some rare earth elements by EDXRF and artificial neural networks
Author(s) -
Schimidt Fernando,
CornejoPonce Lorena,
Bueno Maria Izabel M. S.,
Poppi Ronei J.
Publication year - 2003
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.662
Subject(s) - backpropagation , artificial neural network , rare earth , component (thermodynamics) , regression , matrix (chemical analysis) , biological system , mean squared error , computer science , algorithm , artificial intelligence , analytical chemistry (journal) , chemistry , mineralogy , mathematics , chromatography , statistics , physics , thermodynamics , biology
This paper describes the simultaneous determination of Pr, Nd and Sm by EDXRF spectrometry using mixtures of oxides of these metals in a silica matrix. The data were treated by distinct neural network algorithms: back‐propagation (BP), Levenberg–Marquardt (LM) and two variations of back‐propagation (called BP‐SC, single component, and BP‐MC, multiple component), using results from the PLS model (partial least square regression) for comparison. The best applied model was the BP‐SC neural network, which produced relative standard errors of prediction of 17.5% for Pr, 12.5% for Nd and 12.6% for Sm. Copyright © 2003 John Wiley & Sons, Ltd.