Accuracy improvement of iron ore analysis using laser-induced breakdown spectroscopy with a hybrid sparse partial least squares and least-squares support vector machine model

Y. M. Guo, L. B. Guo, Z. Q. Hao, Y. Tang, S. X. Ma, Q. D. Zeng, S. S. Tang, X. Y. Li, Y. F. Lu, X. Y. Zeng

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

The quantitative analysis of iron ore by laser-induced breakdown spectroscopy (LIBS) is usually complicated due to nonlinear self-absorption and matrix effects. To overcome this challenge, a hybrid sparse partial least squares (SPLS) and least-squares support vector machine (LS-SVM) model was proposed to analyze the content of total iron (TFe) and oxides SiO2, Al2O3, CaO, and MgO in iron ore. In this study, 24 samples were used for calibration and 12 for prediction. Sparse partial least squares was used for variable selection and establishing the multilinear regression model between spectral data and concentrations; LS-SVM was used to fit the residual errors of the SPLS regression model to compensate for the nonlinear effects. The model parameters were determined by using the tenfold cross-validation (CV) method. With the hybrid model, the root-mean-square-error of prediction (RMSEP) values of TFe, SiO2, Al2O3, CaO, and MgO were 0.6242, 0.3569, 0.0456, 0.0962, and 0.2157 wt%, respectively. The results showed that the hybrid model yielded better performance than only the conventional SPLS or LS-SVM model. This study demonstrated that the hybrid model is a competitive data processing method for iron ore analysis using LIBS.

Original languageEnglish (US)
Pages (from-to)1330-1335
Number of pages6
JournalJournal of Analytical Atomic Spectrometry
Volume33
Issue number8
DOIs
StatePublished - Aug 2018

Fingerprint

Ore analysis
Laser induced breakdown spectroscopy
Iron ores
Support vector machines
Mean square error

ASJC Scopus subject areas

  • Analytical Chemistry
  • Spectroscopy

Cite this

Accuracy improvement of iron ore analysis using laser-induced breakdown spectroscopy with a hybrid sparse partial least squares and least-squares support vector machine model. / Guo, Y. M.; Guo, L. B.; Hao, Z. Q.; Tang, Y.; Ma, S. X.; Zeng, Q. D.; Tang, S. S.; Li, X. Y.; Lu, Y. F.; Zeng, X. Y.

In: Journal of Analytical Atomic Spectrometry, Vol. 33, No. 8, 08.2018, p. 1330-1335.

Research output: Contribution to journalArticle

Guo, Y. M. ; Guo, L. B. ; Hao, Z. Q. ; Tang, Y. ; Ma, S. X. ; Zeng, Q. D. ; Tang, S. S. ; Li, X. Y. ; Lu, Y. F. ; Zeng, X. Y. / Accuracy improvement of iron ore analysis using laser-induced breakdown spectroscopy with a hybrid sparse partial least squares and least-squares support vector machine model. In: Journal of Analytical Atomic Spectrometry. 2018 ; Vol. 33, No. 8. pp. 1330-1335.
@article{fede161c928d46339ec77aaffd2849c7,
title = "Accuracy improvement of iron ore analysis using laser-induced breakdown spectroscopy with a hybrid sparse partial least squares and least-squares support vector machine model",
abstract = "The quantitative analysis of iron ore by laser-induced breakdown spectroscopy (LIBS) is usually complicated due to nonlinear self-absorption and matrix effects. To overcome this challenge, a hybrid sparse partial least squares (SPLS) and least-squares support vector machine (LS-SVM) model was proposed to analyze the content of total iron (TFe) and oxides SiO2, Al2O3, CaO, and MgO in iron ore. In this study, 24 samples were used for calibration and 12 for prediction. Sparse partial least squares was used for variable selection and establishing the multilinear regression model between spectral data and concentrations; LS-SVM was used to fit the residual errors of the SPLS regression model to compensate for the nonlinear effects. The model parameters were determined by using the tenfold cross-validation (CV) method. With the hybrid model, the root-mean-square-error of prediction (RMSEP) values of TFe, SiO2, Al2O3, CaO, and MgO were 0.6242, 0.3569, 0.0456, 0.0962, and 0.2157 wt{\%}, respectively. The results showed that the hybrid model yielded better performance than only the conventional SPLS or LS-SVM model. This study demonstrated that the hybrid model is a competitive data processing method for iron ore analysis using LIBS.",
author = "Guo, {Y. M.} and Guo, {L. B.} and Hao, {Z. Q.} and Y. Tang and Ma, {S. X.} and Zeng, {Q. D.} and Tang, {S. S.} and Li, {X. Y.} and Lu, {Y. F.} and Zeng, {X. Y.}",
year = "2018",
month = "8",
doi = "10.1039/c8ja00119g",
language = "English (US)",
volume = "33",
pages = "1330--1335",
journal = "Journal of Analytical Atomic Spectrometry",
issn = "0267-9477",
publisher = "Royal Society of Chemistry",
number = "8",

}

TY - JOUR

T1 - Accuracy improvement of iron ore analysis using laser-induced breakdown spectroscopy with a hybrid sparse partial least squares and least-squares support vector machine model

AU - Guo, Y. M.

AU - Guo, L. B.

AU - Hao, Z. Q.

AU - Tang, Y.

AU - Ma, S. X.

AU - Zeng, Q. D.

AU - Tang, S. S.

AU - Li, X. Y.

AU - Lu, Y. F.

AU - Zeng, X. Y.

PY - 2018/8

Y1 - 2018/8

N2 - The quantitative analysis of iron ore by laser-induced breakdown spectroscopy (LIBS) is usually complicated due to nonlinear self-absorption and matrix effects. To overcome this challenge, a hybrid sparse partial least squares (SPLS) and least-squares support vector machine (LS-SVM) model was proposed to analyze the content of total iron (TFe) and oxides SiO2, Al2O3, CaO, and MgO in iron ore. In this study, 24 samples were used for calibration and 12 for prediction. Sparse partial least squares was used for variable selection and establishing the multilinear regression model between spectral data and concentrations; LS-SVM was used to fit the residual errors of the SPLS regression model to compensate for the nonlinear effects. The model parameters were determined by using the tenfold cross-validation (CV) method. With the hybrid model, the root-mean-square-error of prediction (RMSEP) values of TFe, SiO2, Al2O3, CaO, and MgO were 0.6242, 0.3569, 0.0456, 0.0962, and 0.2157 wt%, respectively. The results showed that the hybrid model yielded better performance than only the conventional SPLS or LS-SVM model. This study demonstrated that the hybrid model is a competitive data processing method for iron ore analysis using LIBS.

AB - The quantitative analysis of iron ore by laser-induced breakdown spectroscopy (LIBS) is usually complicated due to nonlinear self-absorption and matrix effects. To overcome this challenge, a hybrid sparse partial least squares (SPLS) and least-squares support vector machine (LS-SVM) model was proposed to analyze the content of total iron (TFe) and oxides SiO2, Al2O3, CaO, and MgO in iron ore. In this study, 24 samples were used for calibration and 12 for prediction. Sparse partial least squares was used for variable selection and establishing the multilinear regression model between spectral data and concentrations; LS-SVM was used to fit the residual errors of the SPLS regression model to compensate for the nonlinear effects. The model parameters were determined by using the tenfold cross-validation (CV) method. With the hybrid model, the root-mean-square-error of prediction (RMSEP) values of TFe, SiO2, Al2O3, CaO, and MgO were 0.6242, 0.3569, 0.0456, 0.0962, and 0.2157 wt%, respectively. The results showed that the hybrid model yielded better performance than only the conventional SPLS or LS-SVM model. This study demonstrated that the hybrid model is a competitive data processing method for iron ore analysis using LIBS.

UR - http://www.scopus.com/inward/record.url?scp=85051077630&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051077630&partnerID=8YFLogxK

U2 - 10.1039/c8ja00119g

DO - 10.1039/c8ja00119g

M3 - Article

AN - SCOPUS:85051077630

VL - 33

SP - 1330

EP - 1335

JO - Journal of Analytical Atomic Spectrometry

JF - Journal of Analytical Atomic Spectrometry

SN - 0267-9477

IS - 8

ER -