Quantitative analysis of sinters using laser-induced breakdown spectroscopy (LIBS) coupled with kernel-based extreme learning machine (K-ELM)

This work explores the combination of LIBS technology and K-ELM algorithm for the quantitative analysis of total iron (TFe) content and alkalinity of sinter. The main components of sinter ore samples were qualitatively identified from the National Institute of Standards and Technology (NIST) database. 30 sinter ore samples were compressed into pellets and prepared for LIBS measurements. 20 sintered samples were used as calibration samples, and their LIBS spectral data were used as input variables to construct the calibration model, and the other 10 sinter samples were used as test set samples. In order to verify the prediction ability of the sintered sample calibration model, the performance of a kernel-based extreme learning machine (K-ELM) and partial least square (PLS) models were compared by means of root mean square error (RMSE). The experimental results showed that the K-ELM model is superior to the partial least square regression (PLSR) model in quantitative analysis of TFe and alkalinity, both for the calibration set and the test set. Correlation coefficients obtained by the K-ELM model are above 0.9, and the RMSEs are relatively lower. The method proposed in this paper can quickly and effectively realize quantitative analysis of total iron content and alkalinity in a sinter, and can be used for the analysis and control of metallurgical raw materials, thus reducing analysis time and saving production costs.

[1]  Israel Schechter,et al.  Absolute analysis of particulate materials by laser-induced breakdown spectroscopy. , 1997, Analytical chemistry.

[2]  Pak Kin Wong,et al.  Modeling and optimization of biodiesel engine performance using kernel-based extreme learning machine and cuckoo search , 2015 .

[3]  S. J. Tsai,et al.  Spatially resolved, laser-induced breakdown spectroscopy, development, and application for the analysis of Al and Si in nickel-based alloys. , 2006, Analytical chemistry.

[4]  R. Barnes,et al.  Standard Normal Variate Transformation and De-Trending of Near-Infrared Diffuse Reflectance Spectra , 1989 .

[5]  D. L. Death,et al.  Multi-element analysis of iron ore pellets by Laser-induced Breakdown Spectroscopy and Principal Components Regression , 2008 .

[6]  R. Noll,et al.  Laser-induced breakdown spectroscopy—From research to industry, new frontiers for process control , 2008 .

[7]  Chunhua Yan,et al.  Classification of iron ore based on acidity and alkalinity by laser induced breakdown spectroscopy coupled with N-nearest neighbours (N3) , 2016 .

[8]  J. Castillo,et al.  Silicon determination by inductively coupled plasma atomic emission spectrometry after generation of volatile silicon tetrafluoride. , 1998, Talanta.

[9]  Chee Kheong Siew,et al.  Extreme learning machine: Theory and applications , 2006, Neurocomputing.

[10]  F. Feret.  Routine analysis of iron ores by X-ray spectrometry , 1982 .

[11]  Dorian Pyle,et al.  Data Preparation for Data Mining , 1999 .

[12]  Chunhua Yan,et al.  Acidity analysis of iron ore based on calibration-free laser-induced breakdown spectroscopy (CF-LIBS) combined with a binary search algorithm (BSA) , 2016 .

[13]  F. J. Fortes,et al.  The development of fieldable laser-induced breakdown spectrometer: No limits on the horizon , 2010 .

[14]  Reinhard Noll,et al.  Liquid steel analysis with laser-induced breakdown spectrometry in the vacuum ultraviolet. , 2003 .

[15]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[16]  Guanghui Niu,et al.  Classification of iron ores by laser-induced breakdown spectroscopy (LIBS) combined with random forest (RF) , 2015 .

[17]  Hongming Zhou,et al.  Extreme Learning Machine for Regression and Multiclass Classification , 2012, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[18]  Huijuan Lu,et al.  A kernel extreme learning machine algorithm based on improved particle swam optimization , 2017, Memetic Comput..

[19]  R. Noll,et al.  Laser-induced breakdown spectrometry — applications for production control and quality assurance in the steel industry , 2001 .

[20]  David A. Cremers,et al.  Laser-Induced Breakdown Spectroscopy—Capabilities and Limitations , 2009 .

[21]  R. Pattrick,et al.  Acid leaching and dissolution of major sulphide ore minerals: processes and galvanic effects in complex systems , 2004, Mineralogical Magazine.

[22]  J. Gurell,et al.  Laser induced breakdown spectroscopy for fast elemental analysis and sorting of metallic scrap pieces using certified reference materials , 2012 .

[23]  C. Shand,et al.  Portable X-ray fluorescence analysis of mineral and organic soils and the influence of organic matter , 2014 .

[24]  Han Xu,et al.  The basicity analysis of sintered ore using laser-induced breakdown spectroscopy (LIBS) combined with random forest regression (RFR) , 2017 .

[25]  Y F Lu,et al.  Acidity measurement of iron ore powders using laser-induced breakdown spectroscopy with partial least squares regression. , 2015, Optics express.

[26]  L. Cabalín,et al.  Assessment of statistical uncertainty in the quantitative analysis of solid samples in motion using laser-induced breakdown spectroscopy☆ , 2010 .

[27]  Nicoló Omenetto,et al.  Laser-Induced Breakdown Spectroscopy (LIBS), Part II: Review of Instrumental and Methodological Approaches to Material Analysis and Applications to Different Fields , 2012, Applied spectroscopy.

[28]  Z. Yamani,et al.  Study of hazardous metals in iron slag waste using laser induced breakdown spectroscopy , 2007, Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering.

[29]  Steven J. Spencer,et al.  Comparison of principal components regression, partial least squares regression, multi-block partial least squares regression, and serial partial least squares regression algorithms for the analysis of Fe in iron ore using LIBS , 2012 .