Four-metal-element quantitative analysis and pollution source discrimination in atmospheric sedimentation by laser-induced breakdown spectroscopy (LIBS) coupled with machine learning

Abstract

The laser-induced breakdown spectroscopy (LIBS) technique coupled with machine learning was proposed to perform four metal elements quantitative analysis and pollution source discrimination in atmospheric sedimentation.

Application of laser-induced breakdown spectroscopy with a generalized regression neural network and LASSO-type methods for estimation of arsenic and chromium in soil

“…Two important parameters of RF that are commonly enhanced by out-of-bag (OOB) estimation are the number of trees ( n tree ) and the number of diverse descriptors attempted at each split ( m try ). 39…”

A hybrid method combining discharge-assisted laser induced breakdown spectroscopy with wavelet transform for trace elemental analysis in liquid targets

“…† Wavelet transform de-noise Noise, baseline dri, and overlapping peaks frequently occur in LIBS, due to the experimental environment, spectral detecting system, and sample complexity, which extremely degrade the quantitative analysis ability. 41 Therefore, it is necessary to preprocess LIBS spectra to improve the quantitative analysis sensitivity. In this work, WT is applied for de-noising spectra.…”

“…As for WTDN, the mother wavelet and decomposition layers are very important parameters and these parameters should be optimized. For LIBS, the db wavelet, [41][42][43] Mexican hat (mexh) wavelet, 44 biorthogonal wavelet, 45 and Symlet wavelet 46 are widely used as mother wavelets. In this work, the db function was selected as the mother wavelet.…”

Novel Method Based on Hollow Laser Trapping-LIBS-Machine Learning for Simultaneous Quantitative Analysis of Multiple Metal Elements in a Single Microsized Particle in Air

“…Zhang et al used machine learning combined with LIBS to quantify/classify urban atmospheric sediment samples (Pb, Cu, Zn, and Al), confirming that it is a promising method to achieve atmospheric sedimentation analysis. 20 At present, there are many multivariate calibration methods, such as partial least squares (PLS), 21 artificial neural network (ANN), 22 support vector machine (SVM), 23 extreme learning machine (ELM), 24 random forest (RF), 25 and so on. RF, regarded as a new classification and regression algorithm on the basis of statistical learning theory, was put forward by Leo Breiman in 2001.…”

“…The multivariate calibration method based on machine learning is an effective way to extract useful information from complex LIBS spectral data and reduce interference factors such as background signals, noise, and overlapping peaks. Zhang et al used machine learning combined with LIBS to quantify/classify urban atmospheric sediment samples (Pb, Cu, Zn, and Al), confirming that it is a promising method to achieve atmospheric sedimentation analysis . At present, there are many multivariate calibration methods, such as partial least squares (PLS), artificial neural network (ANN), support vector machine (SVM), extreme learning machine (ELM), random forest (RF), and so on.…”