Efficient weakly supervised LIBS feature selection method in quantitative analysis of iron ore slurry.

On-stream analysis of the element content in ore slurry plays an important role in the control of the mineral flotation process. Therefore, our laboratory developed a LIBS-based slurry analyzer named LIBSlurry, which can monitor the iron content in slurries in real time. However, achieving high-precision quantitative analysis results of the slurries is challenging. In this paper, a weakly supervised feature selection method named spectral distance variable selection was proposed for the raw spectral data. This method utilizes the prior information that multiple spectra of the same slurry sample have the same reference concentration to assess the important weight of spectral features, and features selected by this prior can avoid over-fitting compared with a traditional wrapper method. The spectral data were collected on-stream of iron ore concentrate slurry samples during the mineral flotation process. The results show that the prediction accuracy is greatly improved compared with the full-spectrum input and other feature selection methods; the root mean square error of the prediction of iron content can be decreased to 0.75%, which helps to realize the successful application of the analyzer.

Model transfer method based on piecewise direct standardization in laser-induced-breakdown spectroscopy.

A large number of certified samples are usually required to build models in the quantitative analysis of complicated matrices in laser-induced-breakdown spectroscopy (LIBS). Because of differences among instruments, including excitation and collection efficiencies, a quantitative model made on one instrument is difficult to apply directly to other instruments. Each instrument requires a large number of samples to model, which is very labor intensive and will hinder the rapid application of the LIBS technique. To eliminate the differences in spectral data from different instruments and reduce the cost of building new models, a piecewise direct standardization method combined with partial least squares (PLS_PDS) is studied in this work. Two portable LIBS instruments with the same configuration are used to obtain spectral data, one of which is called a master instrument because its calibration model is directly built on a large number of labeled samples, and the other of which is called a slave instrument because its model is obtained from the master instrument. The PLS_PDS method is used to build a transfer function of spectra between the master instrument and slave instrument to reduce the spectral difference between two instruments, and thus one calibration model can adapt to different instruments. Results show that for multiple elemental analyses of aluminium alloy samples, the number of samples required for slave modeling was reduced from 51 to 14 after model transferring by PLS_PDS, and the quantitative performance of the slave instrument was close to that of the master instrument. Therefore, the model transfer method can obviously reduce the sample number of building models for slave instruments, and it will be beneficial to advance the application of LIBS.

An Image Auxiliary Method for Quantitative Analysis of Laser-Induced Breakdown Spectroscopy.

Improving both the stability and accuracy of laser-induced breakdown spectroscopy (LIBS) is an issue in quantitative analysis. For certain environments outside of the laboratory, consistently and exactly maintaining the distance from the optical system to the sample surface is difficult, and fluctuations of this distance severely affect the stability of the spectrum. In this work, the principal components of the plasma images are extracted and used to correct the spectral line intensities as an auxiliary method to reduce spectral fluctuation. The presented image auxiliary method is combined with univariate analysis and multivariate analysis, and the element concentrations of Cu, Mn, V, and Cr in steel samples are analyzed. For univariate analysis, all the determination coefficients ( R2) of the four elements exceed 0.99, whereas the average relative standard deviations (RSDs) of the intensities decrease from 30.45, 23.14, 27.03, and 22.04%, to 2.13, 3.38, 2.49, and 3.58%, respectively. For the multivariate analysis, the R2 values for Cu, Mn, V, and Cr also all exceed 0.99, and the average RSDs of the predicted concentrations of the validation samples decrease to 2.87, 3.82, 2.86, and 6.51%, respectively.

[Automatic Method for Selecting Characteristic Lines Based on Genetic Algorithm to Quantify Laser-Induced Breakdown Spectroscopy].

Selecting proper characteristic lines from enormous spectral intensities is crucially important to implement quantitative analysis of Laser-Induced Breakdown Spectroscopy using internal standard method. Manual selecting of characteristic lines by researchers is time consuming and energy consuming, which cannot guarantee the best result. An automatic method to select analytical and reference lines for internal standard method from the original spectra based on Genetic Algorithm was proposed in this paper. This method was utilized to select analytical and reference lines for internal standard methods from LIBS of Mn, Ni, Cr, Si and Fe of low alloy steels. The optimal characteristic lines optimized by this method were the analytical line 403.306 8 nm of Mn and the corresponding reference line 368.745 7 nm of Fe, the analytical line 288.157 7 nm of Si and the corresponding reference line 427.176 1 nm of Fe, the analytical line 286.510 0 nm of Cr and the corresponding reference line 272.753 9 nm of Fe and the analytical line 352.453 6 nm of Ni and the corresponding reference line 358.698 5 nm of Fe, respectively. Then these elements were quantified by the internal standard method using these selected lines. The results showed that this proposed method for selecting characteristic lines can automatically select the optimal analytical and reference lines and could guarantee the best quantitative result obtained by internal standard method.

[In-Situ Analysis of Solid Steel Samples with Remote Double-Pulse Laser-Induced Breakdown Spectroscopy System].

In order to realize real-time, online monitoring of the component of steel and other metal smelting process, we designed a remote double-pulse laser-induced breakdown spectroscopy (LIBS) analysis system which can realize non-contact remote measurement and component analysis for long distance sample. The paper first tests the system on solid standard steel samples, which provides basis for online monitoring the component of molten steel. The experimental results show：laser focal spot is about 1mm in long distance; double-pulse ablation depth is deeper than single pulse's; the optimum delay of double-pulse is non-consistent in different distances; the enhancement effect of double- pulse in 3.1 m is better than that in 2.1 m，and the maximum enhancement is 5.19 of Ti(Ⅰ) 319.99 nm; the calibration curve of R2 is about 0.99, RSD being less than 5%, RMSE being less than 0.021%, LOD being less than 500 ppm for most elements in 2.1 m, which is better than that in 3.1 m.

[Quantitative analysis of alloy steel based on laser induced breakdown spectroscopy with partial least squares method].

In the present paper both the partial least squares (PLS) method and the calibration curve (CC) method are used to quantitatively analyze the laser induced breakdown spectroscopy data obtained from the standard alloy steel samples. Both the major and trace elements were quantitatively analyzed. By comparing the results of two different calibration methods some useful results were obtained: for major elements, the PLS method is better than the CC method in quantitative analysis; more importantly, for the trace elements, the CC method can not give the quantitative results due to the extremely weak characteristic spectral lines, but the PLS method still has a good ability of quantitative analysis. And the regression coefficient of PLS method is compared with the original spectral data with background interference to explain the advantage of the PLS method in the LIBS quantitative analysis. Results proved that the PLS method used in laser induced breakdown spectroscopy is suitable for quantitative analysis of trace elements such as C in the metallurgical industry.

Study on laser-induced breakdown spectroscopy in high-pressure helium gas environment for deep ocean applications.

Laser-induced breakdown spectroscopy (LIBS) has gained wide acceptance as an in situ detection technique for elements. However, in deep-sea applications, the sensitivity of LIBS detection will be reduced due to the high-pressure environment and the nature of water. To address the negative effects of high-pressure water, this study used the method of draining the water from the sample surface by passing high-pressure helium gas, which allowed the plasma excitation environment to be converted from high-pressure water to high-pressure gas. The available spectral signals of solid samples at 60 MPa gas pressure were obtained for the first time, and the peak intensity and spectral broadening of the spectral lines at different pressures were analyzed for comparison. We found a nonlinear decrease in the spectral intensity and a gradual increase in the spectral broadening during the pressure increase. We also investigated the effect of laser energy on the intensity and width of the spectral lines in a high-pressure helium environment and found that increasing the laser energy in a high-pressure environment enhanced the spectral intensity and that the change in laser energy almost did not affect the line width. Finally, by observing the plasma images at different pressures with different energies, this study found that the laser penetrated the high-pressure helium gas in advance and leaved a column of light in the gas, and the plasma was slowly made smaller as the ambient pressure increases. This finding explained the cause of laser energy loss and demonstrated that the LIBS signal intensity can be improved by increasing the laser energy and shortening the laser transmission distance in a high-pressure gas environment.

[Research on the analytical line auto-selection for quantitative analysis of materials with laser-induced breakdown spectroscopy].

To realize auto-selection of analytical lines for quantitative analysis of materials with laser-induced breakdown spectroscopy, two parameters, i. e. the relative detected-to-theory intensity ratio (RDTIR) and wavelength difference of detected and theory (WDDT) were defined. The spectral lines seriously disturbed by self-absorption and spectral interference were excluded automatically by setting reasonable thresholds of RDTIR and WDDT. By analyzing the experimental data of high-alloy steel (GBW01605), the analytical lines of iron (Fe), chromium (Cr), nickel (Ni), manganese (Mn) and copper (Cu) were selected, and the results were in line with the principle of lines selection.

Composition analysis of ceramic raw materials using laser-induced breakdown spectroscopy and autoencoder neural network.

In the ceramic production process, the content of Si, Al, Mg, Fe, Ti and other elements in the ceramic raw materials has an important impact on the quality of the ceramic products. Exploring a method that can quickly and accurately analyze the content of key elements in ceramic raw materials is of great significance to improve the quality of ceramic products. In this work, laser-induced breakdown spectroscopy (LIBS) is used for rapid analysis of ceramic raw materials. The chemical element composition and content of ceramic raw materials are quite different, which leads to serious matrix effects. Building an artificial neural network model is an effective way to solve the complex matrix effects, but model training can easily lead to overfitting due to the high number of spectral features and the limited number of samples. In order to solve this problem, we propose a feature extraction method that combines the linear regression (LR) and the sparse and under-complete autoencoder (SUAC) neural network. This LR + SUAC method performs nonlinear feature extraction and dimension reduction on high-dimensional spectral data. The spectral data dimension is reduced from 8188 to 100 through the LR layer, and further reduced to 32 through the SUAC encoding layer. Further, a quantitative analysis model for the elemental composition of ceramic raw materials is established by the combination of LR + SUAC and Back Propagation Neural Network (BPNN). Since the input data dimension and redundant information are greatly reduced by LR + SUAC, the overfitting problem of BPNN is greatly reduced. Experiment results showed that the LR + SUAC + BPNN method obtained the best quantitative analysis performance compared with several other methods in the cross-validation process.

[Quantitative analysis of Mn and Si of alloy steels by laser-induced breakdown spectroscopy].

The concentration of Mn and Si in different kinds of steels was determined by laser-induced breakdown spectroscopy (LIBS). The multivariate quadratic nonlinear function was adopted for calibration. Samples including common alloy steels, stainless steels and carbon tool steels were analyzed. The matrix effect was serious because of large difference in compositions of different kinds of steels and strong line overlaps in steel spectra. Therefore, the common calibration methods that only use one analytical line to calibrate the complex chemical compositions of alloy steels will lose much information. The multivariate calibration methods, however, can utilize more information of spectra, successfully reduce the matrix effect and improve the measurement repeatability and accuracy of LIBS. Compared with the common calibration method based on one analytical line, the relative standard deviation was reduced from above 20% to below 10%, and the accuracy was increased by more than 5 times for Mn and more than 6 times for Si.

[Simultaneous quantitative analysis of multielements in Al alloy samples by laser-induced breakdown spectroscopy].

The multielement components of some aluminium alloy samples were quantified by using laser-induced breakdown spectroscopy (LIBS). The Nd : YAG pulsed laser was used to produce plasma in ambient air. The spectral range of 200-980 nm was simultaneously obtained through a multichannel grating spectrometer and CCD detectors. The authors studied the influences of time delays, energy of the laser, and depth profile of elements in samples on spectral intensity, and optimized the experimental parameters based on the influence analysis. With the optimal experimental parameters, the authors made the calibration curves by four certified aluminum alloy samples for eight elements, Si, Fe, Cu, Mn, Mg, Zn, Sn, and Ni, and quantified the composition of an aluminum sample. The obtained maximum relative standard deviation (RSD) was 5.89%, and relative errors were--20.99%-15%. Experimental results show that LIBS is an effective technique for quantitative analysis of aluminum alloy samples, though the improved accuracy of the quantitative analysis is necessary.

A method for resolving overlapped peaks in laser-induced breakdown spectroscopy (LIBS).

Spectral peak overlapping is a basic problem in analytical data processing of laser-induced breakdown spectroscopy (LIBS). Curve fitting is the typical method of resolving overlapped peaks. For preventing ambiguous fitting, appropriate initial values must be known. The aim of this work was to present a method that could be used to determine appropriate initial values of the curve-fitting method by using fractional differential theory. According to the variation of characteristic points of Lorentzian peaks at different fractional differential orders, parameter estimators were obtained that were used to calculate the initial values of the curve-fitting method. As it is a widely used optimization method, the Levenberg-Marquardt method was used in curve fitting. Simulation and LIBS experimental results proved that the proposed method of the initial value estimation can effectively resolve the overlapped peaks in LIBS data processing.

Correction of self-absorption effect in calibration-free laser-induced breakdown spectroscopy by an internal reference method.

A simplified procedure for correcting self-absorption effect was proposed in calibration-free laser-induced breakdown spectroscopy (CF-LIBS). In typical LIBS measurement conditions, the plasma produced is often optically thick, especially for the strong lines of major elements. The selection of self-absorption lines destroys the performance of CF-LIBS, and the familiar correction method based on the curve of growth is complex in implementation. The procedure we proposed, named internal reference for self-absorption correction (IRSAC), first chose an internal reference line for each species, then compared other spectral line intensity of the same species with the reference line to estimate the self-absorption degrees of other spectral lines, and finally achieved an optimal correction by a regressive algorithm. The self-absorption effect of the selected reference line can be ignored, since the reference line with high excitation energy of the upper level is slightly affected by the self-absorption. Through the IRSAC method, the points on the Boltzmann plot become more regular, and the evaluations of the plasma temperature and material composition are more accurate than the basic CF-LIBS.