Incorporating empirical knowledge into data-driven variable selection for quantitative analysis of coal ash content by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS)has become a widely used atomic spectroscopic technique for rapid coal analysis.However,the vast amount of spectral information in LIBS contains signal uncertainty,which can affect its quantification performance.In this work,we propose a hybrid variable selection method to improve the performance of LIBS quantification.Important variables are first identified using Pearson's correlation coefficient,mutual information,least absolute shrinkage and selection operator(LASSO)and random forest,and then filtered and combined with empirical variables related to fingerprint elements of coal ash content.Subsequently,these variables are fed into a partial least squares regression(PLSR).Additionally,in some models,certain variables unrelated to ash content are removed manually to study the impact of variable deselection on model performance.The proposed hybrid strategy was tested on three LIBS datasets for quantitative analysis of coal ash content and compared with the corresponding data-driven baseline method.It is significantly better than the variable selection only method based on empirical knowledge and in most cases outperforms the baseline method.The results showed that on all three datasets the hybrid strategy for variable selection combining empirical knowledge and data-driven algorithms achieved the lowest root mean square error of prediction(RMSEP)values of 1.605,3.478 and 1.647,respectively,which were significantly lower than those obtained from multiple linear regression using only 12 empirical variables,which are 1.959,3.718 and 2.181,respectively.The LASSO-PLSR model with empirical support and 20 selected variables exhibited a significantly improved performance after variable deselection,with RMSEP values dropping from 1.635,3.962 and 1.647 to 1.483,3.086 and 1.567,respectively.Such results demonstrate that using empirical knowledge as a support for datadriven variable selection can be a viable approach to improve the accuracy and reliability of LIBS quantification.

Estimating the yield stress of soft materials via laser-induced breakdown spectroscopy

Taking three typical soft samples prepared respectively by loose packings of 77-,95-,and 109-μm copper grains as examples,we perform an experiment to investigate the energy-dependent laser-induced breakdown spectroscopy(LIBS)of soft materials.We discovered a reversal phenomenon in the trend of energy dependence of plasma emission intensity:increasing initially and then decreasing separated by a well-defined critical energy.The trend reversal is attributed to the laser-induced recoil pressure at the critical energy just matching the sample's yield strength.As a result,a one-to-one correspondence can be well established between the samples'yield stress and the critical energy that is easily obtainable from LIBS measurements.This allows us to propose an innovative method for estimating the yield stress of soft materials via LIBS with attractive advantages including in-situ remote detection,real-time data collection,and minimal destructive to sample.

Laser-induced breakdown spectroscopy as a method for millimeter-scale inspection of surface flatness

A non-contact method for millimeter-scale inspection of material surface flatness via Laser-Induced Breakdown Spectroscopy(LIBS)is investigated experimentally.The experiment is performed using a planished surface of an alloy steel sample to simulate its various flatness,ranging from 0 to 4.4 mm,by adjusting the laser focal plane to the surface distance with a step length of 0.2 mm.It is found that LIBS measurements are successful in inspecting the flatness differences among these simulated cases,implying that the method investigated here is feasible.It is also found that,for achieving the inspection of surface flatness within such a wide range,when univariate analysis is applied,a piecewise calibration model must be constructed.This is due to the complex dependence of plasma formation conditions on the surface flatness,which inevitably complicates the inspection procedure.To solve the problem,a multivariate analysis with the help of Back-Propagation Neural Network(BPNN)algorithms is applied to further construct the calibration model.By detailed analysis of the model performance,we demonstrate that a unified calibration model can be well established based on BPNN algorithms for unambiguous millimeter-scale range inspection of surface flatness with a resolution of about 0.2 mm.

Estimating the grain size of microgranular material using laser-induced breakdown spectroscopy combined with machine learning algorithms

Recent work has validated a new method for estimating the grain size of microgranular materials in the range of tens to hundreds of micrometers using laser-induced breakdown spectroscopy(LIBS).In this situation,a piecewise univariate model must be constructed to estimate grain size due to the complex dependence of the plasma formation environment on grain size.In the present work,we tentatively construct a unified calibration model suitable for LIBS-based estimation of those grain sizes.Specifically,two unified multivariate calibration models are constructed based on back-propagation neural network(BPNN)algorithms using feature selection strategies with and without considering prior information.By detailed analysis of the performances of the two multivariate models,it was found that a unified calibration model can be successfully constructed based on BPNN algorithms for estimating the grain size in the range of tens to hundreds of micrometers.It was also found that the model constructed with a priorguided feature selection strategy had better prediction performance.This study has practical significance in developing the technology for material analysis using LIBS,especially when the LIBS signal exhibits a complex dependence on the material parameter to be estimated.

Characteristics of laser-induced breakdown spectroscopy of liquid slag

Rapid online analysis of liquid slag is essential for optimizing the quality and energy efficiency of steel production. To investigate the key factors that affect the online measurement of refined slag using laser-induced breakdown spectroscopy(LIBS), this study examined the effects of slag composition and temperature on the intensity and stability of the LIBS spectra. The experimental temperature was controlled at three levels: 1350℃, 1400℃, and 1450℃. The results showed that slag composition and temperature significantly affected the intensity and stability of the LIBS spectra. Increasing the Fe content and temperature in the slag reduces its viscosity, resulting in an enhanced intensity and stability of the LIBS spectra. Additionally, 42 refined slag samples were quantitatively analyzed for Fe, Si, Ca, Mg, Al, and Mn at 1350℃, 1400℃, and 1450℃.The normalized full spectrum combined with partial least squares(PLS) quantification modeling was used, using the Ca Ⅱ 317.91 nm spectral line as an internal standard. The results show that using the internal standard normalization method can significantly reduce the influence of spectral fluctuations. Meanwhile, a temperature of 1450℃ has been found to yield superior results compared to both 1350℃ and 1400℃, and it is advantageous to conduct a quantitative analysis of the slag when it is in a “water-like” state with low viscosity.

Analysis of Pulverized Coal by Laser-Induced Breakdown Spectroscopy

Laser-induced breakdown spectroscopy （LIBS） has been used to detect atomic species in various enviromnents. The quantitative analysis （C, H, O, N and S） of representative coal samples are being carried out with LIBS, and the effects of particle size are analyzed. A powerful pulse Nd：YAG laser is focused on the coal sample at atmosphere pressure, and the emission spectra from laser-induced plasmas are measured by time-resolved spectroscopy, and the intensity of analyzed spectral lines is obtained through observing the laser plasma with a delay time of 0.4 #s. The experimental results show that the slope of calibration curve is nearly 1 when the concentration of the analyzed element is relatively low, and the slope of curve is nearly 0.5 when the concentration of C is higher than other elements. In addition, using the calibration-free model without self-absorption effect, the results show that the decreasing of particle size leads to an increase of the plasma temperature.

Signal processing for real-time identification of similar metals by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS) is regarded as a promising technique for realtime sorting of scrap metals due to its capability of fast multi-elemental and in-air analysis. This work reports a method for signal processing which ensures high accuracy and high speed during similar metal sorting by LIBS. Similar metals such as aluminum alloys or stainless steel are characterized by nearly the same constituent elements with slight variations in elemental concentration depending on metal type. In the proposed method, the original data matrix is substantially reduced for fast processing by selecting new input variables(spectral lines) using the information for the constituent elements of similar metals. Specifically, principal component analysis(PCA) of full-spectra LIBS data was performed and then, based on the loading plots, the input variables of greater significance were selected in the order of higher weights for each constituent element. The results for the classification test with aluminum alloy, copper alloy,stainless steel and cast steel showed that the classification accuracy of the proposed method was nearly the same as that of full-spectra PCA, but the computation time was reduced by a factor of 20 or more. The results demonstrated that incorporating the information for constituent elements can significantly accelerate classification speed without loss of accuracy.

Detection of heavy metals in water samples by laser-induced breakdown spectroscopy combined with annular groove graphite flakes

The use of laser-induced breakdown spectroscopy(LIBS) for the analysis of heavy metals in water samples is investigated. Some factors such as splashing, surface ripples, extinction of emitted intensity, and a shorter plasma lifetime will influence the results if the water sample is directly measured. In order to avoid these disadvantages and the ‘coffee-ring effect', hydrophilic graphite flakes with annular grooves were used for the first time to enrich and concentrate heavy metals in water samples before being analyzed by LIBS. The proposed method and procedure have been evaluated to concentrate and analyze cadmium, chromium, copper, nickel, lead,and zinc in a water sample. The correlation coefficients were all above 0.99. The detection limits of 0.029, 0.087, 0.012, 0.083, 0.125, and 0.049 mgl^(-1) for Cd, Cr, Cu, Ni, Pb, and Zn,respectively, were obtained in samples prepared in a laboratory. With this structure, the heavy metals homogeneously distribute in the annular groove and the relative standard deviations are all below 6%. This method is very convenient and suitable for online in situ analysis of heavy metals.

Rapid online analysis of trace elements in steel using a mobile fiber-optic laser-induced breakdown spectroscopy system

A mobile fiber-optic laser-induced breakdown spectrometer(FO-LIBS) prototype was developed to rapidly detect a large quantity of steel material online and quantitatively analyze the trace elements in a large-diameter steel tube.Twenty-four standard samples and a polynomial fitting method were used to establish calibration curve models.The R^2 factors of the calibration curves were all above 0.99,except for Cu,indicating the elements’ strong self-absorption effect.Five special steel materials were rapidly detected in the steel mill.The average absolute errors of Mn,Cr,Ni,V,Cu,and Mo in the special steel materials were 0.039,0.440,0.033,0.057,0.003,and0.07 wt%,respectively,and their average relative errors fluctuated from 2.9% to 15.7%.The results demonstrated that the performance of this mobile FO-LIBS prototype can be compared with that of most conventional LIBS systems,but the more robust and flexible characteristics of the FO-LIBS prototype provide a feasible approach for promoting LIBS from the laboratory to the industry.

A comparison of single shot nanosecond and femtosecond polarization-resolved laser-induced breakdown spectroscopy of Al

Aluminum samples have been analyzed by femtosecond polarization-resolved laser-induced breakdown spectroscopy （fs-PRLIBS）. We compare the obtained spectra with those obtained from nanosecond PRLIBS （ns-PRLIBS）. The main specific features of fs-PRLIBS are that a lower plasma temperature leads to a low level of continuum and no species are detected from the ambient gas. Furthermore, signals obtained by fs-PRLIBS show a higher stability than those of ns-PRLIBS. However, more elements are detected in the ns-PRLIBS spectra.

Quantitative analysis of C, Si, Mn, Ni, Cr and Cu in low-alloy steel under ambient conditions via laser-induced breakdown spectroscopy

A diode-pumped solid-state laser （DPSSL） with a high energetic stability and long service life is applied to ablate the steel samples instead of traditional Nd：YAG laser pumped by a xenon lamp, and several factors, such as laser pulse energy, repetition rate and argon flow rate, that influence laser-induced breakdown spectroscopy （LIBS） analytical performance are investigated in detail. Under the optimal experiment conditions, the relative standard deviations for C, Si, Mn, Ni, Cr and Cu are 3.3%-8.9%, 0.9%-2.8%, 1.2%-4.1%, 1.7%-3.0%, 1.1%-3.4% and 2.5%-8.5%, respectively, with the corresponding relative errors of 1.1%-7.9%, 1.0%-6.3%, 0.4%-3.9%, 1.5%-6.3%, 1.2%-4.0% and 1.2%-6.4%. Compared with the results of the traditional spark discharge optical emission spectrometry technique, the analytical performance of LIBS is just a little inferior due to the less stable laser-induced plasma and smaller amount of ablated sample by the laser. However, the precision, detection limits and accuracy of LIBS obtained in our present work were sufficient to meet the requirements for process analysis. These technical performances of higher stability of output energy and longer service life for DPSSL, in comparison to the Q-switch laser pumped by xeon lamp, qualify it well for the real time online analysis for different industrial applications.

A rapid classification method of aluminum alloy based on laser-induced breakdown spectroscopy and random forest algorithm

As an important non-ferrous metal structural material most used in industry and production,aluminum(Al) alloy shows its great value in the national economy and industrial manufacturing.How to classify Al alloy rapidly and accurately is a significant, popular and meaningful task.Classification methods based on laser-induced breakdown spectroscopy(LIBS) have been reported in recent years. Although LIBS is an advanced detection technology, it is necessary to combine it with some algorithm to reach the goal of rapid and accurate classification. As an important machine learning method, the random forest(RF) algorithm plays a great role in pattern recognition and material classification. This paper introduces a rapid classification method of Al alloy based on LIBS and the RF algorithm. The results show that the best accuracy that can be reached using this method to classify Al alloy samples is 98.59%, the average of which is 98.45%. It also reveals through the relationship laws that the accuracy varies with the number of trees in the RF and the size of the training sample set in the RF. According to the laws, researchers can find out the optimized parameters in the RF algorithm in order to achieve,as expected, a good result. These results prove that LIBS with the RF algorithm can exactly classify Al alloy effectively, precisely and rapidly with high accuracy, which obviously has significant practical value.

Time-resolved spectroscopy of femtosecond laser-induced Cu plasma with spark discharge

The combination of spark discharge and laser-induced breakdown spectroscopy (LIBS) is called spark discharge assisted LIBS.It works under laser-plasma triggered spark discharge mode,and shows its ability to enhance spectral emission intensity.This work uses a femtosecond laser as the light souuce,since femtosecond laser has many advantages in laser-induced plasma compared with nanosecond laser,meanwhile,the study on femtosecond LIBS with spark discharge is rare.Time-resolved spectroscopy of spark discharge assisted femtosecond LIBS was investigated under different discharge voltages and laser energies.The results showed that the spectral intensity was significantly enhanced by using spark discharge compared with LIBS alone.And,the spectral emission intensity using spark discharge assisted LIBS increased with the increase in the laser energy.In addition,at low laser energy,there was an obvious delay on the discharge time compared with high laser energy,and the discharge time with positive voltage was different from that with negative voltage.

Study of pressure effects on ocean in-situ detection using laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS) has attracted extensive attention as a new technique for in-situ marine application. In this work, the influence of deep-sea high pressure environment on LIBS signals was investigated by using a compact LIBS-sea system developed by Ocean University of China for the in-situ chemical analysis of seawater. The results from the field measurements show that the liquid pressure has a significant effect on the LIBS signals. Higher peak intensity and larger line broadening were obtained as the pressure increases. By comparing the variations of the temperature and salinity with the LIBS signals, a weak correlation between them can be observed. Under high pressure conditions, the optimal laser energy was higher than that in air environment. When the laser energy exceeded 17 mJ, the effect of laser energy on the signal intensity weakened. The signal intensity decreases gradually at larger delays. The obtained results verified the feasibility of the LIBS technique for the deep-sea in-situ detection, and we hope this technology can contribute to surveying more deep-sea environments such as the hydrothermal vent regions.

Improvement in classification accuracy of stainless steel alloys by laser-induced breakdown spectroscopy based on elemental intensity ratio analysis

Laser-induced breakdown spectroscopy(LIBS) is a useful technique for accurate sorting of metal scrap by chemical composition analysis.In this work,a method for intensity-ratiobased LIBS classification of stainless steel applicable to highly fluctuating LIBS signal conditions is proposed.The spectral line pairs for intensity ratio calculation are selected according to elemental concentration and upper levels of emission lines.It is demonstrated that the classification accuracy can be significantly improved from that of full-spectra principal component analysis or intensity-based analysis.The proposed method is considered to be suited to an industrial scrap sorting system that requires minimal maintenance and low system price.

Determination of Iron in Water Solution by Time-Resolved Femtosecond Laser-Induced Breakdown Spectroscopy

The influence of the energy of femtosecond laser pulses on the intensity of Fe I （371.99 nm） emission line and the continuous spectrum of the plasma generated on the surface of Fe^3＋ water solution by a Ti： sapphire laser radiation with pulse duration 〈45 fs and energies up to 7 mJ is determined. A calibration curve was obtained for Fe3＋ concentration range from 0.5 g/L to the limit of detection in water solution, and its saturation was detected for concentrations above 0.25 g/L, which is ascribed to self-absorption. The 3σ- limit of detection obtained for Fe in water solution is 2.6 mg/L in the case of 7 mJ laser pulse energy. It is found that an increase of laser pulse energy insignificantly affects on LOD in the time-resolved LIBS and leads to a slight improvement of the limit of detection.

Improved Measurement Performance of Inorganic Elements in Coal by Laser-Induced Breakdown Spectroscopy Coupled with Internal Standardization

Laser-induced breakdown spectroscopy was employed to determine the inorganic elements in coal. To improve the measurement＇s accuracy and precision, a new internal standardization scheme, which we named changed internal standardization, was compared with the traditional internal standardization and no internal standardization for the analysis of inorganic elements. The new internal standardization scheme used the atomic line of carbon at 247.86 nm and the molecular band of CN at 388.34 nm and C2 at 516.32 nm to normalize the lines of inorganic elements that were distributed in the same spectral channel. The performance of the utilization of the new internal standardization scheme was evaluated using a set of coal samples, including twenty calibration samples and five validation samples. The results show that the coefficients of determination R2 and the slope of calibration models coupled with changed internal standard- ization are better than that of the calibration models coupled with fixed internal standardization and no internal standardization. Moreover, the measurement accuracy and reproducibility of changed internal standardization for the analysis of five validation samples also yielded further improvement. The results that we obtained suggest that changed internal standardization could compensate for the matrix effects, as well as the influence of the difference in the spectral response of the light collection system.

Rock and Soil Classification Using PLS-DA and SVM Combined with a Laser-Induced Breakdown Spectroscopy Library

Laser-induced breakdown spectroscopy （LIBS） has become a powerful technology in geological applications. The correct identification of rocks and soils is critical to many geological projects. In this study, LIBS database software with a user-friendly and intuitive interface is developed based on Windows, consisting of a database module and a sample identification module. The database module includes a basic database containing LIBS persistent lines for elements and a dedicated geological database containing LIBS emission lines for several rock and soil reference standards. The module allows easy use of the data. A sample identification module based on partial least squares discriminant analysis （PLS-DA） or support vector machine （SVM） algorithms enables users to classify groups of unknown spectra. The developed system was used to classify rock and soil data sets in a dedicated database and the results demonstrate that the system is capable of fast and accurate classification of rocks and soils, and is thus useful for the detection of geological materials.

Laser-induced breakdown spectroscopy applied to the characterization of rock by support vector machine combined with principal component analysis

Laser-induced breakdown spectroscopy（LIBS） is a versatile tool for both qualitative and quantitative analysis.In this paper,LIBS combined with principal component analysis（PCA） and support vector machine（SVM） is applied to rock analysis.Fourteen emission lines including Fe,Mg,Ca,Al,Si,and Ti are selected as analysis lines.A good accuracy（91.38% for the real rock） is achieved by using SVM to analyze the spectroscopic peak area data which are processed by PCA.It can not only reduce the noise and dimensionality which contributes to improving the efficiency of the program,but also solve the problem of linear inseparability by combining PCA and SVM.By this method,the ability of LIBS to classify rock is validated.

Elemental and proximate analysis of coal by x-ray fluorescence assisted laser-induced breakdown spectroscopy

Although laser-induced breakdown spectroscopy(LIBS),as a fast on-line analysis technology,has great potential and competitiveness in the analysis of chemical composition and proximate analysis results of coal in thermal power plants,the measurement repeatability of LIBS needs to be further improved due to the difficulty in controlling the stability of the generated plasmas at present.In this paper,we propose a novel x-ray fluorescence(XRF) assisted LIBS method for high repeatability analysis of coal quality,which not only inherits the ability of LIBS to directly analyze organic elements such as C and H in coal,but also uses XRF to make up for the lack of stability of LIBS in determining other inorganic ash-forming elements.With the combination of elemental lines in LIBS and XRF spectra,the principal component analysis and the partial least squares are used to establish the prediction model and perform multi-elemental and proximate analysis of coal.Quantitative analysis results show that the relative standard deviation(RSD) of C is 0.15%,the RSDs of other elements are less than 4%,and the standard deviations of calorific value,ash content,sulfur content and volatile matter are 0.11 MJ kg,0.17%,0.79% and 0.41%respectively,indicating that the method has good repeatability in determination of coal quality.This work is helpful to accelerate the development of LIBS in the field of rapid measurement of coal entering the power plant and on-line monitoring of coal entering the furnace.