Quantitative Analysis Of Nonmetallic Elements In Coal By Laser Induced Breakdown Spectroscopy

Coal is one of the main energy sources of traditional industry and people's life.The demand for coal resources in China is also increasing year by year.For coal resources,coal quality is one of the important indicators to evaluate its quality,but coal quality analysis is a very complex work.It is of great demand and value to develop an on-line coal quality detection technology which can quickly detect various industrial and elemental analysis indicators of coal,with high accuracy and accuracy,relatively low operation and maintenance costs and no nuclear radiation hazards.Non-metallic elements in coal not only determine the calorific value of coal,but also closely related to other industrial indicators.Rapid and accurate measurement of non-metallic elements in coal is of great significance for efficient utilization and clean operation of coal.With the rapid development of laser technology,Laser-induced breakdown spectroscopy(LIBS)has also made unprecedented progress in the field of rapid analysis of material elements,bringing new technical means for rapid analysis of elements in coal.Firstly,this paper introduces the principle,method and main problems of LIBS technology.For the problem of high detection limit and susceptibility to continuous background noise in LIBS technology for coal non-metallic element analysis,a baseline correction algorithm based on mathematical morphology and iterative weighted least squares method in image processing field is proposed.First,morphological algorithm is used to obtain an initial rough baseline;on this basis,asymmetric weighted least squares method is used.By adjusting the baseline,a smooth and reasonable baseline signal can be obtained.By subtracting the original spectral signal from the baseline signal,the corrected spectral signal can be obtained.The accuracy and robustness of the baseline are validated by simulated spectral data.The results show that the baseline correction algorithm can effectively identify baseline drift and has good anti-jamming performance.In order to overcome the influence of matrix effect and realize the rapid simultaneous quantitative determination of carbon,hydrogen and sulphur in coal,a 1064nm Nd:YAG solid-state laser was used to ablate nine national standard samples of coal in air.Spectra in the range of 188.885~980.391 nm were collected by Echelle stepped iCCD high resolution spectrometer.Selected 188.885~308.008 nm?655~660 nm?867.5~869nm wavelength range spectra combined with partial least squares regression synchronous detection of carbon,hydrogen and sulfur in coal three non-metallic elements.The results show that the determination coefficients R~2 of the correction models for C,H and S are 0.9997,0.9957 and 0.9969,RMSEC is 0.1369,0.0416 and 0.0534,and the average relative errors ARE are 0.15%,0.90%and 2.58%.In order to test the predictive accuracy of the model,three coal samples were used to verify the reliability of the model.The results showed that the determinant coefficients R~2 of predictive concentration and real concentration of C,H and S elements were 0.9625,0.9928 and 0.9920,the RMSEP of predictive root mean square error was 1.9547,0.1529 and 0.0839,and the average relative error ARE was 2.68%,8.52%and 9.96%.The results show that laser induced breakdown spectroscopy combined with partial least squares regression can be used to quantitatively detect non-metallic elements in coal.