Coal Quality Analyzer Based On Laser Induced Breakdown Spectroscopy

Coal is one of the most abundant fossil fuels on Earth and one of the main energy sources for coal-fired power plants in China.Its nature and quality have a significant impact on the production and operation efficiency,safety,and environmental pollution of coal-fired power plants.Therefore,detecting the composition of coal in different coal forming ages and mining areas is beneficial for the judgment of coal usage methods and the improvement of utilization efficiency.Laser Induced Breakdown Spectroscopy(LIBS)is a material detection technology based on atomic emission spectroscopy.In recent years,because of its advantages such as no sample pretreatment,simple and fast analysis,real-time detection of solid,liquid and gas samples,it has developed rapidly in the field of online analysis,and has good application potential in the field of coal quality detection in coal-fired power plants.The matrix effect and self absorption effect caused by the special physical and chemical properties of coal can affect the accuracy of LIBS coal quality quantitative analysis.Therefore,it is necessary to effectively process the collected spectral information.Firstly,this article builds a LIBS experimental platform for coal quality analysis based on the principle of laser induced breakdown spectroscopy technology,including components such as solid-state lasers,fiber optic spectrometers,and three-dimensional mobile platforms.Experimental parameters such as delay time,laser energy,defocusing distance,and receiving angle were adjusted and optimized.The results indicate that the delay time is 1 μ s.The spectral intensity is most suitable when the laser energy is100 m J,the defocusing distance is about 1mm,and the receiving angle is 45 °,which has the best excitation effect on coal samples.Secondly,in order to overcome the problem of some coal samples being difficult to be made into coal cakes through a tablet press due to their low carbon content,this article designs a multi-point cyclic single excitation experimental scheme using a mechanical disk as the carrier of coal powder.The results show that this scheme can meet the analysis requirements of common analysis methods such as standard curve method,internal standard method,and free calibration method for metal elements,but the fitting effect of the main element carbon element is poor.To improve the carbon element calibration effect,this article proposes the DSC-PLSR(Double Spectral Correction Partial Least Square Regression)calibration algorithm to predict and analyze the element content in coal powder.The former DSC method is divided into two steps.The first step is to perform an adaptive iterative weighted penalty least squares method(P-air PLS)based on P-operation assistance on the LIBS signal,effectively removing background noise and baseline signals during the LIBS measurement process,and mathematically restoring the true spectral information;The second step is the double spectrum correction of plasma temperature compensation and spectral normalization,so that it not only further corrects the signal in the mathematical level,but also compensates the complex matrix effect in the physical level as much as possible.The two-step coordination can improve the Goodness of fit of carbon from nonlinear to above 0.948,which is much higher than the common calibration methods such as internal standard method.The latter is based on the PLSR method for carbon calibration model reconstruction.The experimental results show that the average prediction error of the DSC-PLSR calibration algorithm for carbon content of different unknown samples is only 0.41%,which is greatly improved compared to the traditional PLSR model.Finally,this article develops a coal quality analyzer upper computer measurement and control software based on the Winform framework of C # and the dynamic link library of the spectrometer.The software can effectively identify the spectrometer model,activate channels,synchronize initialization parameters,and achieve user independent calibration.The data layer selects a visual Access database for spectral information exchange.After testing,the software has successfully implemented various algorithms proposed in this article,completed long-term calibration and prediction work,and adapted to the coal quality analysis instrument described in this article.