Study On Analysis Method For Improving The Measurement Accuracy Of Coal Quality By Using LIBS

Based on the importance of coal quality analysis for coal-fired power plants and the limitations of existing coal quality analysis technologies,Laser-Induced Breakdown Spectroscopy(LIBS)technology that can achieve rapid and multi-element simultaneous detection was applied to the analysis of the coal quality.However,due to the matrix effects of LIBS technology,the accuracy of LIBS measurement needs to be further improved.In this work,we carried out the research on methods to improve the accuracy of coal quality analysis based on LIBS,mainly including the following two aspects: one is LIBS technology combined with effective chemometrics methods that used to improve the accuracy of quantitative analysis of coal quality index;the other is the tandem system of LIBS and Laser Ablation Inductively coupled Plasma Time-of-Flight Mass Spectrometry(LA-ICP-TOF-MS),and data fusion technology to improve the accuracy of coal quality discrimination analysis.Firstly,the spectral characteristics of different coal samples are analyzed.Then the obtained spectral data are subjected to the uniform standardization processing.For the quantitative analysis of calorific value and proximate analysis(volatile and ash),through the analysis of the influencing factors of each indicator and the matrix effect of LIBS,different chemometric methods(Multivariate Linear Regression(MLR),Partial Least Squares Regression(PLSR),Support Vector Machine Regression(SVR))are used in the quantitative analysis of each index.The comparison of the results of different model methods showed that the SVR quantitative analysis method based on K-fold Cross Validation(K-CV)parameter optimization can effectively improve the accuracy of quantitative analysis of calorific value,ash content and volatile content.For the quantitative analysis of the main elements(C,H)in coal,a quantitative analysis model based on dominant factors combined with deviation correction is proposed.According to the excitation characteristics analysis of C,it is proposed to establish the MLR model by combining the carbon atoms and carbon molecular spectrum as the dominant factors.For the H element,the atomic hydrogen,carbon and carbon molecular spectrum are chosen as the dominant factors.Furthermore,in order to improve the accuracy of the analysis,PLSR and SVR models are used to correct the deviations generated by the MLR model.The obtained results are compared with MLR,PLSR,and SVR,indicating that the model based on dominant factors combined with deviation correction can effectively improve the accuracy of the quantitative analysis of C and H.Furtherly,based on the complementarity of LIBS and LA-ICP-TOF-MS technology,a tandem system of combining LIBS and LA-ICP-TOF-MS is proposed,and the coal quality is discriminated through data fusion.Through coal classification strategie,the coal quality of LIBS data,TOF data,TOF+LIBS data and TOF/LIBS data obtained by different fusion methods are analyzed.In all cases,the results obtained with the combined TOF and LIBS data were found to be superior to those obtained with the individual TOF or LIBS data.The nonlinear SVM model combined with TOF and LIBS data provided the best coal classification performance.Finally,we summarized the full text,and puts forward suggestions and prospects for the next research work.