High Stability Analysis Of Coal Quality By NIRS-XRF Bispectral Combination

Timely knowledge of the calorific value of coal and its related industrial indicators such as ash content,volatile matter,and total sulfur is crucial for coal purchasing based on quality,pricing according to quality,and enhancing coal utilization efficiency.Traditional national standard coal quality analysis methods lag significantly behind industrial production needs due to their poor timeliness.Therefore,there is an urgent need to develop rapid and highly stable coal quality testing methods.One promising approach is the use of Near-infrared spectroscopy(NIRS),which offers advantages such as fast analysis speed,high efficiency,great stability,and low cost.NIRS can detect primary harmonic absorption spectra of hydrogen-containing groups like C-H,O-H,and N-H in coal,as well as multiple harmonic absorption spectra of C-O,C-N,and C-C.However,it is unable to measure the ash content of coal.Another technique,X-ray fluorescence spectrometry(XRF),provides stable analysis of inorganic ash-forming elements in coal such as Si,Al,Ca,Fe,and Mg.However,it cannot analyze light elements like C,H,and N.To address these limitations,this study proposes an innovative approach that combines the strengths of NIRS and XRF in a dual-spectrum analysis method.By integrating the stable detection of organic groups in coal offered by NIRS with the stable detection of ash-forming elements in coal provided by XRF,this approach enables comprehensive analysis of all coal quality indicators.Additionally,it ensures highly repeatable measurements.The modeling method employed in this study involves preprocessing and fusing the dual-spectrum data,which is then used as input variables for Partial Least Squares Regression(PLSR).This allows for the modeling and prediction analysis of coal heating value,ash content,volatile matter,total sulfur,and other indicators.To verify the reliability of this combined method and the prediction model,multiple sets of coal samples were tested in experiments.The results demonstrate that this method meets the national standard requirements for the measurement repeatability of all industrial indicators in coal.This paper initially elucidates the significant position of coal resources in China’s energy structure and the necessity of fully and rationally processing and utilizing coal resources.It provides a comprehensive overview of the applications of various analysis methods in coal quality analysis and provides a detailed description of the measurement principles and research status of NIRS and XRF.Subsequently,the NIRS-XRF coal quality analysis experiment is introduced,including sample selection,experimental setup,parameter settings,spectral calibration methods,etc.The advanced nature of the NIRS-XRF dual-spectrum combined coal quality analysis method is discussed.Furthermore,the paper presents the optical preprocessing and modeling analysis methods for NIRS-XRF dual-spectrum combination.This includes the preprocessing methods for NIRS spectra and XRF energy spectra,such as the selection of effective wavelengths and spectral fusion.Calibration modeling is established using the PLSR method.Additionally,evaluation of the predictive performance of the developed models is conducted using indicators such as determination coefficient(R~2),root mean square error of validation(RMSEV),average relative deviation(ARE),and standard deviation(SD).A comparison and discussion of the prediction results for coal heating value indicators using the NIRS-XRF combined method and single NIRS are also provided.Finally,the paper introduces a coal quality analyzer developed based on the NIRS-XRF dual-spectrum combination method.This includes the functional structure,software interface,operating procedures,etc.Furthermore,a corresponding in-plant coal control system scheme is designed based on this coal quality analyzer.Utilizing this control system,multiple sampling points on coal trucks can be sampled,prepared,measured,and stored,enabling comprehensive monitoring and control of the quality of incoming coal.This system can be widely applied in"high-carbon"industries such as thermal power generation,coal chemical industry,metallurgy,cement,and coking,thereby assisting in achieving China’s carbon neutrality goals.The innovation of this article lies in:1.Technological innovation:a NIRS-XRF dual-spectral high-stability analysis technology for coal quality was developed,and the organic groups and inorganic ash-forming elements in coal were measured with high stability by NIRS and XRF,respectively,and the repeatability of industrial indicators such as coal calorific value,ash,volatile content,and sulfur reached the requirements of national standards.2.Method innovation:A set of pretreatment and coal quality prediction modeling methods suitable for NIRS-XRF dual spectrum are proposed.By preprocessing the two sets of spectra by standard normal transformation and SG smoothing,they were fused as input variables of PLSR for preliminary modeling of the full spectrum,and then the effective band of NIRS was selected according to the regression coefficient,and after fusion and normalization with the XRF spectral line of gray-forming elements,PLSR was reused for calibration modeling,which greatly improved the reliability and accuracy of the analysis results.