Study On Coal Ash Content Detection Method Based On Spectrum Analysis Technology

Coal ash is one of the important indexes for evaluating coal quality,In the increasingly competitive coal sales market,The ash level plays a key role in setting the price of commercial coal in Coal Preparation Plant.Therefore,how to quickly detect the ash content of coal has become a very important problem in the process of production management.The traditional coal detection method can not get the coal ash in real time,At present,the real time ash detection methods are mostly based on radiation method,which is harmful to the environment and people.In this paper,the light source with less energy is used as energy source,Based on the optical properties of coal surface,the relationship between coal ash and reflectance spectra of coal surface is studied,And try to establish a green and efficient ash detection model.In this study,the relationship between the ash content of coal sample and the spectral reflectance of the sample surface was studied by using the micro fiber spectrometer to collect the reflectance spectrum of the coal surface,This kind of method is a new method of coal ash detection.Therefore,the research on the parameters and performance of the micro fiber spectrometer is discussed in this paper.The spectrometer parameters include the average number of times,the integration time,the width of the smoothing window.The single factor experiment was designed and the results were analyzed,Finally,the optimum parameters of the micro spectrometer were determined and used as the following conditions for the acquisition of coal samples.At the same time,the baseline stability and spectral repeatability were tested,Based on the analysis of test results,We can fully understand the performance of the micro fiber spectrometer before the sample collection.In order to investigate the influence of coal physical properties on the surface reflectance spectra of coal samples,A single factor experiment was designed to investigate the effect of coal moisture content,coal particle size and coal seam density on the surface reflectance of coal samples.In the process of the influence of coal water content on the spectral reflectance of coal samples,Different test samples were prepared by adding different volumes of clear water in dry coal samples,the surface reflectance spectra of each sample were collected,Then,the relationship between the water content of coal and the average reflectance of coal samples was established and analyzed.In the experiment of the influence of the coal sample compactness on the reflection spectrum of the coal sample,the compactness of the coal sample is measured by the compression height of the coal sample.and the average surface reflectance of coal samples with different compression height were collected,Finally,the influence of coal seam density on the reflection spectrum of coal samples is obtained.In the study of the influence of the sample size on the spectral reflectance of the sample,the sample with different particle size was used as the test sample,The average spectral reflectance of different coal samples was collected and analyzed,Finally,the influence of coal particle size on the surface reflectance of coal samples is obtained.The main component of gangue is inorganic mineral,the ash content of coal is mainly determined by the content of inorganic minerals in coal,Firstly,the influence of gangue content on the spectral reflectance of the samples was investigated,the coal ash content was used to characterize the gangue content,the results show that there is a great correlation between gangue content and surface reflectance of coal samples.This conclusion lays the foundation for the establishment of ash detection model.There may be differences in different types of inorganic minerals,In order to investigate the effect of different inorganic minerals on the spectral reflectance of samples,aluminosilicate mineral montmorillonite,kaolinite,quartz and limonite silicate minerals,four kinds of inorganic minerals was selectedas the research object.Single factor experiments were designed to investigate the effect of single mineral content on the surface reflectance of the samples.The design of orthogonal test to explore the influence of kaolinite,quartz stone,three kinds of inorganic mineral reflectivity on the sample surface.Finally,using clustering algorithm combined with the coal gangue surface reflection spectra of sample data to establish coal gangue identification model.Coal gangue identification model has higher prediction accuracy for unknown samples.In this study,90 samples of different coal ash samples were collected as the sample,Using partial least squares method to establish ash detection model,Using the model to test 10 unknown samples.RMSEC,RMSEP and R were chosen as the indexes to evaluate the stability and accuracy of the model,The establishment of ash detection model based on PLS which RMSEC=2.8025,RMSEP=2.45,R=0.9949.In addition,different spectral preprocessing methods were used to deal with the original spectral data,the experimental results show that the best results are obtained by using the two order derivative to deal with the original spectral data,After the two derivative spectral data processing,the ash detection model RMSEC=2.2790,RMSEP=2.3925,R=0.9956.Finally,the ash detection model is established by the combination of principal component analysis and neural network algorithm,The results of the two methods are compared,The ash detection model based on PCA and BP neural network was better than the ash detection model based on PLS,The model based on PCA and BP neural network had RMSEC=1.6084,RMSEP=1.2422,R=0.9986.