Determination of the calorific value and ash content of coal using the analysis of primary X-ray radiation scattering spectra

Abstract

The study presents a new approach to the determination of the main fuel characteristics of coals, such as calorific value and ash content, using wavelength dispersive X-ray fluorescence analysis (WDXRF) in combination with partial least squares (PLS) regression. The Compton and Rayleigh scattering spectra of the primary radiation of the X-ray tube served as the basis for the quantitative assessment of these indicators.

The effect that spectral resolution has on the quality of the prediction was also assessed in the study. It was found that increasing the spectral resolution increases the accuracy of predicting the calorific value and ash content of coals. The use of a LiF (220) crystal analyzer in combination with a high-resolution collimator provided a root-mean-square error of 0.52 MJ/kg for an average calorific value of 29.06 MJ/kg, and 1.70 wt% for an average ash content of 9.20 wt% in validating the training set data. Although the current results of coal quality analysis obtained on the basis of primary X-ray scattering spectra cannot replace the already accepted methods, the presented approach can provide useful information for rapid assessment of coal quality and can be used in automated systems for quality control of incoming fuel.