A novel coal quality index and its estimation using diffuse reflectance spectroscopy

Abstract

ABSTRACT An essential component of coal’s efficient use and large-scale environmental impact management is evaluating its quality. This study introduces a novel quality indexing technique that incorporates multiple coal properties, providing a comprehensive measure of coal quality. Conventional methods for assessing coal quality, such as proximate and ultimate analyses, are often time-consuming, costly, and require extensive sample preparation, prompting the need for a more efficient approach. In this study, we used the minimum dataset (MDS) approach to estimate a comprehensive coal quality index (CQI) and show that the diffuse reflectance spectroscopy (DRS) may be used as a rapid, nondestructive technique for estimating CQI as a single quality indicator for coal. A total of 212 coal samples of different ranks and grades are collected from the Tertiary and Gondwana coal basins of India. Ten chemical properties of coal were measured following the conventional laboratory-based methods. The spectral responses of the coal samples were measured across the visible-NIR-SWIR (350–2500 nm) range. The MDS approach was adopted while estimating the CQI from the lab-based coal quality parameters. We also examined different chemometric models for estimating CQI values from spectroscopic data. Results showed that the feature selection-based partial-least-square regression (PLSRFS) model may be used to estimate CQI with coefficient of determination (R2) values as high as 0.7 with root-mean-squared error (RMSE) of 0.06. While CQI serves as a single coal quality parameter, the capability to use the DRS approach as a rapid and noninvasive sensing approach for coal quality assessment provides new opportunity to characterize coal quality in an instant and cost-effective manner, especially, when a large number of coal samples have to be tested. The integration of the CQI with DRS could facilitate better decision-making in coal utilization and management, leading to optimized performance and reduced environmental impact.