Determination of Calorific Value of Mixed Coals by Analysis of Major Elements Using Data Pre-Processing in Laser-Induced Breakdown Spectroscopy

The purpose of this study was to enhance the accuracy calorific value estimation coal by applying data preprocessing methods in laser-induced breakdown spectroscopy (LIBS). Savitzky–Golay (SG)-smoothing and SG derivative were adopted improve prediction model. relationship among original, SG-smoothing-pretreated, derivative-pretreated LIBS their elemental concentrations determined using partial least squares regression (PLSR) In order compare reliability each PLSR model, coefficient determination, root mean square error (RMSE), relative error, RMSE average used. As a result, model processed with method highest, lowest three models. predictability concentration element pre-processed confirmed residual predictive deviation parameter. predicted estimated from elements Dulong’s equation. pretreated showed compared mixed coals obtained via chemical analysis.