Advanced Multi-Parameter Prediction of Coal Quality Using LIBS and Ensemble Machine Learning Techniques

Accurate assessment of coal quality is essential for optimizing combustion efficiency and reducing pollutant emissions in coal-fired power plants. In this study, we developed a laser-induced breakdown spectroscopy (LIBS)-based framework, combined with advanced machine learning techniques to predict key coal quality parameters, including elemental carbon, ash content, volatile matter, total sulfur, and calorific value. After applying spectral preprocessing methods. such as outlier removal and baseline correction, predictive models were established using algorithms including the least squares support vector machine (LS-SVM), which achieved the highest accuracy with an R² of 0.9940 for elemental carbon. The results indicate that the proposed method provides a reliable and efficient alternative for rapid coal quality analysis, with potential for future application in intelligent monitoring and control frameworks.