Abstract:In order to meet the demand of coal quality analysis in power industry, this paper proposes a new technology of coal analysis based on Hadamard near infrared spectroscopy. On the basis of comprehensive consideration of five aspects, such as development of near infrared spectrometer, the design of control and analysis software, the acquisition of coal spectral signal, the extraction of characteristic information and the establishment of quantitative model, the Hadamard near infrared coal quality analysis system is developed. In the study, 41 standard coal samples with different quality indicators are quantitatively analyzed and predicted. The ability of Hadamard near infrared spectroscopy to predict coal indicators in conditions of same particle size is investigated. Based on LS-SVM algorithm, a coal indicators prediction method is proposed, which has a good correlation with the coal index, and the correlation coefficients is above 0.9.