文献类型：会议论文

中文题名：Rapid Coal Quality Detection Based on Near-Infrared Diffuse Reflectance Spectroscopy

作者：Yaning Wang[1];Yuqiang Li[1];Jia Li[1];Zixiang Wang[1];Hanshun Gu[1];Feidi Chen[1];Yunmeng Zhao[1,2]

机构：[1]East China University of Science and Technology, School of Information Science and Engineering;[2]Qingyuan Innovation Laboratory

会议论文集：第43届中国控制会议论文集（12）

会议日期：20240728

会议地点：中国云南昆明

语种：英文

中文关键词：Near-infrared spectroscopy;coal;preprocessing methods;PLS

摘要：This study investigated the application of near-infrared（NIR） diffuse reflectance spectroscopy for rapid and non-destructive detection of ash and sulfur content in coal powder. Partial least squares（PLS） regression models were established using NIR diffuse reflectance spectroscopy. Preprocessing methods included standard normal variate, first derivative,second derivative, and multiplicative scatter correction were discussed. The ash and sulfur content models processed with standard normal variate transformation outperformed. The experimental results demonstrate that the established models for ash and sulfur content have achieved satisfactory performance, with coefficients of determination（R） of 0.95 and 0.96, respectively,and root mean square error of prediction（RMSEP） of 0.68 and 0.18, respectively. The confirmation of NIR spectroscopy modeling for rapid detection of coal powder properties compared to traditional methods indicates faster speed and satisfactory precision, demonstrating its significant applicability in industrial process analysis. It has been confirmed that NIR spectroscopy modeling enables rapid detection of coal powder properties. Compared to traditional methods, it offers faster speed while meeting the required precision standards, demonstrating higher application value.