基于NIRS的煤样定量检测技术研究

Quantitative Detection Research of Coal Samples Based on Near Infrared Spectroscopy

研究了近红外漫反射光谱分析技术运用于煤的灰分和硫分的快速无损检测方法,通过研究煤粉样品的内部成分,利用近红外漫反射光谱结合不同光谱预处理方法建立主成分回归(PCR)定量检测模型。预处理方法包括归一化、一阶微分、二阶微分和多元散射校正。结果发现经归一化处理后的灰分含量模型较优,多元散射校正处理后的硫分含量模型较优,证实了基于主成分回归分析煤粉成分的近红外光谱建模具有较高的应用价值。

Fast and non-destructive of ash and sulfur content of coal samples with diffuse reflectance near-infrared spectroscopy was studied.In order to study on the components of coal samples,this present is concerning quantitative detection of different coal samples combined with different pre-processing methods which was used for Principal component regression（PCR）.And the pre-treatment concerned included standardization,frist derivative,second derivative,multiplicative signal correction.The result indicated that the ash content of the establishment of standardization was the best,at the same time the sulfur content of multiplicative signal correction was also better than others.There was a good repeatability of coal quality analysis by NIRS modeling method based on principal component regression（PCR）.