摘要
为了研究煤质的硫含量,采集了120个煤粉样品的近红外漫反射光谱,建立了偏最小二乘回归结合不同光谱预处理方法的定量数学模型,并与工业检测结果进行对比。结果表明:采用5点平滑处理后的模型效果最佳,相关系数达到0.89695,校正集均方根误差(RMSEC)和预测集均方根误差(RMSEP)分别为0.0406和0.0423,结果表明模型具有较高的相关性、稳定性和预测能力。
In order to investigate the sulfur in coal,the authors collected the near-infrared diffuse reflection spectrum from 120 coal sam-ples,the mathematical model which introduced partial least squares regression (PLSR) combined with different spectrum pretreatment methods were established.Then the paper compared the model with industrial detection.The results showed that,the PLSR modeling of sul-fur by 5 points smooth had better effects.The root mean square error of calibration (RMSEC) and root mean square error of prediction (RMSEP) were 0.0406 and 0.0423,with correlation coefficients of 0.89695 respectively.The results indicated that the model had high rel-evance,stability and accuracy.
出处
《洁净煤技术》
CAS
2014年第6期74-75,79,共3页
Clean Coal Technology
关键词
近红外光谱
偏最小二乘回归
定量
全硫
near-infrared spectroscopy
PLSR
quantitative
total sulfur