近红外光谱法快速测定混合含能材料中邻苯二甲酸二丁酯

Fast analysis of the content of DBP in composite energetic material by near infrared spectroscopy

利用近红外光谱快速分析技术对含能混合材料中邻苯二甲酸二丁酯(DBP)组分质量分数进行定量检测研究。从生产线采集与自行制备混合材料样本共175批次,采集这些样本近红外谱图。通过主成分分析(PCA)等方法剔除异常光谱,通过光谱分析等方法选取波数范围参与偏最小二乘(PLS)法建模。通过模型均方根预测误差(RMSEP)以及RMSEP/维数曲线,结合光谱PCA得分因子随浓度分布判断所选建模参数准确性、优良性。最终确定建模波段7467.2~6514.5cm^-1,6225.2~5411.4cm^-1,预处理方法为多元散射校正(MSC)。分别采用预测标准差(SEP)、Bias值、平均绝对偏差及重复测量标准偏差评估模型准确性与精确性。结果表明,所建新方法有较好准确性与精确性,新方法代替常规气相色谱法检测重复性更好,效率更高,更环保。

Near infrared spectroscopy was used to quantitatively detect the content of dibutylphthalate( DBP) in a composite energetic material. 175 samples were acquired via collection from production line and preparation. Near infrared spectra of these samples were collected. Abnormal spectra were picked out by using principal component analysis( PCA). Wave bands were chosen by spectral analysis,etc. And these wave bands chosen were used to build the partial least square models( PLS).The accuracy of these models was determined by the root-mean-square prediction error( RMSEP) and the curve of RMSEP/dimension. And the distribution of score factor with sample concentration was also used to determine the accuracy of these models. The final wave bands determined was the bands of7467. 2-6514. 5 cm-1 and 6225. 2-5411. 4 cm-1. The preprocessing method determined was the multiplicative scatter correction( MSC). The predictive standard deviation( SEP),value of Bias,and average absolute deviation were used to evaluate the accuracy of the final model determined. And the standard deviation of repeated measures was used to evaluate the precision of the final model. The results showed that the new method built was accurate and precise. Compared with the conventional method of gas chromatography,the new method was precise,fast and environmentally friendly.