利用开路傅里叶变换红外光谱仪进行污染云团遥测时,往往会出现一些未知的干扰物质, 这些干扰的存在使准确地多组分同时定量分析变的很困难。利用标准吸收截面光谱,结合矩阵投 影运算及目标转换因子分析,完成了未知干扰光谱的检出,从而完成多个待测组份的同时定量分析测定。 仿真六氟化硫(SF6 )、甲醇和氨气(NH3 )的混合物光谱,把NH3 作为干扰,预测SF6 和甲醇的浓度程长积 值,预测结果和真实值有相当好的符合度。

It is difficult to accurately realize the multi-component quantitative analysis simultaneously, when applying open path Fourier transform infrared spectroscopy (OP-FTIR) into the remote sensing of cloud cluster because of the existence of some unknown interference materials. Combined the standard spectrum addition with matrix projection operation and target transformation factor analysis, the unknown interference components can be identified and multi-component quantitative analysis was realized simultaneously. The mixed spectrum of SF6 , methanol and NH3 were simulated to predict the concentration of SF6 and methanol while NH3 was regarded as the interference component. The results and the real concentration path-length matched fairly well.