论文部分内容阅读
基线的变化幅度用Monte Carlo方法产生并用一个计算程序模拟产生不同比例双组分混合物的红外光谱。因子分析结果表明,RE(真实误差)函数比IND(Indicator)函数略佳。但若基线吸光度偏差幅度大于0.007,RE和IND检出的P(真实的组分数目)大部分是错误的。本文还讨论了基线变化幅度与RE和IND函数检出因子正确率的关系。
Baseline amplitude variations were generated using the Monte Carlo method and a computational program was used to simulate the infrared spectra resulting in different ratios of two-component mixtures. The factor analysis shows that the RE (real error) function is slightly better than the IND (Indicator) function. However, if the deviation of the baseline absorbance is greater than 0.007, most of the P (true number of components) detected by RE and IND is incorrect. The paper also discusses the relationship between the baseline change amplitude and the correctness of the detection factors of RE and IND functions.