掺假者
弗洛斯
化学计量学
偏最小二乘回归
主成分分析
色谱法
模式识别(心理学)
数学
人工智能
化学
计算机科学
统计
生物化学
芦丁
抗氧化剂
作者
Jing Zhao,Pengdi Cui,Huan Liu,Chunhua Wang,Ming Liu,Gang Li,Mingzhong Pan,Zheng Li,Tongchuan Suo
标识
DOI:10.1016/j.infrared.2019.103139
摘要
Adulteration of Lonicerae Flos (LF) in Lonicerae Japonicae Flos (LJF) is difficult to detect with naked eyes, and significantly affect the clinical efficacy. The rapid screening and quantitative analysis of adulterant LF in LJF is important in the raw material preparation. In this study, Fourier-transform near infrared spectroscopy (FT-NIRs) and chemometrics are used for rapid analysis of adulterant LF in LJF. The powder of the pure LJF, LF and adulteration, and the corresponding extract solutions are prepared. FT-NIR spectra are modeled for comparison. Successive Projections Algorithm (SPA) is used for characteristic bands selection; Savitzky-Golay smoothing (S-G), multiplicative scatter correction (MSC), first deviation (1stD) and standard normal variate (SNV) are adopted for spectra preprocessing; Principal component analysis (PCA), and back propagation artificial neural network (BPANN) are employed as qualitative analysis method for screening whether or not containing adulteration; BPANN and partial least squares (PLS) are used for quantitative analysis. The optimal results of screening and quantitative models are obtained from “S − G + MSC + SNV + BPANN” and “S − G + SNV + BPANN” of powder samples. It indicates that FT-NIRs can provide a rapid and efficient tool for screening and quantitative analysis of adulterant LF in LJF.
科研通智能强力驱动
Strongly Powered by AbleSci AI