论文部分内容阅读
应用近红外光谱结合不同化学计量学方法对同仁乌鸡白凤丸原料、辅料、半成品、制剂进行定性和定量分析。建立原料药材的近红外标准光谱库,利用质量控制(QC)比较检索方法鉴别未知样品。采用主成分分析-马式距离法鉴别乌鸡。建立半成品标准光谱库,用相似度匹配算法筛查阴性(缺味)半成品。采用偏最小二乘法对主要原辅料、半成品和制剂进行了定量分析。采用近红外混合过程分析仪结合移动块标准偏差法在线判断同仁乌鸡白凤丸半成品混合终点。所建立的定性分析模型能够对样品做出准确的分类;定量分析方法结果准确可靠。方法准确、可靠、快速、简便、环保,可通过在线或旁线方式用于工业现场的原位检测。
Near-infrared spectroscopy and different chemometric methods were used to qualitatively and quantitatively analyze the raw materials, excipients, semi-finished products and preparations of Tongji Wujibaifeng pills. The establishment of near infrared spectroscopy library of raw herbs, the use of quality control (QC) comparative search method to identify unknown samples. Identification of black-bone chicken by using principal component analysis-horse distance method. The establishment of semi-finished standard spectral library, using similarity matching algorithm to screen negative (lack of flavor) semi-finished products. Partial least squares method was used to quantitatively analyze the main raw materials, semi-finished products and preparations. Using near infrared mixing process analyzer combined with moving block standard deviation method to judge Tongyao Wuji Baifeng Wan semi-finished product mixing end point on line. The established qualitative analysis model can accurately classify the samples. The quantitative analysis method is accurate and reliable. The method is accurate, reliable, fast, easy and environment-friendly and can be used for in-situ inspection of industrial sites through on-line or bypassing methods.