论文部分内容阅读
采用高效液相色谱获得黄连上清片的色谱指纹图谱,利用基于主成分分析的投影判别法分析实验结果,并用反传人工神经网络对未知样品进行预报。结果表明,不同厂家生产的黄连上清片存在显著差异,主成分分析投影判别法能对样品进行正确分类,从而建立了识别不同厂家黄连上清片的方法,能有效地控制中药黄连上清片的质量。此外,主成分分析还用于优化反传人工神经网络,统计多次预报的结果,表明经过优化的反传人工神经网络能对未知样品的来源进行准确预报。
Chromatographic fingerprints of Huanglian Shangqing Tablets were obtained by high performance liquid chromatography. Projective discriminant analysis based on principal component analysis was used to analyze the experimental results, and unknown samples were predicted using a back propagation artificial neural network. The results show that there are significant differences between the Huanglian Shangqing tablets produced by different manufacturers, and the principal component analysis projection method can correctly classify the samples, thus establishing a method for identifying different manufacturers of Huanglian Shangqing tablets, which can effectively control the traditional Chinese medicine Huanglian Shangqing tablets. the quality of. In addition, principal component analysis is also used to optimize the back-propagation artificial neural network and statistics the results of multiple predictions, indicating that the optimized back propagation artificial neural network can accurately predict the source of unknown samples.