论文部分内容阅读
为实现对新陈莲子的快速鉴别,该文采用自行研制的表面解吸常压化学电离质谱(DAPCI-MS),在无需样品预处理的前提下,直接对新鲜和陈年莲子切面进行质谱检测,获得其化学指纹图谱,并通过主成分分析(PCA)和反向传输人工神经网络技术(BP-ANN)对所获指纹谱图信息进行分析,获得新鲜和陈年莲子的质谱信息特征。结果表明,在负离子模式下,DAPCI-MS结合化学计量学方法,实现了新鲜和陈年莲子的快速鉴别,其测试样本准确率分别为95.0%和91.7%;对不同年份莲子也能够有效地分类判别,2012、2011、2010和2009年莲子测试样本准确率分别为90%,85%,85%和90%。该方法具有分析速度快,信息提取准确,识别精度高等优点,为其他粮食谷物品质的鉴定提供参考。
In order to realize the rapid identification of the new lotus seeds, the surface desorption atmospheric pressure chemical ionization mass spectrometry (DAPCI-MS) developed by ourselves was used to detect the mass spectrum of fresh and aged lotus seeds without sample pretreatment. Their chemical fingerprints were obtained and their fingerprints were analyzed by principal component analysis (PCA) and reverse transmission artificial neural network (BP-ANN) to obtain the mass spectral features of fresh and aged lotus seeds. The results showed that the rapid identification of fresh and aged lotus seeds by DAPCI-MS combined with chemometric method in negative ion mode, the accuracy of the test samples were 95.0% and 91.7% respectively; lotus seeds can also be effectively classified in different years Discriminant, the accuracy of lotus test samples in 2012, 2011, 2010 and 2009 were 90%, 85%, 85% and 90% respectively. The method has the advantages of fast analysis, accurate information extraction and high recognition accuracy, which can provide references for the identification of other grain cereals.