论文部分内容阅读
基于近红外光谱技术结合应用主成分回归法(PCR)、偏最小二乘法(PLS)、BP神经网络法、支持向量机(SVM)等四种方法对来自不同国家和地区的24种聚丙烯-聚丙烯酰胺型保水剂进行了品种的鉴别研究。结果表明:近红外光谱技术结合SVM法可以有效的进行保水剂分类鉴别工作。当光谱范围选择5000cm-1～9000cm-1,经验参数c=8,g=0.0313时,PCA-SVM模型的预测准确率可以达到100%。研究证明此种方法可以应用于保水剂品种的鉴别。
Based on near infrared spectroscopy (FTIR) and principal component analysis (PCR), partial least squares (PLS), BP neural network and support vector machine (SVM), four kinds of polypropylene- Polyacrylamide type water retaining agent for the identification of species. The results showed that near infrared spectroscopy combined with SVM method can effectively classify and identify water-retaining agents. When the spectral range is selected from 5000cm-1 to 9000cm-1, the empirical parameter c = 8, g = 0.0313, the prediction accuracy of PCA-SVM model can reach 100%. Research shows that this method can be applied to the identification of SAPs.