According to the needs of classification and identification of many varieties of soybean in Northeast China, this paper uses the research method of combining theoretical calculation and experimental analysis to carry out the classification and identification of 6 varieties of soybean. Oleic acid and linoleic acid are important components of soybean. Firstly, based on density functional theory, the molecular spatial structures of oleic acid and linoleic acid were constructed, and the theoretical Raman spectra were optimized and calculated by B3LYP/6-31+G(d,p) basis set. Then, the Raman spectra of oleic acid, linoleic acid analytical purity and six varieties of soybean were obtained by experiment, and the theoretical Raman spectra were compared with the experimental Raman spectra. It was found that all varieties of soybean had strong Raman peaks at 1281, 1445, 1662 and 2904 cm-1. Finally, taking the four Raman peaks as the characteristic peaks, the principal component analysis (PCA) and linear discriminant analysis (LDA) were used to visually classify different varieties of soybeans, and the classification accuracy reached 90%. The results show that density functional theory combined with Raman spectroscopy can effectively classify soybean varieties, which provides a certain reference for the development of intelligent agriculture.