Abstract: ： 81 Pu'er tea samples from three different producing areas, such as Xishuangbanna, Pu'er City and Linyi City, were studied. Among them, 54 were corrected and 27 were verified. A single Pu'er tea sample was collected by near-infrared spectroscopy at 1100- The spectral data in the range of 2498 nm was established by principal component cluster analysis and discriminant analysis to establish a qualitative analysis model for Pu'er tea production. The test results show that the correct recognition rate of the correction set of the two analytical models is greater than 90%, which can be used for the identification of Pu'er tea origin. Among them, the discriminant analysis model based on the discriminant analysis method is better, and the correctness rate of the calibration set and the verification set is 98.15% and 100%, which is more suitable for the identification of Pu'er tea origin. Therefore, the application of near-infrared spectroscopy technology can quickly and non-destructively identify the origin of Pu'er tea, providing a new method for the detection of Pu'er tea production.

Key words: near-infrared spectroscopy, Pu'er tea, cluster analysis, discriminant analysis, origin identification