Fourier transform near-infrared spectroscopy(FT-NIR) can reflect the overall molecular composition of microbial cells to identify different types of microorganisms. To establish an accurate, effective method about the differentiation and identification of Alicyclobacillus strains between different species, the present research performed the following studies by FT-NIR:（1）The FT-NIR spectra about seven type stains was clustered for data analysis. After preprocessing, reduction of data was performed by Principal Component Analysis (PCA) and Linear Discriminant Analysis(LDA), exploring the feasibility of differentiation and identification between different species, the result suggested that the PCA model can cluster the seven species of Alicyclobacillus strains correctly and the LDA model Ⅰcan predict the unknown species with 100% accuracy. It evidenced that the method could identify different species of Alicyclobacillus strains preliminary. （2）In order to improve the robustness and practicability of the model, a total of 41 Alicyclobacillus strains including type and isolated strains were prepared for LDA modelⅡ, using the same methods as mentioned before. The result indicated that the LDA model validated by fifteen sample with 86.67% accuracy. It was more perfect and more comprehensive.As a result, the FT-NIR technology combined with chemometrics method can accurately and effectively identify Alicyclobacillus strains between different microbial species.