Tobacco leaf distinguishing grouping method based on characteristic chemical component

The invention relates to a tobacco leaf distinguishing grouping method based on characteristic chemical components, and belongs to the technical field of tobacco leaf raw material classification determination. The method comprises: performing correlation replacement on tobacco leaf intrinsic chemical composition indexes, to obtain preliminarily screened chemical indexes, using a hierarchical clustering method to cluster and grouping the preliminarily screened indexes, and using partial least squares discriminant analysis to further screen a chemical index which has an important contribution on each group, and finally using a Fischer stepwise discriminatory method to establish a discrimination model of each group. The tobacco leaf distinguishing grouping method is simple, and is easy to operate. The method provides a scientific and reasonable tobacco leaf classification application method for cigarette blending formulation, suction evaluation pressure on tobacco leaves of a tobacco leaf formulation technician is reduced in great degree. The method is easy to popularize and apply.