According to the traditional morphological classification, the Rhizoma Chuanxiong quality of traditional chinese medicine is divided into recommended products and non-recommended products; the chromatogram data of Rhizoma Chuanxiong are obtained in the standard condition, we inspect and abstract main data, get the main peaks of linear independent vectors. Get every clustering the most likely data of Rhizoma Chuanxiong with Particle Swarm Optimization of synchronous learning factor and Rosenbrock method of Optimization. Make Rhizoma Chuanxiong criterion data as grey-whitened Gumbel distribution and calculate similarity with the sample data of Gumbel distribution and ensure the finial similarity by weighted average. The program will use Simpson double integra and B-spline interpolation. It need to calculate the similarity between unknown sample and the common pattern of corresponding standard sample when analyze unknown sample. This research combines the plant morphology, chemistry, statistics and calculating technology establishing the pattern identification methods of quality, have good attempt to traditional chinese medicine quality assessment methods and make a basic to the deeper research about Rhizoma Chuanxiong in theory and practice.

According to the traditional morphological classification divide the quality of traditional Chinese medicine Acanthopanax Senticosus Harms into recommended products non-recommended products and shoddy goods distribute the chromatogram data Acanthopanax Senticosus Harms obtained in the condition of standard inspect and abstract main data and get the great peaks of linear independent vectors. Get Acanthopanax Senticosus Harms every clustering centre data with liner reduction PSO and Conjugate gradient optimization. Make Acanthopanax Senticosus Harms criterion data as Gamma distribution and calculate similar by the sample data Normal distribution and ensure the finial similar by angle cosine. That will use lada integral and upper gauss interpolation. It need to calculate the similar between unknown sample and the common pattern of corresponding sample when analyze unknown sample. Make appraise classify and evaluate to unknown sample and make good identification to counterfeit medofenoxate for example gracilistylus koxmo hoo column5 plus fine.

According to the traditional morphological classification, the North Schisandra quality of traditional chinese medicine is divided into recommended products and non-recommended products; Discrete the chromatography data of the North Schisandra which obtained under the condition of standard test and also make the knowledge reduction. Obtaining the great peaks of linear independent vectors and obtaining every clustering centre data of North Schisandra by the fuzzy and Defuzzification methods. Make North Schisandra criterion and sample data as Normal-Weibull distribution to calculate similar. This research combines the plant morphology, chemistry, statistics and calculating technology to establish the pattern identification methods of quality and have good attempt to Chinese traditional machine quality methods. The research results provide the basis for medicine industry and manufacturing.