Abstract:[ABSTRACT] OBJECTIVE: To optimize extraction process of Supplemented Guanjie Zhitong decoction. METHODS: On the basis of single factor investigation, the water addition, extraction time and extraction times were taken as the investigation factors, and the contents of genipine, pinoresinol diglucoside, ferulic acid, cup amaranthone and extract ratio were used as evaluation indexes, and the AHP-entropy weight method was used to determine the weight coefficients of each index. Optimize the extraction process parameters of Guanjie Zhitong decoction through orthogonal experiments and genetic algorithm (GA) backpropagation neural network (BP neural network) method, and verify and compare the process parameters obtained from the two methods. Results: The optimal process parameters obtained by orthogonal experiments were as follows: water was added 6 times, extracted 3 times, each time 0.5 h, and the comprehensive score was 90.21 points (RSD was 1.38%); The optimal process parameters obtained by GA-BP neural network optimization were as follows: water addition 6 times, extraction time 1.5 h, extraction times 4 times, and the comprehensive score was 99.26 points (RSD was 0.09%), slightly higher than the orthogonal experiment, combined with the actual production demand, it was finally determined that the optimal extraction process of joint analgesic decoction was to add 6 times of water and extract 3 times for 0.5 h each time. Conclusion: The extraction process parameters optimized by orthogonal experiments combined with GA-BP neural network are stable and reliable, which can provide reference for subsequent research and development. |