Network analysis of ethanol precipitation process for Schisandrae Chinensis Fructus

• DOI: 10.4268/cjcmm20141715
• Publication Date: 2014-12-02T08:22:03Z
• Authors: Yi Zhong; Jieqiang Zhu; Xiaohui Fan; Li‐Yuan Kang; Zheng Li
• Keywords: Ethanol precipitation; Ethanol content

A set of central composite design experiments were designed by using four factors which ethanol amount, concentration, refrigeration temperature and time. The relation between these with the target variables retention rate schizandrol A, soluble solids content, removal fructose glucose analyzed Bayesian networks, amount concentration found as critical process parameters. Then a network model was built 2 inputs 4 outputs back propagation artificial neural networks optimized genetic algorithms. R2 MSE from training 0.983 8 0.001 1. test 0.975 9 8. results showed that analysis method could be used for modeling Schisandrae Chinensis Fructus precipitation identify operating