| The liquid medicine such as Chinese medicine injection is a complicated mixed solution which is closely related to people’s life.With the increasing emphasis on health,improving the quality of medicine such as medicine injection becomes the important problem in the research field of Chinese medicine production.In this article,Ophiopogon japonicas is the study object,and the precipitation process of ophiopogon polysaccharide is studied through experiments;the neural network model based on BP algorithm of the precipitation process of ophiopogon polysaccharide is established according to the measured data;according to the simulation results of the model,the model is optimized.Firstly,experimental research on the polysaccharide contained in the Ophiopogon japonicas liquid was carried out.Experiments are divided into two parts,one part is that using the alkaline hydrolysis and alcohol deposition method to make polysaccharide separate from the Ophiopogon japonicas liquid.The other part is that detecting the quantity of polysaccharide by phenol sulfuric acid method.These experiments collect 110 groups of data,and these data provide foundation for the process modeling.Secondly,the basic theory of BP neural network and the feature of selecting model parameters are introduced,and two kinds of improved BP algorithm are introduced according to the shortcomings of BP neural network.After analyzing the precipitation process of Ophiopogon japonicas polysaccharide,the parameters of the polysaccharide precipitation process model were determined.And then the neural network model is established.At the last,the simulation results are analyzed.Finally,in order to improve the accuracy of the model,the BP neural network model based on genetic algorithm and BP neural network model based on group genetic algorithm are built.Last,by comparing the simulation results,the accuracy of the BP neural network model which based on group genetic algorithm is obviously improved. |
PDF Full Text Request