木糖醇发酵液脱色的优化工艺

Technology for Optimizing Decoloration of Xylitol Fermentation Liquor

根据木糖醇发酵脱色实验 ,借助均匀设计法 ,确定人工神经网络的隐层神经元数、学习速度和动量因子等模型参数 .构建一个能用于预测、优化木糖醇发酵液脱色过程的 3层数学模型 ( 5- 8-1) ,并通过遗传算法寻优 ,获得效果更好的脱色工艺 .

Artificial neural network and genetic algorithm are applied to the technology for optimizing decoloration of xylitol fermentation liquor. Based on experiment of xylitol fermentation and decoloration, a 5 8 1 three level mathematical model is formed for predicting and optimizing decoloration of xylitol fermentation. It is formed with the help of uniform design method by which such model parameters as number of neurons in the implicit level of artificial neural network, speed of learning, and momentum factors can be determined. The technology with better decoloration effect can be obtained by genetic algorithm which seeks optimization.