摘要
为实现新霉素发酵培养基的优化,用均匀设计安排实验,将所得实验数据作为训练样本建立一个结构为6-5-1的神经网络并进行验证。在此基础上进一步利用神经网络和遗传算法偶合寻优得到了新霉素发酵培养基的最优配比,在此条件下的新霉素摇瓶发酵效价为19210u/mL,比优化前提高了23%。
Uniform design was taken to arrange experiments,in order to optimize the fermentation medium for neomycin.The experimental data were used as training samples to establish a 6-5-1 neural network and verify it.On this basis,the optimal ratio of neomycin fermentation medium was obtained by coupling optimization of neural network and genetic algorithm.Under this condition,the potency of neomycin shaking flask fermentation was as follows:19 210 u/mL,23%higher than before.
作者
童冰
郭萌
TONG Bing;GUO Meng
出处
《漳州职业技术学院学报》
2019年第2期104-108,共5页
Journal of Zhangzhou Institute of Technology
基金
漳州职业技术学院科研计划资助项目(ZZY1509)
关键词
新霉素
神经网络
遗传算法
培养基优化
neomycin
neural network
genetic algorithms
medium optimization
