BP神经网络和遗传算法在红发夫酵母培养基优化中的应用.pdf

2008 14 6 ): 834~837 2008-12-25 应用与环境生物学报 ， ( Chin J Appl Environ Biol=ISSN 1006-687X DOI: 10.3724/SP.J.1145.2008.00834 BP神经网络和遗传算法在红发夫酵母 培养基优化中的应用 1, 2 1* 彭 锋 李安明 1 2 ( 中国科学院成都生物研究所 成都 610041; 中国科学院研究生院 北京 100049) 摘 要 运用BP神经网络对红发夫酵母发酵培养基组成进行建模以及预测类胡萝卜素产量，在此基础上采用遗传算 法对此模型进行全局寻优 . 得到红发夫酵母发酵培养基的最佳配比为：蔗糖45.10 g/L ，硫酸铵3.00 g/L ，硫酸镁 0.80 g/ L ，磷酸二氢钾 1.40 g/L，酵母膏3.00 g/L ，氯化钙0.50 g/L ，类胡萝卜素产量达到8.20 mg/L，干重达到9.47 g/L. 采用上述 方法优化后的培养基使类胡萝卜素的产量比起始培养基提高了95.90%. 图4 表 1 参 13 关键词 红发夫酵母；虾青素；BP神经网络；遗传算法；培养基优化 CLC Q936 Optimization of Carotenoid Fermentation Medium of Phaffi a rhodozyma by BP Neural Network and Genetic Algorithm PENG Feng1, 2 LI Anming1* 1 ( Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu 610041, China) (2 Graduate University of Chinese Academy of Sciences, Beijing 100049, China) Abstract In order to improve the carotenoid yield of Phaffi a rhodozyma, Uniform Design was used for trial design of the formula medium components, then back-propagation neural network was established to model the relationships between the carotenoid yield and medium components and predict the concentration of medium components. Genetic algorithm (GA) was applied for integrated optimization of the model. The optimum combination of the medium was obtained: sucrose 45.10 g/L, ammonium sulfate 3.00 g/L, magnesium sulfate 0.80 g/L, potassium dihydrogen phosphate 1.40 g/L, ye