基于遗传算法的羊肚菌液体发酵动力学模型的建立

发酵动力学研究是实现发酵过程最优化控制及发酵过程放大的前提条件。本研究对羊肚菌液体深层发酵动力学进行了研究, 在Matlab软件平台上, 应用遗传算法对比了真菌生长较常用的Monod与Logistic方程在描述羊肚菌生长动力学时的优劣, 并对羊肚菌的生长、胞外多糖产生和基质消耗模型进行了参数估计。结果表明, Logistic方程与试验数据拟和情况更好, 并给出了羊肚菌液体深层发酵的动力学模型具体形式, 经验证, 模型的平均误差为5.8%。利用遗传算法选择羊肚菌动力学模型, 并进行参数估计与其他方法相比具有快速、搜索面广、接近全局最优解的特点, 在处理分批发酵动力学问题上具有不可比拟的优势, 发酵动力学模型的建立为发酵过程优化及放大奠定了基础。

Genetic Algorithm for Fermentation Kinetics of Submerged Fermentation by Morchella

Fermentation kinetics is important for optimizing control and up-scaling fermentation process. We studied submerged fermentation kinetics of Morchella. Applying the genetic Algorithm in the Matlab software platform, we compared suitability of the Monod and Logistic models, both are commonly used in process of fungal growth, to describe Morchella growth kinetics. Meanwhile, we evaluated parameters involved in the models for Morchella growth, EPS production and substrate consumption. The results indicated that Logistic model fit better with the experimental data. The average error of this model was 5.8%. This kinetics model can be useful for optimizing and up-scaling fungal fermentation process.