中文摘要：

为了提高微生物发酵单位,提出了支持向量机（SVM）与基于实数编码遗传算法（RGA）相互耦联的优化控制策略。为解决发酵参数之间的耦合问题,进一步引入了模式的概念。SVM建立微生物发酵过程的预估模型,RGA以此模型为适应度函数计算最优控制模式。此策略应用于青霉素发酵过程的优化控制,效价比没有实施优化控制策略的生产结果提高了22.88%。

英文摘要：

To increase fermentation unit,the strategy of optimization control that combines the support vector machine（SVM）with genetic algorithm based on real coding（RGA）is proposed.To solve the coupling of fermentation parameters,the idea of pattern is also introduced.SVM establishes the prediction model for the microbial process,and RGA taking the model as fitness function calculates the optimal control pattern.The results show that the penicillin titer of fermentation process optimized is increased by 22.88% compared with that of fermentation process not optimized.