基于改进多目标粒子群优化算法的发酵过程补料优化控制  

Optimal control of a fed-batch fermentation process based on an improved multi-objective particle swarm optimization algorithm

作　　者：张超然[1] 王建林[1] 赵利强[1] 于涛[1] 

机构地区：[1]北京化工大学信息科学与技术学院,北京100029

出　　处：《北京化工大学学报（自然科学版）》2014年第1期68-73,共6页Journal of Beijing University of Chemical Technology(Natural Science Edition)

摘　　要：提出了一种基于改进多目标粒子群优化算法(MOPSO)的发酵过程补料优化控制方法,并将该方法用于工业酵母发酵过程补料速率的优化控制。改进的MOPSO算法利用约束违反程度信息修正种群的学习进化公式,减缓了部分粒子在约束边界附近的飞行速度,提高了算法的全局搜索能力。仿真实验结果表明,改进的MOPSO算法能够获得最优的补料轨迹,有效地实现了工业酵母发酵过程补料速率的多目标优化控制。An optimization method for feed profiles in a fed-batch fermentation process based on an improved multiobjective particle swarm optimization(MOPSO) algorithm has been derived and the method was used to optimize the feed profiles for a yeast fermentation process. The parameters in the improved MOPSO algorithm were modified, and the speed of constraint boundary particles in the feasible region was slowed down so as to enhance the ability to search for constraint boundaries. The experimental results showed that the proposed method can effectively realize optimal control of the yeast fermentation process by using the optimal feed profiles given by the improved MOPSO.

关 键 词：发酵过程 优化控制 多目标粒子群优化算法 约束处理技术 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]