改进差分进化算法及其在发酵优化中的应用

针对发酵过程的补料优化问题,提出一种改进的差分进化算法.为了克服基本差分进化算法在全局最优点附近搜索速度较慢、精度较低的缺点,引入单纯形加速算子以提高算法收敛速度,而针对算法易过早收敛的缺点引入混沌迁移算子,以提高算法种群多样度,增强算法跳出局部最优解的能力.对于有约束优化问题,利用3个准则进行选择操作,使求得的最优解满足约束条件.将改进的算法用于某一类补料分批发酵过程,提高了发酵最终产物产量,表明了该算法的有效性.

A Modified Differential Evolution Algorithm and Its Application to Optimization of Fed-Batch Fermentation

To solve the feeding optimization problem of a certain fermentation process,a modified differential evolution(MDE) algorithm is proposed.Because the basic differential evolution algorithm becomes slower in the vicinity of the global optimum with low accuracy,a simplex acceleration operator is introduced to expedite the convergence speed of the algorithm.On the other hand,a chaotic migration operator is introduced to prevent the algorithm from prematurity,so as to improve the diversity of populations and enhance the algorithm's capability to rise above what a local optimal solution is available to do.For the constrained optimization problems,three criteria are selected to cause the optimal solution to satisfy constraining conditions.MDE was applied to optimize the feeding rate of a fed-batch fermentation process,and the result revealed that it improves the finished product output and that the algorithm is effective.