基于学习的进化规划算法

Evolutionary Programming Based on Learning

提出基于学习的进化规划算法 ,用以改进普通进化规划算法的性能 .该算法一方面通过学习种群整体的进化信息用以改善种群整体性能 ,具有大范围快速搜索的特点 ;另一方面该算法强调学习种群中个体的进化信息 ,单一个体以当前代的最优个体作为学习目标 ,用以加大当前最优解附近的搜索力度 ,具有局部“细搜”的特点 .该进化规划算法不仅能够加快算法的收敛速度 ,而且能够有效地保证种群的多样性 .用该方法可求解具有多个极值点的函数优化问题 。

To improve the efficiency of evolutionary programming ,a nevol evolutionary programming is proposed in this paper based on learning.On the one hand,The method emphasises on learning the evolutionary information of total population in order to improve the property of total population and is mainly characterized by fast goal search .on the other hand,the method emphasises on learning the information of individual in the population by learning the distance between an individual and the best individual at each generation and is mainly characterized by fast local search.The new algorithm not only keeps the population diversity but also has quicker convergence speed .It is applied to optimize functions with multi modal.The validity of the algorithm is shown by computer simulation results.