摘要
针对进化算法随机盲目搜索的缺点,提出一种新的自适应梯度信息指导交叉的进化算法.该算法首先利用混沌序列初始化种群,在迭代过程中,根据当前最优个体的梯度信息和种群与个体的聚集程度,自适应地确定最优个体的负梯度方向范围,在该范围内随机选择个体与当前最优个体进行算术交叉操作,使交叉后的个体以较大概率向较好解的方向进化.另外,引入自适应变异算子用于平衡算法的开发和探测能力.几个典型测试函数的实验结果表明,新算法具有较高的收敛精度.
Considering the drawback of randomly aimless search of evolutionary algorithm, a novel adaptive gradient guiding crossover evolutionary algorithm （ AGCEA ） is proposed. Chaotic sequences are used in the initialization of the evolutionary population. In the process of population evolution, the proposed algorithm can adaptively determine the negative gradient direction range of best individual according to the gradient information of the current best individual and the aggregation degree of population and individual, and then, it executes arithmetic crossover operator according choosing special individuals from the set range of the negative grads, which is got from the best individual of the current population. The evolving probability is very large. In addition, adaptive mutation operator is introduced to balancing the capabilities of exploration and exploitation. AGCEA is tested on several typical functions, and the experiment results indicate that it has the better convergence accuracy of the proposed algorithm.
出处
《小型微型计算机系统》
CSCD
北大核心
2011年第7期1331-1335,共5页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(60874070
61074069)资助
高等学校博士学科点专项科研基金项目(20070533131)资助
教育部留学回国人员科研启动基金项目资助
湖南省研究生科研创新项目(CX2009B038)资助
关键词
进化算法
自适应
梯度
算术交叉
变异
evolutionary algorithm
adaptive
gradient
arithmetic crossover
mutation
