摘要
遗传算法是解决优化问题的一种重要而有效的方法,在很多领域中得到了广泛的应用。在实际应用过程中,"过早收敛"是遗传算法经常遇到的问题之一,其主要原因是进化过程中个别优秀个体的迅速繁殖导致种群多样性的过早丧失。针对这一问题,提出了一种基于改进种群熵的多样性评价方法,并根据种群多样性评价及个体的适应度,从宏观和微观两方面对个体操作概率进行动态调整。仿真实验表明改进算法具有良好的全局搜索能力,一定程度上避免了过早收敛。
Genetic algorithm is an important and effective way to solve optimization problems and has been used in many fields. Premature is one of the problems that often occur when using genetic alrigothm in practice. The major reason is that some individuals, whose fitnesses are higher, increase too fast thus resulting in the loss of population's diversity too early. This paper introduces an improved measurement of population diversity based on entropy, and an improved adaptive genetic algorithm is presented. The numerical simulations show that the improved algorithm is more effective for realizing the global opitimization and can avoid premature effectively.
出处
《计算机仿真》
CSCD
2008年第2期206-208,231,共4页
Computer Simulation
关键词
遗传算法
种群熵
种群多样性
自适应
Genetic algorithm
Entropy of population
Population diversity
Adaptive