摘要
针对传统遗传算法容易早熟及收敛速度慢的缺陷,提出了一种新的基于信息熵的遗传策略.该策略根据当前种群个体熵与种群熵的变化自适应调整遗传算子的各项参数,从而使得种群多样性得到保证,提高算法的全局搜索能力.试验结果表明了该方法在运行过程中能避免早熟的发生,在处理复杂问题时表现出较高的性能.
To solve the problem of premature convergence and slow convergence flaws in the traditional genetic algorithm,a new genetic strategy which is based on information entropy is proposed. In this strategy, the algorithm will adaptively adjust its operator parameters in accordance with the current individual entropy and the population entropy. It ensures the diversity of the population and improve the global search capabilities. The result shows that this method can avoid the occurrence of premature convergence and provide excellent performance indealing complex problems.
出处
《甘肃联合大学学报（自然科学版）》
2008年第5期69-71,共3页
Journal of Gansu Lianhe University :Natural Sciences
关键词
遗传算法
函数优化
种群熵
早熟
genetic algorithm
function optimization
population entropy
premature convergence
