摘要
针对标准遗传算法在实际应用中存在的早熟问题,设计了标准遗传算法的一种改进形式——模糊自适应遗传算法。该算法是利用种群的方差和熵来衡量种群多样性,并根据每代种群的方差和熵设计模糊推理系统来自适应控制交叉概率和变异概率。通过多峰函数优化问题的仿真实验,表明了该模糊自适应遗传算法的可行性和有效性。
An improved form of normal genetic algorithm： Fuzzy adaptive genetic algorithm is designed. It is aiming at the premature problem that existed in the actual application of normal genetic algorithm. It used the population variance and entropy to measure the diversity of population, and in accordance with the each population of variance and entropy to design the fuzzy reasoning system for adaptive the controlled cross probability and mutation probability. The feasibility and effectiveness of the fuzzy self-adapting genetic algorithm is proved by the simulation testing of a multimodal function optimization problem.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第18期4783-4785,共3页
Computer Engineering and Design
基金
国家自然科学基金项目(60773009)
关键词
模糊推理系统
模糊遗传算法
自适应
熵
遗传算法
fuzzy reasoning system
fuzzy genetic algorithm
adaptive
entropy
genetic algorithm
