摘要
设计了一种新的单亲遗传算法,该算法对个体和基因分别计算其适应值,并将适应值最差的基因 进行变异,从而大大提高了遗传算法的全局收敛速度。以n皇后问题为例,采用这种算法求解3000个皇 后问题所用时间平均约为55min,而求解100个皇后问题所用时间平均只有约124ms。实验表明,该算法 不仅能在短时内找到全局最优解(精确解),而且具有良好的稳定性。
A new one-parent genetic algorithm is designed,by which the fitness of individuals and genesis is calculated. The worst fitness value is changed,thus the global convergent speed is improved greatly. By taking the queen problem for example,the method is used to derive more than 3000 queens problem with time average about 55 min, while 124ms is used for deriving 100 queens problem. Test indicates that the method can not only be used to find out the global optimal solution (precision solution) in a short time, but have excellent stability.
出处
《长江大学学报（自然科学版）》
CAS
2004年第2期74-76,共3页
Journal of Yangtze University(Natural Science Edition)
关键词
单亲遗传算法
遗传变异
基因换位
算法
one-parent genetic algorithm
genetic variation
genesis exchange
evaluation function
