摘要
从种群规模和个体空间的角度分析了影响遗传算子性能的因素,在遗传算法(GA)的基础上设计了一种搜索区域可变、群体规模可变的最优家族遗传算法(OFGA).该算法提出了在优良解附近构造最优家族,最优解搜索将在这个微型空间中进行,在有限的时间内搜索到更优基因的家族将获得生存的权利.由于每一个家族的搜索区域大幅度减缩,伴随着种群规模的减缩,因此提高了算法的收敛速度.家族个体空间大小不变提高了解的精度.最后,给出了3个典型函数的模拟例子,通过与GA的对比结果看到,OFGA在数量级上提高了收敛速度,使最优解的精度也有很大提高,说明新的算法具有应用的潜力.
In the view of the population size and individual space, the factors that affect the performance of genetic operator were analyzed. A novel genetic algorithm (optimum family genetic algorithm), which had the ability to change its search space and population size, was presented based on the GA. In this algorithm, the optimum solution families close to quality individuals were constructed. Search will be done in this micro-space. The family that can search better gene in a limited time will win a new life. The convergent speed of the algorithm can be accelerated because of the reduction of the search space and population size. And the accuracy of the solution can be improved because of the invariability of the individual space. Three typical function tests are given in this paper. The results indicate that the OFGA can greatly improve the accuracy of the solution and the convergent speed by an order of magnitude, which show that this novel algorithm has application prospects.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2004年第1期77-80,共4页
Journal of Xi'an Jiaotong University
基金
陕西省自然科学研究基金资助项目 (2 0 0 1X1 7)
陕西省机械制造装备重点实验室资助项目 (0 3IF0 6)
关键词
遗传算法
种群规模
个体空间
Convergence of numerical methods
Optimization