论文部分内容阅读
首先从理论上对三种传统遗传算子的作用进行了定量分析,在此基础上提出了一种新的遗传算子——扩散算子,并利用模拟退火法给出了扩散概率.这种增加扩散算子的遗传算法,较好地克服了传统遗传算法易发生成熟前收敛和收敛速度过慢的缺点,仿真结果表明了其实用性和有效性.
Firstly, the quantitative analysis of the roles of the three traditional genetic operators is carried out theoretically. On the basis of this, a new genetic operator - diffusion operator is proposed, and the diffusion probability is given by the simulated annealing method. The genetic algorithm to increase the diffusion operator is better than the traditional genetic algorithm to overcome the shortcomings of convergence and convergence before maturity. The simulation results show its practicability and effectiveness.