改进的小生境混合遗传算法在函数优化上的应用

为了提高经典小生境遗传算法的收敛性能,加强局部寻优能力,设计了一种新的小生境混合遗传算法.通过判断算法的在线性能指标Xe(s),将模拟退火算法巧妙地融入算法的后期,并针对小生境遗传算法的特点选用格雷码编码,同时设计了自适应的遗传交叉算子.用一个Shubert多峰值函数对改进的算法进行验证,结果表明:新算法的收敛性能和进化效率得到提高,局部寻优能力也有加强.

Application of improved niche hybrid genetic algorithm in function optimization

In order to improve the convergence of classical niche genetic algorithm,strengthen local optimizing ability,a new niche hybrid genetic algorithm is designed.By judging online performance indicators Xe (s) of algorithm,the simulated annealing algorithm is cleverly fused into later algorithm,and aiming at characteristics of the niche genetic algorithm,choose Gray code coding,at the same time,adaptive genetic crossover operator is designed.Using a Shubert multimodal function to validate the improved algorithm,the results show that the convergence of the new algorithm and evolutionary efficiency is improved,the local optimization ability is also strengthened.