针对高维复杂函数的优化问题,提出一种带有倒位变异的差分进化算法。当个体适应度值连续几代不变时,对前一代的最优个体进行倒位变异,以增强种群的多样性,使其跳出局部最优。数值实验结果表明:该算法全局搜索能力强,收敛速度快,且鲁棒性好。
For complex functions with high dimensions,a differential evolution algorithm with inversion mutation is presented.When the fitness of the particle doesn't change under successive several generations,optimal particle of the previous generation is used to the inversion mutation,so that it can jump out of the local optimization,which enhances the diversity of population.The numerical results show that the proposed algorithm has not only remarkable global searching ability and fast convergence speed, but also strong robustness.