蚁群和微分进化相融合的自适应优化算法

摘要

An self-adaption hybrid optimization algorithm with fusion of Ant Colony(AC) algorithm and Differential Evolution(DE) algorithm is proposed to solve the problem of complicated function global optimization. The new algorithm utilizes DE algorithm with the mutation and crossover operation to avoid AC algorithm premature convergence, and utilizes the pheromone positive feedback effect to speed up evolutionary algorithm search, and automatically adjusts searching range. Experimental results show that compared to the AC algorithm and DE algorithm, this new algorithm greatly improves the global optimization search efficiency.%为解决复杂函数的全局优化问题，提出一种蚁群和微分进化相融合的自适应优化算法。采用微分进化算法的变异和交叉操作避免蚁群算法过早收敛，使用蚁群算法的寻优路径信息素正反馈机制来加速微分进化算法收敛于最优路径，并自动调整搜索范围。实验结果表明，与蚁群算法和微分进化算法相比，该算法全局优化的搜索效率较高。