Abstract: Ant colony algorithm is a novel simulated evolutionary algorithm which can provide a new way of thinking for solving complex combinatorial optimization problems. But the basic ant colony algorithm is of slow convergence and easy-to-stagnation, and easily converges to local solutions. Several optimization strategies proposed in this paper can improve several aspects of selection strategy, local search and pheromone modification to make the algorithm not easy to run into local optimal solution, and can quickly converge to the global optimal solution. Experimental results show that the modified strategies are more reasonable, effective and accurate.

Key words: ant colony algorithm, parameter optimization, pheromone, variation

摘要： 蚁群算法是一种新型的模拟进化算法，为求解复杂的组合优化问题提供了一种新的思路，但基本的蚁群算法收敛速度慢，易于停滞，并且很容易收敛于局部解。提出从几种优化策略对算法的选择策略、局部搜索、信息量修改等方面进行改进，使算法不易陷入局部最优解，并且能较快地收敛到全局最优解。实验结果表明，此改进策略是比较合理、有效和准确的。

关键词: 蚁群算法, 参数优化, 信息素, 变异特征