基于创建解动态控制和局部搜索合并的蚁群算法

为了加快蚁群算法的收敛性和改善解的合理性,提出了一种改进的蚁群算法。该算法提出一种基于动态控制的策略,其目的是确保蚂蚁在搜索前期采用最大概率探索解,而在搜索后期,每只蚂蚁都在当前最优解附近搜索解,这在一定程度上提高了算法的收敛性能;其次,为得到更合理的解,对每只蚂蚁的局部搜索解中加入合并机制,这样集成了多个蚂蚁对最优解的搜索性能。实验结果表明:该方法性能优于传统的蚁群算法。

Improved ant colony algorithm based on dynamic control of solution construction and mergence of local search

To enhance the convergence rate of the ant colony algorithm and the reasonability of solution provided by this algorithm, an improved ant colony optimization algorithm is proposed. The main components of the proposed method include two parts. First, a newly strategy based on the dynamic control of solution construction is adopted. The purpose of this strategy is to ensure ants to exploit the solutions at the beginning of searching procedure with large probability while at the end of the searching procedure the solutions provided by each ant are obtained by searching around the best-so-far solution. By this way, the convergence of the ant colony algorithm is increased. Second, to obtain a more reasonable solution, a mergence mechanism, based on the local search result of each ant, is employed. Using this technique, the performances of exploiting solutions provided by several ants are integrated. The experiments demonstrate that the proposed method has better performance than the conventional ACO algorithm.