Abstract: Concerning the disadvantages of ant colony optimization such as easily plunging into a local optimum and slow convergence speed in continuous optimization, a new Ant Colony Algorithm (ACO) with dimension mutation operator (DMCACO) was presented. In this algorithm, target individuals which led the ant colony to do global rapid search were determined by dynamic and stochastic extraction and the current optimal ant searches in small step nearly. The concept of dimension diversity was defined and the worst of diversity was mutated with introducing the dimension mutation operator: the positions of all ants in this dimension were distributed in the feasible range evenly. The simulation on typical test functions indicates that this algorithm has excellent global optimization and rapid convergence.

Key words: Ant Colony Algorithm (ACO), continuous domain, diversity, dimension mutation, global optimization

摘要： 针对在连续优化中，蚁群算法(ACO)存在的收敛速度慢和易陷入局部最优的问题，提出了一种新的含维变异算子的连续域蚁群算法(DMCACO)。该算法采用动态随机抽取的方法来确定目标个体，引导蚁群进行全局的快速搜索，同时在当前最优蚂蚁邻域内进行小步长的局部搜索。在定义了维多样性概念的基础上，引入维变异算子对维多样性最差的维进行变异：让所有蚂蚁在该维上的位置重新均匀分布在可行区域上。对测试函数所做的仿真实验表明，该算法具有优良的全局寻优能力和快速的收敛能力。

关键词: 蚁群算法, 连续域, 多样性, 维变异, 全局寻优