一种基于粒子群参数优化的改进蚁群算法

蚁群算法是一种应用广泛、性能优良的智能优化算法,其求解效果与参数选取息息相关.鉴于此,针对现有基于粒子群参数优化的改进蚁群算法耗时较大的问题,提出一种新的解决方案.该方案给出一种全局异步与精英策略相结合的信息素更新方式,且通过大量统计实验可以在较大程度上减少蚁群算法被粒子群算法调用一次所需的迭代代数.仿真实验表明,所提出算法在求解较大规模旅行商问题时具有明显的速度优势.

Improved ant colony optimization algorithm based on particle swarm optimization

Ant colony optimization(ACO) algorithm is an intelligent algorithm which has a wide range of applications
and better performance, and its search quaility is closely related with the parameters selection. Therefore, aiming at the
large time-consuming problem of the existing improved ACO alogorithm, a novel ACO algorithm based on particle swarm
optimization(PSO) algorithm is proposed. The new pheromone update method is presented, which combines the global
asynchronous feature and elitist strategy. Moreover, the iteration number of ACO algorithm invoked by PSO algorithm
is reduced significantly by large amounts of statistical experiments. The simulation results show that the proposed ACO
algorithm has obvious advantage in search speed when it is used for solving the large-scale traveling salesman problem.