蚁群算法不确定性分析

Uncertainty analysis of ant colony optimization algorithm

蚁群算法作为一种开创性的生物仿真算法,因其具有并行性、鲁棒性等优良性质得到了广泛的应用。在对蚁群算法进行系统仿真的实验中,发现蚁群算法存在很多不确定因素。这些因素对蚁群算法的性能造成不同程度的影响,作为一种基于实验的研究性的探讨,本文对所发现的不确定因素做了分析,并根据分析结果对蚁群算法作了相应的改进。

Ant Colony Optimization algorithm(ACO) is a kind of innovative biology emulation algorithm. ACO has been used in many fields because it has some excellent qualities,such as parallelism and robustness. In the process of the system emulation experiment,many uncertain factors of ACO were found. These factors would influence qualities of ACO in various degrees. As a kind of learned disquisition based on experiment,the uncertain factors which were found in experiment were analyzed and ACO was improved based on the analysis result.