基于均匀设计的蚁群算法参数设定

Parameter Establishment of an Ant System Based on Uniform Design

蚁群算法的参数设置一直是依靠经验和试验来确定,造成试验工作量大且难以得到最优的参数组合,影响了算法的使用.通过将蚁群算法基本模型的参数设定问题描述成均匀设计中多因素多水平的试验设计,从而能够用较少的试验很快设定算法参数的取值.TSP问题的仿真试验表明,利用均匀设计得到的参数组合可使蚁群算法获得较优的运行性能,说明了该方法的可行性和有效性.

The enactment of the parameters of an ant system is determined by experience and experiment. This leads to heavy work load and makes the optimal combination of the parameters difficult to obtain. Uniform design method is used to convert the problem of parameter establishment into the experimental design of multi-factor and multi-level and reduces the work load of experiment greatly of simulation. The method applied to the traveling salesman problem shows good performance feasibility and effectiveness.