Dual-population genetic algorithm based on periodic slow change in radius parameter
基于半径参数周期性缓慢变化的双种群遗传算法*

Dual-population genetic algorithm introduced main population and reserve pop-ulation,it dominated the reserve population through controlling the change of radius parameter which maintained the diversity of group.However,the radius parameter changed too fast to control the reserve population.For the sake of overcoming this shortcoming,described the change of radius parameter by means of cosin function and proposed a dual-population genetic algorithm based on periodic slow change in radius parameter.The regularity for change of radius parameter was not only advantageous to keep the diversity of group,but also enhanced the local search.In order to estimate the maximum value of radius parameter,took the distance between individual and the centre of main population view as the maximum value of radius parameter.The results obtained show that the improved genetic algorithm is more effective than some current optimization algorithms.