基于种群划分策略的人工蜂群算法的改进研究

Research on Improvement of the Artificial Bee Colony Algorithm Based on Population Partition Strategy

针对人工蜂群算法在高峰多维情况下早熟性收敛和容易陷入局部寻优的问题,提出了基于种群划分策略的人工蜂群算法.此算法利用个体的适应值与子种群的适应值的相似度,将种群划分为同参不同源的子种群,一方面优化了子种群的数量,另一方面保证了种群量的多样性和解的精确度.最后,通过实验性计算,计算了新算法和常规人工蜂群算法的迭代次数,平均误差等参数,验证了新算法的高效低误差性.

Aiming at the problem of premature convergence and easy to fall into local optimization in artificial bee colony algorithm in the case of peak and multidimensional,the artificial bee colony algorithm based on population partition strategy is proposed.In this algorithm,by using the individual fitness and the fitness of the sub population similarity,the population will be divided into different sub populations with the same parameters.On the one hand,the number of sub populations are optimized,on the other hand,the diversity of the population and the accuracy of reconciliation are ensured.Finally,through the experimental calculation,the parameters of the iteration number and average error etc.of the new algorithm and the conventional artificial bee colony algorithm are calculated,and the new algorithm is proved to be high efficient and low error.