论文部分内容阅读
为了克服人工蜂群算法蜜源更新过程中的随机性并保留蜜源中个体序列合理的组合形式,通过分析基本蜂群算法更新公式的机理,提出一种改进GA(Genetic A1gorithm)机制融合的二进制蜂群算法.算法以二进制编码,首先依概率对任意两蜜源进行“去同存异”操作后随机排列,将排列结果放入到其中某个体中形成新个体.然后依概率进行二进制个体的“翻转”操作,上述两种操作从其本质上相当于GA的类交叉和类变异操作;其次利用GA机制收敛性的证明方式在理论上证明算法是收敛的.最后通过应用不同特性的多维基准函数和算法之间的比较验证改进蜂群算法具有良好的收敛能力和鲁棒性.
In order to overcome the randomness of artificial bee colony algorithm in updating honeybees and to keep a reasonable combination of individual sequences in honeybee, a binary bee colony with improved genetic algorithm (GA) is proposed by analyzing the mechanism of basic colony algorithm updating formula. The algorithm is coded in binary system, and firstly randomly arranges the random two neighbors according to the probability, then puts the arrangement result into one of the individuals to form a new individual, and then according to the probability of binary entity “Flip” operation, the above two operations are essentially equivalent to GA’s class-crossing and class-mutation operations; secondly, using GA’s convergence proof proves that the algorithm converges theoretically. Finally, by applying multidimensional The comparison between benchmark function and algorithm shows that the improved bee colony algorithm has good convergence ability and robustness.