一种基于反馈策略的自适应选择人工蜂群算法

雇用蜂觅食策略对人工蜂群算法性能有较大影响,而单一的觅食策略难以适用于所有问题的搜索空间,并且算法运行的不同阶段所适合的搜索策略也不尽相同.因此,如何为一个给定的函数优化问题选择最佳的觅食策略尤为重要.针对这一问题,提出了一种基于反馈的觅食策略自适应人工蜂群算法SSABC,该算法能够在优化过程中为一个给定的优化问题自动选择最佳的觅食策略.实验表明,与经典ABC(artificial bee colony algorithm),PSO(particle swarm optimization),DE(differential evolution),GA(genetic algorithm)算法相比,SSABC算法的寻优能力有较大提高.

A Strategy Self-Adaptive Selection Bee Colony Algorithm Based on Feedback

Employed bee foraging strategies have a greater impact on the performance of artificial bee colony algorithm. The single foraging strategy is difficult to apply to all the search space of the problems. And the different stages of the algorithm performs differently by using different employed bee foraging strategies.How to choose the best foraging strategy is very important for the given function optimization problem. To solve this problem， a strategy self-adaptive selection colony algorithm was presented， based on feedback. The optimal foraging strategy could be automatically selected for the given problem during the optimization process using the praposed algorithm. Experimental results showed that compared with the ABC (artificial bee colony algorithm)， the PSO (particle swarm optimization algorithm)， the DE (differential evolution algorithm)， and the GA (genetic algorithm)， the optimization capability of the SSABC algorithm has been improved greatly.