| In recent years,the swarm intelligent algorithm has made great progress because of solving many large-scale complexity problems.The artificial bee colony algorithm is a newer swarm intelligent algorithm.Because of its strong global convergence,greedy heuristic search characteristics and the rapidity of solving the problem and other superior performance has been widespread concern.This paper briefly introduces the biological background of the artificial bee colony algorithm,and introduces the algorithm model of artificial bee colony algorithm in detail.At present,the basic artificial bee colony algorithm still has the following problems:(1)Artificial bee colony algorithm has the "premature" convergence defect.(2)Artificial bee colony algorithm has a good ability to explore,but the lack of development capacity,local search ability is weak,and the convergence rate is relatively slow.In view of the shortcomings of the basic artificial bee colony algorithm,the main work of this paper is as follows:(1)To improve the choice of hired bee strategy,dynamic adjustment of the depth of mining and breadth of the proportion of exploration.Improve the efficiency of the original artificial bee colony algorithm,and improve the convergence speed of the algorithm.(2)To improve the observation bee selection strategy,the original artificial colony using roulette rules,easy to make the algorithm into a "precocious".In this paper,we first choose a rich area of nectar,and then randomly selected.So as to maintain the diversity of the same time to enhance the understanding of the accuracy.Through the tests on 15 reference functions,the improved algorithm proposed in this paper has improved the robustness,convergence speed and robustness of the algorithm compared with the original artificial bee colony algorithm.Especially when the latitude rises to a certain time,the original artificial bee colony algorithm can not be satisfied,and the improved algorithm can still achieve satisfactory results. |
PDF Full Text Request