Abstract: Artificial colony algorithm has strong ability of exploration, but has poor exploitation ability, low search accuracy, slow convergence speed during the later period. In order to solve the above problems, an artificial colony algorithm based chaotic mechanism is proposed. In order to avoid the imbalance between the exploration and exploitation ability, the historical average optimal solution was added to the search strategy; and chaotic sequence was used in the late period to make the population mutation if the population falls into local extremum, to enhance the exploitation ability and the quality of solutions, maintain the diversity of the population. Experimental results tested on functions show that the improved algorithm is superior to the basic ABC algorithm and other improved algorithms in solving speed and precision.


Key words: swarm intelligence, artificial bee colony algorithm, search strategy, chaotic mutation, function optimization