An Artificial Bee Colony Algorithm Guided by Lévy Flights Disturbance Strategy for Global Optimization

In order to solve the problems of the slow convergence speed, low precision and easy trapping in local optimal solutions in an artificial bee colony (ABC) algorithm, a novel modified ABC algorithm based on the Levy flights disturbance mechanism is proposed in this chapter. It attempts to increase the exploration efficiency of the solution space for global optimization. The modifications focus on the solution construction phase of the artificial bee colony algorithm. In addition, to further balance the search processes of exploration and exploitation, the modification forms of the onlookers and scouts search strategy is proposed in this chapter. It could avoid local optimum. And it also could greatly improve convergence speed and solution precision on the basis of keeping strong global optimization performance of the proposed algorithm. Simulation experiment results based on typical benchmark functions show that the proposed algorithm is more effective to avoid premature convergence and to improve solution precision than some other ABCs and several state-of-the-art algorithms.