Mixed Strategy Of Artificial Bee Colony Algorithm Research

Swarm Intelligence Algorithm is a kind of stochastic optimization algorithm based on behavior of biological swarm, which provides a new method to solve global optimization problem existed in the fields of computer science, management science, control engineering and so on. So it becomes a focus among researchers in a long term.Artificial bee colony (ABC) algorithm is a swarm intelligence optimization algorithm based on the particular intelligent behavior of honeybee swarms. Because of the advantages of its less control parameters, easily programming and simple calculation, it has received more and more attention by scholars, but the research and application on artificial bee colony algorithm is still in primitive stage at present. There are still many problems to be improved and solved. In order to effectively improve performance of artificial bee colony algorithm (ABC), the hybrid strategies from various angles are studied in the paper.Firstly, the selection strategy of ABC algorithm is analyzed in this paper. In order to improve the population diversity and avoid the premature, several selection strategies, such as rank selection strategy, disruptive selection strategy and tournament selection strategy, are analyzed and compared through simulation, and the results show that the modified algorithm outperforms the basic ABC algorithm.Secondly, this paper proposes an improved artificial bee colony (ABC) algorithm called chaotic artificial bee colony algorithm with self-adapting search space (SA-CABC). The main idea is self-adapting adjust search space according to the results of each optimization, and takes use of the randomicity and ergodicity properties of the chaos to break away the local optima, and ultimately finds the global optima. Simulation results show that the SA-CABC algorithm not only accelerates the convergence rate and improves its accuracy, but also effectively avoids the premature convergence. SA-CABC algorithm is better than the basic ABC, and provides excellent performance in dealing high-dimensional complex problems.Finally, a novel Bi-group differential artificial bee colony algorithm (BDABC) which is combined with differential evolution (DE) algorithm is proposed. In this algorithm, an initialization strategy based on the opposition-based learning is applied to diversify the initial individuals in the search space. All of the individuals are divided into two populations randomly, and the evolutions of two sub-groups are parallel performed with different optimization strategies. The interactive learning strategy is introduced to accelerate the convergence speed. Experimental results on six benchmark functions show that the BDABC algorithm can not only effectively avoids the premature convergence, but also significantly improves the global optimization ability and the convergence speed.