Modified Artificial Bee Colony Algorithm And Its Application Research

Artificial Bee Colony(ABC) algorithm is a novel swarm intelligent algorithm inspired by the foraging behaviors of honeybees. And it has attracted the attention of researchers and been widely used in solving many numerical and practical engineering optimization problems. There are so many kinds of swarms in the world. Their intelligence level could vary from one swarm to another swarm. Self-organization and division of labour are two key feature of a swarm system which results in collective behaviour by means of local interactions among simple agents. ABC algorithm has its unique advantages when compared to other swarm based algorithms. For example, the ABC is simple in concept, easy to implement, and has fewer control parameters.Although the ABC algorithm has its own unique advantages, the convergence speed of ABC algorithm will be more slowly and the convergence accuracy will be lower with the high complexity and dimension of optimization problem. The extremal optimization(EO) algorithm is a general, heuristic algorithm based on local search, and has fast convergence speed. In order to improve the convergence speed and accuracy of ABC algorithm, this paper proposed a hybrid ABC-EO algorithm. Because of the EO algorithm’s inherent extremal dynamics mechanism, the hybrid ABC-EO algorithm can find the solution more efficiently.The hybrid ABC-EO algorithm introduced in this paper includes the incorporation of ABC algorithm and EO algorithm, and the incorporation of modified ABC algorithm and EO algorithm(respectively referred to as ABC_EO、IABC_EO). Then the combination of two algorithms also includes fixed numbers of iteration and unfixed numbers of iteration(respectively referred to as number one and number two). In a word, there are four hybrid algorithms applications in unconstrained function optimization problems in this paper(respectively referred to as ABC_EO1、ABC_EO2、IABC_EO1、IABC_EO2), which are the innovation points of this paper. We testified the performance of the proposed approach on six classic test functions and provided comparisons with other meta-heuristics. The experimental results show the superiority of the proposed approach on benchmark test functions.In addition, the hybrid ABC-EO algorithm can optimize constrained function problems. There are four simple constrained test functions for simulation experiments in this paper, and each test function can find the optimal solution, which verifies the good performance of the hybrid ABC-EO algorithm in dealing with the constrained optimization problems.