Improvement And Application Of Artificial Bee Colony Algorithm

Artificial Bee Colony (ABC) algorithm is one of the most recently introduced swarm-based algorithms, which demonstrates good performances for solving optimization problems. Due to its simplicity and ease of implementation, the ABC algorithm has captured much attention and has been applied to solve many practical optimization problems. However, the original ABC still has some drawbacks. In this paper, inspired by Particle Swarm Optimization (PSO) and Demon strategy we present an improved ABC algorithm (IABC) by incorporating the particle best and global best into updating equations to solve unconstrained problem. And also we use the small section method to improve diversity of the swarm. Four benchmark functions are used to validate the performance, and the result indicates that the IABC is better than the original ABC. In additional, we also applied the IABC on predicting longitudinal dispersion coefficients. Datasets are used to demonstrate the performance and applicability of IABC. Experimental results are compared with the previous methods and the original ABC algorithm. Those results indicate that the IABC algorithm performs better than other methods.In the last part of this paper, with the penalty function, Deb’s rule and Multiple parameters perturbation method we propose another improved ABC algorithm for constrained optimization problem (IABC(COP)). A series of well-known constrained problems from relevant literatures are employed to verify the performance of our approach. Experimental results are compared with GA, CSOMGA, HM, PSO and CMA-ES algorithms. Those results indicate that the proposed IABC(COP) algorithm performs better than, or at least comparable to other state-of-art algorithms in terms of the quality of final solutions and the convergence rate.