Research on Performance of Artificial Bee Colony Algorithm Based on Benchmark Test Function
Ren Wang, Keke Shi, Shuqi Wang · IOP Conference Series Materials Science and Engineering · 2020
Abstract Because the artificial bee colony(ABC) algorithm has problems of poor development ability, slow convergence rate and easy to fall into local optimum, an improved ABC algorithm is proposed. The algorithm combines the global optimal guidance strategy, we propose a new location update formula to improve the efficiency of the iterative optimization process, and introduce the mining coefficient in the formula to improve the accuracy of global optimization. In addition, the beta distribution is introduced during the scouting bee phase, which improves the disturbance ability of the algorithm and prevents it from falling into local extremum effectively. Finally, the simulation results of four benchmark functions of Sphere, Rastrigin, Rosenbrock and Griewank show that compared with the traditional algorithm and other improved artificial bee colony algorithm, the proposed algorithm has faster convergence speed and better optimization ability.