Abstract:

Aiming at the shortcoming of artificial bee colony algorithms, such as the low convergence rate and easy to be trapped into the local optimums, an improved algorithm is proposed. First, a new crossover strategy is designed to make the group close to the optimal solution as soon as possible. Then, considering that the parameter of controlling the behavior of the scouts to avoid falling into local optimal setting is difficult and of a greater impact on the performance of the algorithm, a mutation strategy based on opposition-based learning is proposed to replace the scouts’ behavior. The simulation results on 10 standard test functions show that this new improved algorithm can obtain the global optimal solutions for almost all the functions, with fast convergence and good robustness. The performance of this algorithm is significantly better than the existing artificial bee colony algorithms.