In this paper, we concentrate on the modification of the evolved bat algorithm (EBA), designed for solving numerical optimization by utilizing the scheming idea of Artificial Bee Colony algorithm (ABC).Three roles of bat colony and six successive processes are realized to accelerate the convergence characteristic of the modificatory algorithm, namely evolved bat colony algorithm (EBC). In the initialization, two roles (employed bats and onlooker bats) are set with equal probability, and the movement law of EBA is applied for obtaining candidate solutions by the greedy selection strategy. The last modified phase is movement phase of scout bats, the employed bats become scouts with randomly search in this phase, when one solution cannot be further improved any more. In the experiments, five well-know benchmark functions are utilized to evaluate the performance of EBC algorithm. The obtained results show that the EBC algorithm is superior to EBA according to solution quality and robustness.
|Number of pages||8|
|Journal||Journal of Information Hiding and Multimedia Signal Processing|
|Publication status||Published - 1 Jul 2018|
- Greedy selection strategy
- Numerical optimization
- Swarm intelligence