Year: 2020
Type: conference-paper
Abstract: In response to the problems of slow convergence, low convergence accuracy and easy to fall in local optimum of the standard artificial bee colony (ABC) algorithm, this paper proposed the following two improvement measures: Firstly, on the basis of using the current optimal solution, the gradient modification strategy was used to reduce the leading role of the current optimal solution, so that the nectar resources could be fully searched to ensure the diversity of the population; Secondly, the crossover and mutation operation in genetic algorithm was used to replace the random search of scout bees, and the prior information was effectively utilized, which not only improved the search efficiency but also ensured the diversity of population to a certain extent. The numerical experiment results show that the proposed algorithm has good convergence accuracy and robustness, and also presents an excellent performance in multi-dimensional optimization.
Authors: Yi Yu, Yonggang Wu, Xinglong Liu
Institution: Huazhong University of Science and Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
Award: G5126688454
OpenAlex ID: W3022891678