A modified surrogate-assisted multi-swarm artificial bee colony for complex numerical optimization problems |
| |
| Affiliation: | 1. College of Engineering, Huaqiao University, Quanzhou, 362021, China;2. Fujian Provincial Key Laboratory of Data Intensive Computing, Quanzhou 362021, China;3. Fujian Key Laboratory of Autonomous Controllable Software, Quanzhou 362000, China;4. Postdoctoral Workstation of Linewell Software Company Limited, Quanzhou 362000, China;1. Department for Resilience of Maritime Systems, German Aerospace Center, Bremerhaven, Germany;2. Universidade Federal de Viçosa, Brazil;3. Group of Computer Architecture, University of Bremen, Germany;4. Universidade Federal de Minas Gerais, Brazil;5. Centro Federal de Educação Tecnológica de Minas Gerais, Brazil;6. Johannes Kepler University Linz, Austria;7. Cyber-Physical Systems, DFKI GmbH, Bremen, Germany;1. Assistant Professor, Department of ECE, Sri Ranganathar Institute of Engineering and Technology, Coimbatore;2. Associate Professor, Department of Electrical and Electronics Engineering, Hindusthan College of Engineering and Technology, Coimbatore |
| |
| Abstract: | Artificial bee colony (ABC) algorithm has been widely used in solving complex optimization due to its few control parameters and outstanding global search capability. However, ABC suffers from slow convergence rate, which limits its real-world applications. To overcome such disadvantage, this paper proposes a surrogate-assisted multi-swarm artificial bee colony (SAMSABC). The multiple swarm strategy is employed to further keep the diversity. To enhance the local exploitation capability, the orthogonal method is utilized to provide a guide vector. Moreover, to avoid wasting the computation resources, the fitness estimation strategy for artificial bee colony algorithm, as a surrogate-assistance technology, is designed. Finally, the experimental results of 20 benchmark functions verify its outstanding performance on solving complex numerical optimization problems. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
