Improved biogeography-based optimization with random ring topology and Powell's method |
| |
Institution: | 1. School of Science, Guilin University of Technology, Guilin 541004, China;2. School of Mathematics and Statistics, Xidian University, Xi''an 710071, China;3. School of Science, Xi''an University of Posts and Telecommunications, Xi''an 710121, China;4. School of Mathematics and Computer Science, Shaanxi University of Technology, Hanzhong 723000, PR China |
| |
Abstract: | Biogeography-based optimization (BBO) is a competitive population optimization algorithm based on biogeography theory with inherently insufficient exploration capability and slow convergence speed. To overcome limitations, we propose an improved variant of BBO, named PRBBO, for solving global optimization problems. In PRBBO, a hybrid migration operator with random ring topology, a modified mutation operator, and a self-adaptive Powell's method are rational integrated together. The hybrid migration operator with random ring topology, denoted as RMO, is created by using local ring topology to replace global topology, which can avoid the asymmetrical migration operation and enhance potential population diversity. The self-adaptive Powell's method is amended by using self-adaptive parameters for suiting evolution process to enhance solution precision quickly. Extensive experimental tests are carried out on 24 benchmark functions to show effectiveness of the proposed algorithm. Simulation results were compared with original BBO, ABC, DE, other variants of the BBO, and other state-of-the-art evolutionary algorithms. Finally, the effectiveness of operators on the performance of PRBBO is also discussed. |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |
|