Hybridizing Invasive Weed Optimization with Firefly Algorithm for Unconstrained and Constrained Optimization Problems
This study presents a hybrid invasive weed firefly optimization (HIWFO) algorithm for global optimization problems. Unconstrained and constrained optimization problems with continuous design variables are used to illustrate the effectiveness and robustness of the proposed algorithm. The firefly...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JATIT
2017
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/17172/ http://umpir.ump.edu.my/id/eprint/17172/ http://umpir.ump.edu.my/id/eprint/17172/1/fkp-2017-nmyahya-%20Hybridizing%20Invasive%20Weed%20Optimization.pdf |
| Summary: | This study presents a hybrid invasive weed firefly optimization (HIWFO) algorithm for global optimization
problems. Unconstrained and constrained optimization problems with continuous design variables are used
to illustrate the effectiveness and robustness of the proposed algorithm. The firefly algorithm (FA is effective in local search, but can easily get trapped in local optima. The invasive weed optimization (IWO) algorithm, on the other hand, is effective in accurate global search, but not in local search. Therefore, the idea of hybridization between IWO and FA is to achieve a more robust optimization technique, especially to compensate for the deficiencies of the individual algorithms. In the proposed algorithm, the firefly method is embedded into IWO to enhance the local search capability of IWO algorithm that already has very good exploration capability. The performance of the proposed method is assessed with four well-known unconstrained problems and four practical constrained problems. Comparative assessments of performance of the proposed algorithm with the original FA and IWO are carried out on the unconstrained problems and with several other hybrid methods reported in the literature on the practical constrained problems, to illustrate its effectiveness. Simulation results show that the proposed HIWFO algorithm h
as superior searching quality and robustness than the
approaches considered. |
|---|