Abro, Abdul Ghani
(2013)
Performance Enhancement Of Artificial Bee Colony Optimization Algorithm.
PhD thesis, Universiti Sains Malaysia.
Abstract
Artificial Bee Colony (ABC) algorithm is a recently proposed bio-inspired optimization algorithm, simulating foraging phenomenon of honeybees. Although literature works have revealed the superiority of ABC algorithm on numerous benchmark functions and real-world applications, the standard ABC and its variants have been found to suffer from slow convergence, prone to local-optima traps, poor exploitation and poor capability to replace exhaustive potential-solutions. To overcome the problems, this research work has proposed few modified and new ABC variants; Gbest Influenced-Random ABC (GRABC) algorithm systematically exploits two different mutation equations for appropriate exploration and exploitation of search-space, Multiple Gbest-guided ABC (MBABC) algorithm enhances the capability of locating global optimum by exploiting so-far-found multiple best regions of a search-space, Enhanced ABC (EABC) algorithm speeds up exploration for optimal-solutions based on the best so-far-found region of a search-space and Enhanced Probability-Selection ABC (EPS-ABC) algorithm, a modified version of the Probability-Selection ABC algorithm, simultaneously capitalizes on three different mutation equations for determining the global-optimum. All the proposed ABC variants have been incorporated with a proposed intelligent scout-bee scheme whilst MBABC and EABC employ a novel elite-update scheme.
Actions (login required)
 |
View Item |
