Development of a modified bat algorithm using elite opposition — Based learning

Abstract

This research work presents the development of a modified bat algorithm (mBA) using elite opposition — based learning. The bat algorithm (BA), which is a nature inspired meta-heuristic algorithm, works on the basis of the echolocation behavior of bat. It, however, has a poor exploration capability leading to it easily getting stuck in local optima. The mBA is developed by modifying the BA with elite opposition — based learning (EOBL) in order to diversify the solution search space and the inertial weight in order to improve its exploitation capability. The performance of the proposed mBA was compared with that of the standard BA using seven benchmark optimization test functions. The simulation results showed that the mBA is superior to the standard BA by obtaining global optimal result of most of the test functions. All simulations were carried out using MATLAB R2013b.