Bat algorithm lags behind other modern metaheuristic algorithms in terms of search efficiency, due to premature convergence. Once trapped in any sub-optimal region, the algorithm is unable to escape because of deficiency in population diversity. To address this, an enhanced Bat Algorithm (EBA) is introduced in this paper. The EBA algorithm comes with adaptive exploration and exploitation capability, as well as, additional population diversity. This enables EBA improve its convergence ability to find even better solutions towards the end of search process, where standard BA is often trapped. To illustrate effectiveness of the proposed method, EBA is applied on non-linear, non-convex economic dispatch problem with a power generation system comprising of twenty thermal units. The experimental results suggest that EBA not only saved power generation cost but also reduced transmission losses, more efficiently as compared to original BA and other methods reported in literature. The EBA algorithm also showed enhanced convergence ability than BA towards the end of iterations.
|Title of host publication||Recent Advances on Soft Computing and Data Mining|
|Subtitle of host publication||Proceedings of the Fourth International Conference on Soft Computing and Data Mining (SCDM 2020), Melaka, Malaysia, January 22–23, 2020|
|Editors||Rozaida Ghazali, Nazri Mohd Nawi, Mustafa Mat Deris, Jemal H. Abawajy|
|Number of pages||10|
|Publication status||Published - 5 Dec 2019|
|Name||Advances in Intelligent Systems and Computing|