Extension of Bat Algorithm on Standard Benchmark Functions
Anjali, Deepak Garg1, Richa Singh2, Sarika Bathija3
1Anjali, National Institute of Technology, Kurukshetra, Haryana, India.
2Deepak Garg, National Institute of Technology, Kurukshetra, Haryana, India.
3Richa Singh, National Institute of Technology, Kurukshetra, Haryana, India.
4Sarika Bathija, National Institute of Technology, Kurukshetra, Haryana, India.
Manuscript received on January 02, 2020. | Revised Manuscript received on January 15, 2020. | Manuscript published on January 30, 2020. | PP: 802-807 | Volume-8 Issue-5, January 2020. | Retrieval Number: E5889018520/2020©BEIESP | DOI: 10.35940/ijrte.E5889.018520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Meta heuristics are superior methods of finding, producing and even modifying heuristics that are able to solve various optimization problems. All Meta-heuristic algorithms are influenced by the nature. These types of algorithms tend to mimic the behaviour of biotic components in nature and are emerging as an effective way of solving global optimization algorithms. We have reviewed that no any algorithm is best for all applications due to lack of generality (no. of parameters), non-dynamic input values. So, this paper studied BAT algorithm deeply and found weakness in terms of non-dynamic pulse rate and loudness. In order to avoid being trapped into local optima these inputs are made dynamic with inclusion of levy Flight too. Performance of this proposed Modified BAT approach is evaluated using few standard benchmark functions. For justifying the superiority of Modified BAT, its performance has been compared with standard Bat algorithm too. From simulation it is found that dynamic pulse rate and dynamic loudness improve the performance of Bat algorithm in terms of results without being stuck at local optima and is more general.
Keywords: Global Optima, Local Optima, Heuristics, Meta-heuristic, NP-Hard, Optimization Algorithm.
Scope of the Article: VLSI Algorithms.