Crossover-Free Differential Evolution Algorithm to study the impact of Mutation Scale Factor Parameter
Dhanya M Dhanalakshmy1, G. Jeyakumar2, C. Shunmuga Velayutham3

1Dhanya M Dhanalakshmy, Department of Computer Science and Engineering, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetham, (Tamil Nadu), India.
2G. Jeyakumar, Department of Computer Science and Engineering, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetham, (Tamil Nadu), India.
3C.Shunmuga Velayutham3, Department of Computer Science and Engineering, Amrita School of Engineering, Coimbatore, Amrita Vishwa Vidyapeetham, (Tamil Nadu), India.

Manuscript received on 23 March 2019 | Revised Manuscript received on 30 March 2019 | Manuscript published on 30 March 2019 | PP: 1728-1737 | Volume-7 Issue-6, March 2019 | Retrieval Number: F2324037619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© 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: The Differential Evolution (DE) algorithm, which is one of the popular optimization algorithms in the category of Evolutionary Algorithms (EAs), is known for its simplicity and wide applicability. Analysing and understanding the working nature of DE algorithm, for its further improvement, is an active research area in Evolutionary Computing (EC) field. In particular studying the role of its control parameters and their effects in its performance needs more attention. As an attempt in this direction, this paper presents evidences to showcase the role of the Scale Factor (F) parameter of DE algorithm through the plots generated based on the studies made from experimental results obtained through a well formulated experimental setup. The experimental set up includes five different benchmarking functions and a crossover-free DE algorithm, in which the crossover component is removed, for capturing better insights about the impact of F. The empirical evidences for the observed inferences are plotted as graphs.
Keywords: Differential Evolution, Parameter Study, Mutation Scale Factor, Nature of Convergence, Premature Convergence, Successful Convergence and Stagnation
Scope of the Article: Software Engineering Case Study and Experience Reports