An Adaptive Mutation Strategy for Real Coded Genetic Algorithm. | [Sindh University Research Journal • 2017]

Author(s):

1. I. A. KOREJO: Institue of Mathematics and Computer Science, University of Sindh, Jamshoro, Pakistan

2. K. BROHI: Institue of Mathematics and Computer Science, University of Sindh, Jamshoro, Pakistan

3. A. A. CHANDIO: Institue of Mathematics and Computer Science, University of Sindh, Jamshoro, Pakistan

4. F. A. MEMON: Institue of Mathematics and Computer Science, University of Sindh, Jamshoro, Pakistan

5. A. R. NANGRAJ: Institue of Mathematics and Computer Science, University of Sindh, Jamshoro, Pakistan

Abstract:

Generally, single mutation operator is used by the conventional genetic algorithms (GAs), so it determine that all individuals in the population apply the same mechanism. Due to this mechanism, simple gas suffer from the premature convergence problem, it is easy to be trapped in to local optima.This paper proposes adaptive mutation (AMGAs) scheme for GAs to automatically select the most suitable operator while solving the problem.It also proposes different mutation strategies to deal with different situations in the search space by using amgas. To analysis the performance of propose approach, twnenty well known and applied benchmark problems are used. The experimental results show that amgas greatly balance the performance of GAs on different set problems.

Page(s): 255-260

DOI: DOI not available

Published: Journal: Sindh University Research Journal, Volume: 49, Issue: 2, Year: 2017

Keywords:

operator adaptation , Genetic algorithms

References:

References are not available for this document.

Citations

Citations are not available for this document.

Citations

Downloads

Views