Modified Mutation in Asynchronous Differential Evolution

Modified Mutation in Asynchronous Differential Evolution

Vaishali (Amity University Rajasthan, Jaipur, India) and Tarun K. Sharma (Amity University Rajasthan, Jaipur, India)
Copyright: © 2018 |Pages: 12
DOI: 10.4018/IJAEC.2018010103

Abstract

This article presents a variant of an asynchronous differential evolution (ADE) algorithm for solving global optimization problems. The proposed algorithm uses the mean of two randomly chosen variables as a third variable to perform a mutation operation. The modification in a mutation operation is done to exploit the chosen random variables and to accelerate the convergence. The proposed algorithm is tested on a set of benchmark functions and compared with differential evolution (DE) and ADE. Results and comparisons show that the proposed work outperforms other algorithms in terms of number of function evaluation, standard deviation and convergence rate. Results are validated through non-parametric statistical analysis.
Article Preview
Top

Introduction

Evolutionary and population based metaheuristic algorithms have become very popular from last two decades. And these algorithms have proved their efficiency over other algorithms in finding the optimal solution more quickly (Back, 1996). These metaheuristic algorithms and their variants (Rajpurohit et al. 2017) have been used to optimize various benchmark functions, engineering design problems and real-life problems (Gupta et al., 2017; Kumar et al., 2013; Sharma & Pant, 2017; Sharma & Pant, 2017; Sharma & Pant, 2013; Sharma et al., 2013). In this paper Asynchronous differential evolution (ADE) algorithm has been used which supports asynchronous strategy to solve optimization problems (Zhabitskaya & Zhabitsky, 2012). ADE has been derived from Differential Evolution (DE). DE, a stochastic population based optimization algorithm was introduced by (Storn & Price,1995). DE is a generation based evolution strategy in which selection, mutation and crossover operations are performed on the population synchronously (Storn & Price,1997). Over the time DE has been proved its efficient in solving global and real-time optimization problems. Because DE is simple yet robust, it has been used efficiently in various fields of engineering like communication (Storn, 1996), mechanical engineering (Rogalsky et al., 2000), pattern recognition (Ilonen et al.,2003) and other fields (Dehmollaian, 2011; Gao et al., 2014; Gao et al., 2016; Monakhov et al., 2016).

ADE is well suited for parallelization and global optimization problems (Zhabitskaya & Zhabitsky, 2012). ADE also performs selection, mutation and crossover operations but asynchronously. ADE is not generation based, it iteratively improves the population. Unlike DE, the better vectors become part of the population as soon as they are found.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2020): 2 Released, 2 Forthcoming
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing