A Modified Differential Evolution Algorithm and Its Application to Engineering Problems

Author

Ali, Musrrat ; Pant, Millie ; Abraham, Ajith

Author_Institution

Dept. of Paper Technol., Indian Inst. of Technol. Roorkee, Saharanpur, India

fYear

2009

fDate

4-7 Dec. 2009

Firstpage

196

Lastpage

201

Abstract

In the present study a Modified Differential Evolution (MDE) algorithm is proposed. This algorithm is different in three ways from basic DE. For initialization it utilizes opposition-based learning while in basic DE uniform random numbers serve this task. Secondly, in basic DE mutant individual is random while in MDE it is tournament best and finally MDE utilizes only one set of population as against two sets as used in basic DE. The performance of proposed algorithm is investigated and compared with basic differential evolution. The experiments conducted shows that proposed algorithm outperform the basic DE algorithm in all the benchmark problems and real life applications.

Keywords

learning (artificial intelligence); optimisation; DE; modified differential evolution algorithm; mutant individual; opposition-based learning; uniform random numbers; Computer applications; Computer industry; Constraint optimization; Containers; Design optimization; Integer linear programming; Laboratories; Pattern recognition; Printing; Testing; differential evolution; mutation operator; opposition based learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Soft Computing and Pattern Recognition, 2009. SOCPAR '09. International Conference of

Conference_Location

Malacca

Print_ISBN

978-1-4244-5330-6

Electronic_ISBN

978-0-7695-3879-2

Type

conf

DOI

10.1109/SoCPaR.2009.48

Filename

5370090