DocumentCode :
527329
Title :
A new hybrid evolutionary algorithm with quasi-simplex technique
Author :
Zhang, Guo-li ; Lu, Hai-yan ; Zhang, Guang-quan
Author_Institution :
Dept. of Math. & Phys., North China Electr. Power Univ., Baoding, China
Volume :
4
fYear :
2010
fDate :
11-14 July 2010
Firstpage :
1811
Lastpage :
1816
Abstract :
This paper proposes a new parallel search algorithm using an evolutionary algorithm and quasi-simplex techniques (EAQST) for non-linear constrained function optimization. EAQST produces the offspring in parallel by using the Gaussian mutation, the Cauchy mutation and the quasi-simplex technique. Experimental studies on typical benchmark functions have shown that EAQST has very better performance than the compared algorithm.
Keywords :
evolutionary computation; Cauchy mutation; EAQST; Gaussian mutation; hybrid evolutionary algorithm; non-linear constrained function optimization; parallel search algorithm; quasi-simplex technique; Algorithm design and analysis; Cybernetics; Evolutionary computation; Machine learning; Machine learning algorithms; Optimization; Reflection; Cauchy mutation; Evolutionary algorithm; Gaussian mutation; Quasi-simplex;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics (ICMLC), 2010 International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-1-4244-6526-2
Type :
conf
DOI :
10.1109/ICMLC.2010.5580962
Filename :
5580962
Link To Document :
بازگشت