TY - GEN
T1 - DE-FPA
T2 - 2014 International Conference on High Performance Computing and Applications, ICHPCA 2014
AU - Chakraborty, Dwaipayan
AU - Saha, Sankhadip
AU - Dutta, Oindrilla
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2015/2/18
Y1 - 2015/2/18
N2 - In this paper, a new hybrid population based algorithm (DE-FPA) is proposed with the combination of differential evolution optimization algorithm and flower pollination algorithm. The main idea is to integrate the natural evolution characteristics of the population in differential evolution algorithm with the pollination behavior of flowering plant in flower pollination algorithm to synthesize the strength and power of both the algorithms. The hybrid algorithm is robust in the sense that the globalization takes place in evolution. Some benchmark test functions are utilized here to compare the hybrid algorithm with the individual DE and FPA algorithms in searching the best solution. The results show the hybrid algorithm possesses a better capability in searching for the sufficiently good solution and to escape from local optima. In addition to that, a novel concept of dynamic adaptive weight is introduced for faster convergence than the individual algorithms, thereby making the hybrid one competent.
AB - In this paper, a new hybrid population based algorithm (DE-FPA) is proposed with the combination of differential evolution optimization algorithm and flower pollination algorithm. The main idea is to integrate the natural evolution characteristics of the population in differential evolution algorithm with the pollination behavior of flowering plant in flower pollination algorithm to synthesize the strength and power of both the algorithms. The hybrid algorithm is robust in the sense that the globalization takes place in evolution. Some benchmark test functions are utilized here to compare the hybrid algorithm with the individual DE and FPA algorithms in searching the best solution. The results show the hybrid algorithm possesses a better capability in searching for the sufficiently good solution and to escape from local optima. In addition to that, a novel concept of dynamic adaptive weight is introduced for faster convergence than the individual algorithms, thereby making the hybrid one competent.
UR - http://www.scopus.com/inward/record.url?scp=84925424511&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84925424511&partnerID=8YFLogxK
U2 - 10.1109/ICHPCA.2014.7045350
DO - 10.1109/ICHPCA.2014.7045350
M3 - Conference contribution
AN - SCOPUS:84925424511
T3 - 2014 International Conference on High Performance Computing and Applications, ICHPCA 2014
BT - 2014 International Conference on High Performance Computing and Applications, ICHPCA 2014
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 22 December 2014 through 24 December 2014
ER -