TY - GEN
T1 - Chaotic co-evolutionary algorithm based on differential evolution and particle swarm optimization
AU - Zhang, Meng
AU - Zhang, Weiguo
AU - Sun, Yong
PY - 2009
Y1 - 2009
N2 - A chaotic co-evolutionary algorithm based on differential evolution and particle swarm optimization (CCDEPSO) is proposed. CCDEPSO is a two-population co-evolutionary algorithm, in which the individuals of one population are evolved according to differential evolution algorithm (DE) and the other individuals are evolved according to particle swarm optimization algorithm (PSO). In order to realize co-evolving of the two sub-populations, an information sharing scheme by sharing the fitness value and the corresponding position value of the two sub-populations is introduced. Furthermore, in order to improve the searching speed and avoid the results trapping in local optima prematurely, chaotic initialization and chaotic perturbation based on Tent map are introduced in CCDEPSO. The comparative testing experiments are performed by means of CCDEPSO, DE, CPSO and PSO algorithms on six benchmark functions. Experimental results demonstrate that the global optimization ability of CCDEPSO is better than other algorithms above mentioned.
AB - A chaotic co-evolutionary algorithm based on differential evolution and particle swarm optimization (CCDEPSO) is proposed. CCDEPSO is a two-population co-evolutionary algorithm, in which the individuals of one population are evolved according to differential evolution algorithm (DE) and the other individuals are evolved according to particle swarm optimization algorithm (PSO). In order to realize co-evolving of the two sub-populations, an information sharing scheme by sharing the fitness value and the corresponding position value of the two sub-populations is introduced. Furthermore, in order to improve the searching speed and avoid the results trapping in local optima prematurely, chaotic initialization and chaotic perturbation based on Tent map are introduced in CCDEPSO. The comparative testing experiments are performed by means of CCDEPSO, DE, CPSO and PSO algorithms on six benchmark functions. Experimental results demonstrate that the global optimization ability of CCDEPSO is better than other algorithms above mentioned.
KW - Co-evolutionary
KW - Differential evolution
KW - Information sharing
KW - Particle swarm optimization
KW - Tent map
UR - http://www.scopus.com/inward/record.url?scp=70450176952&partnerID=8YFLogxK
U2 - 10.1109/ICAL.2009.5262798
DO - 10.1109/ICAL.2009.5262798
M3 - 会议稿件
AN - SCOPUS:70450176952
SN - 9781424447954
T3 - Proceedings of the 2009 IEEE International Conference on Automation and Logistics, ICAL 2009
SP - 885
EP - 889
BT - Proceedings of the 2009 IEEE International Conference on Automation and Logistics, ICAL 2009
T2 - 2009 IEEE International Conference on Automation and Logistics, ICAL 2009
Y2 - 5 August 2009 through 7 August 2009
ER -