Title :
A Novel Enhanced Particle Swarm Optimization Method with Diversity and Neighborhood Search
Author :
Dang Cong Tran ; Zhijian Wu ; Hui Wang
Author_Institution :
State Key Lab. of Software Eng., Wuhan Univ., Wuhan, China
Abstract :
By learning from different particle in local neighborhood and global neighborhood Particle Swarm Optimization (PSO) algorithm achieved a trade-off between exploration and exploitation abilities. In this paper, we propose a new approach by combining diversity mechanism and neighborhood search strategies, called a novel enhanced PSO method with Diversity and Neighborhood Search (EPSODNS). In this paper we propose a new local neighborhood search strategy that promotes the exploitation ability of algorithm. Our experiments are conducted on test benchmarks include 13 well-known numerical benchmarks. The results show that EPSODNS obtains a better majority of the test problems.
Keywords :
particle swarm optimisation; search problems; EPSODNS; PSO algorithm; diversity mechanism; global neighborhood; local neighborhood search strategy; particle swarm optimization; Benchmark testing; Convergence; Diversity reception; Search problems; Sociology; Statistics; Topology; diversity; global optimization; neighborhood search; particle swarm optimization;
Conference_Titel :
Systems, Man, and Cybernetics (SMC), 2013 IEEE International Conference on
Conference_Location :
Manchester
DOI :
10.1109/SMC.2013.38