The limited mutation particle swarm optimizer

Authors:
Chunhe Song;Hai Zhao;Wei Cai;Haohua Zhang;Ming Zhao
Affiliations:
College of Information Science and Engineering, Northeastern University, Shenyang, China;College of Information Science and Engineering, Northeastern University, Shenyang, China;College of Information Science and Engineering, Northeastern University, Shenyang, China;College of Information Science and Engineering, Northeastern University, Shenyang, China;College of Information Science and Engineering, Northeastern University, Shenyang, China
Venue:
LSMS'07 Proceedings of the Life system modeling and simulation 2007 international conference on Bio-Inspired computational intelligence and applications
Year:
2007

Citing 3
Cited 0

The gregarious particle swarm optimizer (G-PSO)

Proceedings of the 8th annual conference on Genetic and evolutionary computation
Dissipative particle swarm optimization

CEC '02 Proceedings of the Evolutionary Computation on 2002. CEC '02. Proceedings of the 2002 Congress - Volume 02
Self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Similar with other swarm algorithms, the PSO algorithm also suffers from premature convergence. Mutation is a widely used strategy in the PSO algorithm to overcome the premature convergence. This paper discusses some induction patterns of mutation (IPM) and typical algorithms, and then presents a new PSO algorithm - the Limited Mutation PSO algorithm. Basing on a special PSO model depicted as "social-only", the LMPSO adopts a new mutation strategy - limited mutation. When the distance between one particle and the global best location is less than a threshold predefined, some dimensions of the particles will mutate under specific rules. The LMPSO is compared to other five different types of PSO with mutation strategy, and the experiment results show that the new algorithm performances better on a four-function test suite with different dimensions.