Enhanced particle swarm optimizer incorporating a weighted particle

Nai Jen Li, Wen June Wang, Chen-Chien James Hsu, Wei Chang, Hao Gong Chou, Jun Wei Chang

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

This study proposes an enhanced particle swarm optimizer incorporating a weighted particle (EPSOWP) to improve the evolutionary performance for a set of benchmark functions. In conventional particle swarm optimizer (PSO), there are two principal forces to guide the moving direction of each particle. However, if the current particle lies too close to either the personal best particle or the global best particle, the velocity is mainly updated by only one term. As a result, search step becomes smaller and the optimization of the swarm is likely to be trapped into a local optimum. To address this problem, we define a weighted particle for incorporation into the particle swarm optimization. Because the weighted particle has a better opportunity getting closer to the optimal solution than the global best particle during the evolution, the EPSOWP is capable of guiding the swarm to a better direction to search the optimal solution. Simulation results show the effectiveness of the EPSOWP, which outperforms various evolutionary algorithms on a selected set of benchmark functions. Furthermore, the proposed EPSOWP is applied to controller design and parameter identification for an inverted pendulum system as well as parameter learning of neural network for function approximation to show its viability to solve practical design problems.

Original languageEnglish
Pages (from-to)218-227
Number of pages10
JournalNeurocomputing
Volume124
DOIs
Publication statusPublished - 2014 Jan 26

Fingerprint

Benchmarking
Learning
Pendulums
Evolutionary algorithms
Particle swarm optimization (PSO)
Identification (control systems)
Neural networks
Controllers
Direction compound

Keywords

  • Convergence
  • Inverted pendulum system
  • Neural network
  • PID controller design
  • Particle swarm optimization (PSO)
  • Weighted particle

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Enhanced particle swarm optimizer incorporating a weighted particle. / Li, Nai Jen; Wang, Wen June; Hsu, Chen-Chien James; Chang, Wei; Chou, Hao Gong; Chang, Jun Wei.

In: Neurocomputing, Vol. 124, 26.01.2014, p. 218-227.

Research output: Contribution to journalArticle

Li, Nai Jen ; Wang, Wen June ; Hsu, Chen-Chien James ; Chang, Wei ; Chou, Hao Gong ; Chang, Jun Wei. / Enhanced particle swarm optimizer incorporating a weighted particle. In: Neurocomputing. 2014 ; Vol. 124. pp. 218-227.
@article{749fcffbcb084cd8b47aaec096cde352,
title = "Enhanced particle swarm optimizer incorporating a weighted particle",
abstract = "This study proposes an enhanced particle swarm optimizer incorporating a weighted particle (EPSOWP) to improve the evolutionary performance for a set of benchmark functions. In conventional particle swarm optimizer (PSO), there are two principal forces to guide the moving direction of each particle. However, if the current particle lies too close to either the personal best particle or the global best particle, the velocity is mainly updated by only one term. As a result, search step becomes smaller and the optimization of the swarm is likely to be trapped into a local optimum. To address this problem, we define a weighted particle for incorporation into the particle swarm optimization. Because the weighted particle has a better opportunity getting closer to the optimal solution than the global best particle during the evolution, the EPSOWP is capable of guiding the swarm to a better direction to search the optimal solution. Simulation results show the effectiveness of the EPSOWP, which outperforms various evolutionary algorithms on a selected set of benchmark functions. Furthermore, the proposed EPSOWP is applied to controller design and parameter identification for an inverted pendulum system as well as parameter learning of neural network for function approximation to show its viability to solve practical design problems.",
keywords = "Convergence, Inverted pendulum system, Neural network, PID controller design, Particle swarm optimization (PSO), Weighted particle",
author = "Li, {Nai Jen} and Wang, {Wen June} and Hsu, {Chen-Chien James} and Wei Chang and Chou, {Hao Gong} and Chang, {Jun Wei}",
year = "2014",
month = "1",
day = "26",
doi = "10.1016/j.neucom.2013.07.005",
language = "English",
volume = "124",
pages = "218--227",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Enhanced particle swarm optimizer incorporating a weighted particle

AU - Li, Nai Jen

AU - Wang, Wen June

AU - Hsu, Chen-Chien James

AU - Chang, Wei

AU - Chou, Hao Gong

AU - Chang, Jun Wei

PY - 2014/1/26

Y1 - 2014/1/26

N2 - This study proposes an enhanced particle swarm optimizer incorporating a weighted particle (EPSOWP) to improve the evolutionary performance for a set of benchmark functions. In conventional particle swarm optimizer (PSO), there are two principal forces to guide the moving direction of each particle. However, if the current particle lies too close to either the personal best particle or the global best particle, the velocity is mainly updated by only one term. As a result, search step becomes smaller and the optimization of the swarm is likely to be trapped into a local optimum. To address this problem, we define a weighted particle for incorporation into the particle swarm optimization. Because the weighted particle has a better opportunity getting closer to the optimal solution than the global best particle during the evolution, the EPSOWP is capable of guiding the swarm to a better direction to search the optimal solution. Simulation results show the effectiveness of the EPSOWP, which outperforms various evolutionary algorithms on a selected set of benchmark functions. Furthermore, the proposed EPSOWP is applied to controller design and parameter identification for an inverted pendulum system as well as parameter learning of neural network for function approximation to show its viability to solve practical design problems.

AB - This study proposes an enhanced particle swarm optimizer incorporating a weighted particle (EPSOWP) to improve the evolutionary performance for a set of benchmark functions. In conventional particle swarm optimizer (PSO), there are two principal forces to guide the moving direction of each particle. However, if the current particle lies too close to either the personal best particle or the global best particle, the velocity is mainly updated by only one term. As a result, search step becomes smaller and the optimization of the swarm is likely to be trapped into a local optimum. To address this problem, we define a weighted particle for incorporation into the particle swarm optimization. Because the weighted particle has a better opportunity getting closer to the optimal solution than the global best particle during the evolution, the EPSOWP is capable of guiding the swarm to a better direction to search the optimal solution. Simulation results show the effectiveness of the EPSOWP, which outperforms various evolutionary algorithms on a selected set of benchmark functions. Furthermore, the proposed EPSOWP is applied to controller design and parameter identification for an inverted pendulum system as well as parameter learning of neural network for function approximation to show its viability to solve practical design problems.

KW - Convergence

KW - Inverted pendulum system

KW - Neural network

KW - PID controller design

KW - Particle swarm optimization (PSO)

KW - Weighted particle

UR - http://www.scopus.com/inward/record.url?scp=84885853221&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885853221&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2013.07.005

DO - 10.1016/j.neucom.2013.07.005

M3 - Article

AN - SCOPUS:84885853221

VL - 124

SP - 218

EP - 227

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -