Particle swarm optimization by using coefficient of variation for terminating scheme and it's application to electric power leveling systems

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Recently, metaheuristics becomes practical technique for various optimization problems. Metaheuristics is the technique of obtaining an approximate solution efficiently for optimization problems. Genetic Algorithm (GA), Simulated Annealing (SA), Particle Swarm Optimization (PSO), etc. are typical metaheuristics. PSO is an effective method among them in terms of the processing time and the accuracy in solution. The processing time of PSO depends on the number of the particle and the generation. The terminating condition in the conventional PSO is the preset maximum number of generations. Thus, even though the particles have already converged to the optimal solution, the search continues up to the preset one. This paper proposes a novel scheme of terminating in PSO by using the coefficient of variation (CV) which achieves the shortening of the processing time. The proposed method judges the convergence of the swarm by using CV, and terminates the search. In this paper, the validity is shown by the benchmark problems and the proposed method is applied to the optimization problem for the electric power leveling systems in the rolling mills.

Original languageEnglish
JournalIEEJ Transactions on Power and Energy
Volume133
Issue number5
DOIs
Publication statusPublished - 2013

Fingerprint

Particle swarm optimization (PSO)
Leveling (machinery)
Processing
Rolling mills
Simulated annealing
Genetic algorithms

Keywords

  • Coefficient of variation
  • Optimization
  • Particle swarm optimization
  • Terminating condition
  • Time shortening

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Energy Engineering and Power Technology

Cite this

@article{98c46828f3c6411cb1f0176bdded0663,
title = "Particle swarm optimization by using coefficient of variation for terminating scheme and it's application to electric power leveling systems",
abstract = "Recently, metaheuristics becomes practical technique for various optimization problems. Metaheuristics is the technique of obtaining an approximate solution efficiently for optimization problems. Genetic Algorithm (GA), Simulated Annealing (SA), Particle Swarm Optimization (PSO), etc. are typical metaheuristics. PSO is an effective method among them in terms of the processing time and the accuracy in solution. The processing time of PSO depends on the number of the particle and the generation. The terminating condition in the conventional PSO is the preset maximum number of generations. Thus, even though the particles have already converged to the optimal solution, the search continues up to the preset one. This paper proposes a novel scheme of terminating in PSO by using the coefficient of variation (CV) which achieves the shortening of the processing time. The proposed method judges the convergence of the swarm by using CV, and terminates the search. In this paper, the validity is shown by the benchmark problems and the proposed method is applied to the optimization problem for the electric power leveling systems in the rolling mills.",
keywords = "Coefficient of variation, Optimization, Particle swarm optimization, Terminating condition, Time shortening",
author = "Yohei Makino and Akiko Takahashi and Jun Imai and Shigeyuki Funabiki",
year = "2013",
doi = "10.1541/ieejpes.133.488",
language = "English",
volume = "133",
journal = "IEEJ Transactions on Power and Energy",
issn = "0385-4213",
publisher = "The Institute of Electrical Engineers of Japan",
number = "5",

}

TY - JOUR

T1 - Particle swarm optimization by using coefficient of variation for terminating scheme and it's application to electric power leveling systems

AU - Makino, Yohei

AU - Takahashi, Akiko

AU - Imai, Jun

AU - Funabiki, Shigeyuki

PY - 2013

Y1 - 2013

N2 - Recently, metaheuristics becomes practical technique for various optimization problems. Metaheuristics is the technique of obtaining an approximate solution efficiently for optimization problems. Genetic Algorithm (GA), Simulated Annealing (SA), Particle Swarm Optimization (PSO), etc. are typical metaheuristics. PSO is an effective method among them in terms of the processing time and the accuracy in solution. The processing time of PSO depends on the number of the particle and the generation. The terminating condition in the conventional PSO is the preset maximum number of generations. Thus, even though the particles have already converged to the optimal solution, the search continues up to the preset one. This paper proposes a novel scheme of terminating in PSO by using the coefficient of variation (CV) which achieves the shortening of the processing time. The proposed method judges the convergence of the swarm by using CV, and terminates the search. In this paper, the validity is shown by the benchmark problems and the proposed method is applied to the optimization problem for the electric power leveling systems in the rolling mills.

AB - Recently, metaheuristics becomes practical technique for various optimization problems. Metaheuristics is the technique of obtaining an approximate solution efficiently for optimization problems. Genetic Algorithm (GA), Simulated Annealing (SA), Particle Swarm Optimization (PSO), etc. are typical metaheuristics. PSO is an effective method among them in terms of the processing time and the accuracy in solution. The processing time of PSO depends on the number of the particle and the generation. The terminating condition in the conventional PSO is the preset maximum number of generations. Thus, even though the particles have already converged to the optimal solution, the search continues up to the preset one. This paper proposes a novel scheme of terminating in PSO by using the coefficient of variation (CV) which achieves the shortening of the processing time. The proposed method judges the convergence of the swarm by using CV, and terminates the search. In this paper, the validity is shown by the benchmark problems and the proposed method is applied to the optimization problem for the electric power leveling systems in the rolling mills.

KW - Coefficient of variation

KW - Optimization

KW - Particle swarm optimization

KW - Terminating condition

KW - Time shortening

UR - http://www.scopus.com/inward/record.url?scp=84877076323&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84877076323&partnerID=8YFLogxK

U2 - 10.1541/ieejpes.133.488

DO - 10.1541/ieejpes.133.488

M3 - Article

AN - SCOPUS:84877076323

VL - 133

JO - IEEJ Transactions on Power and Energy

JF - IEEJ Transactions on Power and Energy

SN - 0385-4213

IS - 5

ER -