Study on inertia weight decreasing strategy of particle swarm optimization based on inverse coseca

The standard particle swarm optimization (pso), which introduces inertia weight w, is an effective method to find the extreme value of the function. However, particle swarm optimization (pso) has some disadvantages. When dealing with optimization problems, pso lacks effective parameter control and i...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of physics. Conference series Jg. 1486; H. 7; S. 72004 - 72009
Hauptverfasser: Liu, Huiming, Liu, Yong, Zhen, Jiaqi
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Bristol IOP Publishing 01.04.2020
Schlagworte:
Algorithms
Convergence
Extreme values
Inertia
Local optimization
Machine learning
Optimization
Particle swarm optimization
Physics
Probability distribution functions
Random numbers
Weight
ISSN:1742-6588, 1742-6596
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The standard particle swarm optimization (pso), which introduces inertia weight w, is an effective method to find the extreme value of the function. However, particle swarm optimization (pso) has some disadvantages. When dealing with optimization problems, pso lacks effective parameter control and is prone to fall into local optimization, which leads to low convergence accuracy. In, this paper, put forward a new improved particle swarm optimization (pso) algorithm, The nonlinear decreasing inertia weight by the CSC function strategy, at the same time to join the beta distribution on random Numbers, thus to balance the global search and local search ability of the algorithm. The learning factor is the changed asynchronously to improve the learning ability of the algorithm. By adopting Griewank, Rastrigrin, J.D. Schaffer three standard test functions to simulate experiment, at the same time and the basic particle swarm algorithm the inertia weight in a fixed value, the linear regressive LDIW and nonlinear regressive comparison. The experimental results show that the nonlinear decreasing strategy with dynamic adjustment of inverse cosecant function can improve the convergence speed and stability.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1742-6588
1742-6596
DOI:10.1088/1742-6596/1486/7/072004