Title :
Introduce a New Inertia Weight for Particle Swarm Optimization
Author :
Ememipour, Jafar ; Nejad, M. Mehdi Seyed ; Ebadzadeh, M. Mehdi ; Rezanejad, Javad
Author_Institution :
Dept. of Comput. & IT Eng., Azad Univ. of Qazvin, Qazvinjran, China
Abstract :
PSO has a few parameters to adjust such as inertia weight, velocity and constant factors. Among these parameters, inertia weight is very important and has a great potential to develop. During the last decade, various methods like fuzzy, constant, linear methods were proposed to adjust an inertia weight. This paper proposes a new strategy to calculate inertia weight based on decreasing exponential method. Our method merely uses an iteration to make an inertia weight and it is fast and has highly accurate results rather than other strategies. Our results are tested on well-known benchmarks. Numerical results demonstrated our claim.
Keywords :
particle swarm optimisation; PSO; decreasing exponential method; inertia weight; particle swarm optimization; Benchmark testing; Birds; Educational institutions; Genetic mutations; Information technology; Java; Marine animals; Particle swarm optimization; Random number generation; inertia weight; optimization; particle swarm optimization;
Conference_Titel :
Computer Sciences and Convergence Information Technology, 2009. ICCIT '09. Fourth International Conference on
Conference_Location :
Seoul
Print_ISBN :
978-1-4244-5244-6
Electronic_ISBN :
978-0-7695-3896-9
DOI :
10.1109/ICCIT.2009.297