Answers are grounded in this paper and cite their sources.
The ultimate success of particle swarm optimization depends on the velocity values of previous particles. Velocity is multiplied with inertia weight coefficient, and has a significant effect on search capability of the particle swarm optimization. When looking at previous studies that are carried out to calculate this coefficient, it is seen that inertia weight coefficient has been handled in several ways. In this article; a new ensemble inertia weight calculation strategy is proposed that uses other constant, random, linear decreasing, global local best, simulated annealing and chaotic inertia weight calculation methods. Other methods results are combined and used to make a final output decision in a proper way. In experimental tests, 30 common optimization benchmark test problems are used. Proposed ensemble strategy is proven by statistical tests and gives successful results in all optimization benchmark test problems.
No takes yet. Share an insight, caveat, or question.
İbrahim Berkan Aydilek (2018) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: