Analysis of Vector Evaluated Particle Swarm Optimization Guided by Non-Dominated Solutions : Inertia Weight, Cognitive, and Social Constants
Recently, an improved Vector Evaluated Particle Swarm Optimisation (VEPSO) algorithm is introduced by redefining the swarm's leader as non-dominated solutions. The improved VEPSO algorithm is named as VEPSOnds. Since a parameter tuning of a heuristic algorithm is normally difficult. Hence, in t...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | en |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/6759/1/Analysis_of_Vector_Evaluated_Particle_Swarm_Optimization_Guided_by_Non-.PDF http://umpir.ump.edu.my/id/eprint/6759/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Recently, an improved Vector Evaluated Particle Swarm Optimisation (VEPSO) algorithm is introduced by redefining the swarm's leader as non-dominated solutions. The improved VEPSO algorithm is named as VEPSOnds. Since a parameter tuning of a heuristic algorithm is normally difficult. Hence, in this paper, three important parameters of the improved VEPSO, which are inertia weight, cognitive constant, and social constant, are analyzed. The results suggest that the inertia weight should gradually degrade from 1.0 to 0.4, and both cognitive and social constants to be random value in between 1.5 and 2.5. |
|---|