Abstract: Particle Swarm Optimization (PSO) is a
stochastic search algorithm inspired from the natural behavior of insects
and birds. Due to its few controlling parameters and easiness in
implementations, PSO is very popular among other optimal algorithms.
However, PSO is often trapped into local optima while solving high
dimensional, complicated inverse and multimodal objective problems. To
tackle this difficulty, an improved PSO, having an adaptive, dynamic and an
improved parameter, is proposed. The adaptive and dynamic parameters will
bring balance between the exploration and exploitation search abilities
while the improved parameter controls the diversity of the population at the
final stages of the search process. The experimental results demonstrate
that the performance of the proposed PSO is better as compared to other well
designed variants.
Keywords: Adaptive and dynamic learning parameters, best particle, dynamic inertia weight, PSO, global optimization
