Year: 2015
Type: conference-paper
Abstract: Particle Swarm Optimization (PSO) is a population based optimal method and very simple in both theory and numerical implementation. Nowadays, PSO has been recognized as a paradigm for numerical optimizations; however, PSO is easily trapped into a local optimum when solving multidimensional and complex problems. In order to overcome this difficulty, this paper presents a modified PSO with a dynamic inertia weight and an adaptive mutation operator. To verify the proposed PSO, we test it numerically on a set of well known bench mark functions as well as on an engineering problem, as to which it has shown better performance and efficiency while compared to the basic PSO and Beta PSO.
DOI: 10.22323/1.259.0004
Author: Shiyou Yang
Institution: Zhejiang University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W2564281898