Particle swarm method as a new tool for structural optimization

Jármai, Károly (2005) Particle swarm method as a new tool for structural optimization. JOURNAL OF COMPUTATIONAL AND APPLIED MECHANICS, 6 (2). pp. 237-256. ISSN 1586-2070

	Text Jarmai_JCAM_2005.pdf Restricted to Repository staff only Download (2MB) | Request a copy
Preview	Text 6-2-JarmaiK-V6N2.pdf - Published Version Download (248kB) | Preview

Abstract

A new and promising optimization technique is introduced: the particle swarm optimization (PSO). In this evolutionary technique the social behavior of birds is imitated. The technique is modified in order to be efficient in technical applications. It calculates discrete optima, uses dynamic inertia reduction and craziness at some particles. The efficiency of the technique is shown in the optimum design of a stringer-stiffened shell under bending and compression. The PSO is built into an interactive program system, where several optimization techniques are employed. The program system includes multiobjective optimization techniques as well. Results show that PSO is a reliable and robust technique to find optima with highly non-linear constraints. In the cost calculation 2D and 3D curve fitting is employed to determine the production time.

Actions (login required)

Edit Item