Applying a modified particle swarm optimizer to discrete truss topology optimization

Bin Yang,Qilin Zhang
DOI: https://doi.org/10.1109/CISE.2010.5677255
2010-01-01
Abstract:Particle Swarm Optimization (PSO) is a new paradigm of Swarm Intelligence which proposes that the cooperation of individuals promotes the evolution of the swarm. Recently, a modified Particle Swarm Optimizer (MLPSO) has been succeeded in solving truss topological optimization problems with continuous design variable and competitive results were obtained. Since most of structural problems involve discrete design variables, an effect strategy is involved in MLPSO in order to operate on discrete variables. The performance of the proposed algorithm is then tested by two discrete truss topology optimization problems which belong to the most challenging fields of optimization. The proposed optimization algorithm exhibited competitive performance compared with the best results so far obtained due to it improved global searching ability. Final part is conclusion and outlook. ©2010 IEEE.
What problem does this paper attempt to address?