Optimization of truss structures by a stochastic method

¹ M3M, Université de Technologie de Belfort Montbèliard, 90000 Belfort, France
² The Key Laboratory of Contemporary Design and Integrated Manufacturing Technology, Northwestern, Polytechnical University, Xi'an, China

Corresponding author: dominique.chamoret@utbm.fr

Received: 8 July 2008
Accepted: 12 October 2008

Abstract

This paper presents an original stochastic method of optimization (PGSL) for optimum design of truss structures. PGSL (Probabilistic Global Search Lausanne, developed at the Swiss Federal Institute of Technology at Lausanne) has been inter-faced with a finite element code, then applied and tested on several size optimization problems of plane truss structures. This method uses a probability distribution function (PDF) to sample the search space and select the best structure. A PDF represents the probability of finding a good solution at any given point in the search domain. The better solutions are more likely to be found in the neighbourhood of good solutions; hence, probabilities are increased in regions where good solutions are found. The main advantage of this method is to avoid the sensitivity analysis and to converge to the global optimum.