Multi-objective optimization of nonlinear inelastic steel trusses with continuous design variables using NSGA-II

Abstract

Thank to the rapid development of computer technology, optimization design has been increasingly widely applied in all areas of life. In structural optimization, due to the variety and even contradiction between the desired goals of design engineers, multi-objective optimization attracts more and more attention from scientists. In this paper, the author presents the multi-objective optimization design problem for steel truss with two basic objectives: the truss mass and the horizontal displacement of the top of the truss is minimized. The design variables are the cross-sectional areas of the structure that are in the defined ranges. Direct analysis is used to capture the nonlinear behavior as well as calculate the maximum load-carrying capacity of the structure. The optimization problem set up above is solved by NSGA-II (Non-dominated sorting GA-II) algorithm. A 47-bar planar truss is studied to illustrate the optimal design problem. The results show that the NSGA-II algorithm is very effective in solving the given optimization problem.