Optimization of steel frame mass using direct analysis and self-adaptive pbest differential evolution algorithm

Abstract

Optimum design of construction works in general and steel frame structures in particular has attracted increasing attention, thanks to the strong development of computer science as well as the increasing cost of construction materials. For steel frame structures, the total cost/volume of the building is usually minimized based on the design variables that are the typical cross sections of the components. Besides, the behavior of the steel frame is non-linear and inelastic, so the optimization problem of the steel frame is highly non-linear. On that basis, in this paper, a meta-heuristic optimization algorithm based on DE differential evolution algorithm aims to optimize the steel frame mass using direct analysis technique. Two algorithms Jaya and swarm optimization (PSO) are used to evaluate the efficiency of the algorithm built through an example of a 3-span x 10-storey steel frame. The calculation results show that the proposed method has found the optimal solution better and uses rather much less number of structural analysis times than Jaya and PSO algorithms.

Key words: Steel frame; direct analysis; Meta- heuristic; differential evolution