Optimum design of truss structures by Material Generation Algorithm with discrete variables

Abstract

Optimization is an act of decision-making for reaching a point in which the overall behaviour of the considered system is acceptable by the experts. In this paper, the optimum design of truss structures is considered utilizing the Material Generation Algorithm (MGA) as one of the recently developed metaheuristic algorithms in which the basic and advanced principles of chemistry have been used as an inspirational concept. For numerical investigations, the benchmark problems in the truss optimization field, including the 25-bar, 72-bar and 200-bar truss structures, are used while a penalty method is utilized accordingly for constraint handling purposes. Multiple optimization runs are conducted for statistical purposes, and the results are compared to the results of other metaheuristic algorithms. The obtained results proved that the MGA could provide very acceptable and optimal design sections for the considered problems, which results in the lowest possible weight.