Discrete topology optimization of planar Cable-Truss structures based on genetic algorithms

Abstract eng:
This paper demonstrates the application of Genetic Algorithms to design optimal lightweight Cable-Truss structures, which are structures composed by bars and prestressed cables and offer a high potential in robotics. The optimal lightweight structure shape is determined through a discrete topology optimization process which starts from a ground grid of nodes and interconnect them using cables or bars in order to obtain optimal results. The optimal solution is considered to have the lower mass and highest stiffness, such relation is expressed in the parameter stiffness-to-mass ratio. The objective function of the optimization problem evaluates the bending stiffness and the mass of the feasible solutions searching for the maximum stiffness-to-mass ratio. Symmetric structural response is desired once that in movable machines the majority of the structures are moving parts in which forces can assume different directions during working cycle, as a result the algorithm must find solutions with are symmetric in the axis perpendicular to the loading direction. Simulations are also presented showing comparisons between Cable-Truss and Truss structures under the same boundary conditions, population and iterations. Structural static response is computed using nonlinear finite element iterative procedure. Examples with optimized modular layout of a 2D robotic arm are shown, which presents improvement of Cable-Truss structures in comparison with Truss structures in all cases that have been simulated.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket