Fuzzy-Genetic System Applied to Topology Optimization of Cable-Trusses Modular Design

This paper demonstrates an application of a hybrid system in the designing of lightweight cable-truss structures. The optimized lightweight structure shape is determined by a discrete topology optimization process, in which the optimized solution is considered to have the lower mass and highest stiffness. The optimization process is performed by a genetic algorithm supported by fuzzy logic. Such hybridization allows the inclusion of expertise into the evolutionary search, which permits to create a primary rank that decreases the number of evaluations without losing the credibility of the solutions. The hybrid system is applied for optimization of a 2D robotic arm. The results obtained in the study case highlight the potential benefits of the considered fuzzy-genetic system over genetic algorithms. In addition, the presented hybrid system has led to structures with higher performance for similar boundary conditions, population and number of iterations.