Drop-off Location Optimization in Hybrid CFRP/GFRP Composite Tubes Using Design of Experiments and SunFlower Optimization Algorithm.

This paper presents a novel optimization strategy using Design of Experiments and SunFlower optimization algorithm in order to achieve the better drop-off location in composites tubes used in applications to lower limb prosthesis. The main difficulty in using drop-offs is related to finding an ideal location to the drop-offs that provides higher structural performance. Furthermore, in a single structure there are a variety of possibilities for drop-off location. The statistical approach combines 4 design variables related to drop-off location and 1 categorical variable, which is responsible for providing the type of employed fiber in the tubular structure that can be hybrid manufactured with carbon (CFRP) and glass (GFRP) or not (only CFRP). Based on combinations between the design and categorical variables, numerical analyses using the Finite Element Method were carried out to provide the response variables with regard to structural behavior, such as failure index, nonlinear buckling load, mass and first natural frequency. Two different types of experiments were executed in the design of experiments, the factorial design which identified the significance and curvature of response variables. In the second experiment, the Response Surface Methodology revealed the main effects, the significance of design variables and their interactions considering only the response variables that showed significance. Finally, a multiobjective optimization strategy was elaborated to indicate the better drop-off location using the SunFlower algorithm. [ABSTRACT FROM AUTHOR]