Design optimization of thin-walled composite structures based on material and fiber orientation

Abstract

This paper presents the design optimization of thin-walled composite laminate structures to maximize stiffness at minimum material cost. To this end, a computational design optimization framework for advanced composites has been developed based on a novel Decoupled Discrete Material Optimization. The framework uses non-convex and convex sequential approximation optimization. The design variables are the piecewise patch orientations and fiber materials. The performance of the proposed optimization framework is analyzed to test the convergence and the radius filter effect. The results demonstrate that the proposed framework is much more robust than the conventional Discrete Material Optimization. The framework is used to find the optimal design of cantilevered composite beams with different geometries, boundary conditions and load cases. A comparison of different fiber and material layouts is also investigated for a multiple load case. The results suggest that a suitable solution presents a grid patch design with 25% CFRP material that exhibits a significant structural improvement at a minimum additional cost. Despite having the maximum material cost, the beam composed of fully of carbon glass fiber reinforced polymer in annular design exhibits very close structural performance to the design with the minimum compliance.