STRUCTURAL OPTIMIZATION OF THERMOMECHANICAL PROPERTIES FOR SHORT-FIBER-REINFORCED COMPOSITES

Composite materials are heterogeneous materials on the microscopic level. For short-fiber reinforced composites, the fiber-matrix and fiber-fiber interactions play a significant role in the strength and global stiffness of the material. A unit cell model is developed to study the influence of fiber clustering patterns on the local stress distribution and global composite properties. Design variables are established to account for the variations in fiber clustering patterns, aspect ratio, volume fraction, and fiber packing. Design constraints are used to represent composite moduli, thermal coefficients of expansion, maximum interfacial stress and thermal residual stress. The model is analyzed by the p-version finite element method. The study is limited to elastic analysis. The p-version finite element method offers rapid convergence on localized stress fields using only coarse mesh and allows geometric shape variations without remeshing. Solution accuracy can be checked systematically using a series of solutions with increasing numerical complexity that is generated from a given mesh. A parametric study of the effect of fiber duster patterns, fiber packing, and fiber aspect ratio is conducted. The study shows that the fiber cluster pattern and fiber packing has a significant impact on all of the composite properties studied.