A multi-scale framework for layered composites with thermo-rheologically complex behaviors

A multi-scale modeling framework is proposed for generating the effective nonlinear thermo-viscoelastic responses of multi-layered and thick-section composites. The modeling framework is demonstrated for multi-layered composite systems consisting of two alternating fiber-reinforced polymeric layers: unidirectional fiber (roving) and continuous filament mat (CFM). The viscoelastic behavior of the polymeric matrix is considered as stress-dependent and thermo-rheologically complex. Two simplified 3D micromechanical models with periodic boundary conditions are formulated for the roving and CFM. In addition, a sublaminate model is developed for the 3D effective homogenized through-thickness response of the roving and CFM layers. This framework provides an effective time-dependent material response while simultaneously recognizing the corresponding deformation at the microconstituents under overall thermo-mechanical loadings at the macro-level. Stress correction algorithms are developed at each scale to enhance computational efficiency and accuracy. Short-term (30 min) creep tests on off-axis multi-layered specimens are also conducted under combined stresses and temperatures to calibrate in-situ fiber and matrix properties and verify the predictions of the multi-scale framework. A time-shifting method is applied to create long-term material behaviors from the available short-term creep data. Verification of the multi-scale material framework is also done for the overall long-term responses. Finally, long-term structural analyses under thermo-mechanical loadings are conducted by integrating the multi-scale material framework to finite element (FE) structural analyses.