Multiple spatio-temporal scale modeling of composites subjected to cyclic loading

This manuscript presents a multiscale modeling methodology for failure analysis of composites subjected to cyclic loading conditions. Computational homogenization theory with multiple spatial and temporal scales is employed to devise the proposed methodology. Multiple spatial scales address the disparity between the length scale of material heterogeneities and the overall structure, whereas multiple temporal scales with almost periodic fields address the disparity between the load period and overall life under cyclic loading. The computational complexity of the multiscale modeling approach is reduced by employing a meso-mechanical model based on eigendeformation based homogenization with symmetric coefficients in the space domain, and an adaptive time stepping strategy based on a quadratic multistep method with error control in the time domain. The proposed methodology is employed to simulate the response of graphite fiber-reinforced epoxy composites. Model parameters are calibrated using a suite of experiments conducted on unidirectionally reinforced specimens subjected to monotonic and cyclic loading. The calibrated model is employed to predict damage progression in quasi-isotropic specimens. The capabilities of the model are validated using acoustic emission testing.