Journal of Multiscale Modelling

In this study, a novel fluidic concept was presented to resemble the cerebral microvascular in four types to assess its complexity by using centrifugal platform. The setup consisted of a microstructured disk with a round mixing chamber rotating on a macroscopic drive unit. The left and right internal carotid arteries (L.ICA and R.ICA) and basilar artery (BA) are two isolated vascular system supplying circle of Willis (CoW). The left and right middle cerebral arteries (L.MCA and R.MCA), left and right anterior cerebral arteries (L.ACA and R.ACA), and finally left and right posterior cerebral arteries (L.PCA and R.PCA) constitute efferent arteries of CoW. In this study, cerebral microvascular was investigated by microfluidics approach. The results revealed that a more complex mixing chamber provides normal pixel percentage distribution with respect to the other ones. The outcomes of this study may have implications not only for perception of the intracranial vascular hemodynamic in healthy circumstance, but also for diagnosing the diseases in the blood circulatory system of the human body.

A numerical homogenization scheme is presented to model glass particle-toughened polymer materials under dynamic loading. A constitutive law is developed for the polymer material and validated by comparing the results to experimental test data. A similar constitutive law as that of the polymer material with unknown material parameters is assumed for the glass particle-toughened polymer. A homogenization scheme is used to determine the unknown material parameters from the boundary value problem (BVP) of a representative volume element. Unlike the standard computational homogenization scheme, the proposed numerical homogenization scheme can be used after localization occurs in the material. The proposed multiscale model is then verified against direct numerical simulation.

A multi-scale micromechanics approach along with finite element (FE) model predictive tool is developed to analyze low-energy-impact damage footprint and compression-after-impact (CAI) of composite laminates which is also tested and verified with experimental data. Effective fiber and matrix properties were reverse-engineered from lamina properties using an optimization algorithm and used to assess damage at the micro-level during impact and post-impact FE simulations. Progressive failure dynamic analysis (PFDA) was performed for a two step-process simulation. Damage mechanisms at the micro-level were continuously evaluated during the analyses. Contribution of each failure mode was tracked during the simulations and damage and delamination footprint size and shape were predicted to understand when, where and why failure occurred during both impact and CAI events. The composite laminate was manufactured by the vacuum infusion of the aero-grade toughened Benzoxazine system into the fabric preform. Delamination footprint was measured using C-scan data from the impacted panels and compared with the predicated values obtained from proposed multi-scale micromechanics coupled with FE analysis. Furthermore, the residual strength was predicted from the load-displacement curve and compared with the experimental values as well.