A Reliable Fracture Angle Determination Algorithm for Extended Puck’s 3D Inter-Fiber Failure Criterion for Unidirectional Composites

Determination of the fracture angle and maximum exposure value of extended Puck’s 3D inter-fiber failure (IFF) criterion is of great importance for predicting the failure mechanism of unidirectional fiber-reinforced composites. In this paper, a reliable semi-analytical algorithm (RSAA) is presented for searching fracture angle and corresponding exposure value for the extended Puck’s failure criterion. One hundred million cases are tested for verifying the accuracy of the present and other algorithms on Python using the strength-value-stress-state combinations more universal than those in previous literatures. The reliability of previous algorithms is discussed and counterexamples are provided for illustration. The statistical results show RSAA is adequate for implementation in extended Puck’s criterion and much more reliable than previous algorithms. RSAA can correctly predict the results with a probability of over 99.999%.     

Effect of the 3rd Dimension within the Representative Volume Element (RVE) on Damage Initiation and Propagation during Full-Phase Numerical Simulations of Single and Multi-Phase Steels

In this research, the effect of 2D and 3D Representative Volume Element (RVE) on the ductile damage behavior in single-phase (only ferrite) and dual-phase (ferrite and martensite) steels is analyzed. Physical and fitting parameters of the constitutive model for bcc-ferrite and bcc-martensite phases are adapted from the already published work. Crystal plasticity (CP) based numerical simulations without damage consideration are run and, later, ductile damage criteria for the ferrite phase is defined for all cases. The results of the non-damage (-nD-) and damage (-D-) simulations are compared to analyze the global and local differences of evolving stresses and strains. It is observed that for the same model parameters defined in all cases, damage initiation occurs at the overall higher global strain in the case of 3D compared to 2D. Based on statistical data analysis, a systematic comparison of local results is carried out to conclude that the 3D RVEs provide better quantitative and qualitative results and should be considered for such full phase simulations. Whereas 2D RVEs are simple to analyze and provide appropriate qualitative information about the damage initiation sites.     

Predictive Computational Model for Damage Behavior of Metal-Matrix Composites Emphasizing the Effect of Particle Size and Volume Fraction

In this paper, an integrated numerical model is proposed to investigate the effects of particulate size and volume fraction on the deformation, damage, and failure behaviors of particulate-reinforced metal matrix composites (PRMMCs). In the framework of a random microstructure-based finite element modelling, the plastic deformation and ductile cracking of the matrix are, respectively, modelled using Johnson–Cook constitutive relation and Johnson–Cook ductile fracture model. The matrix-particle interface decohesion is simulated by employing the surface-based-cohesive zone method, while the particulate fracture is manipulated by the elastic–brittle cracking model, in which the damage evolution criterion depends on the fracture energy cracking criterion. A 2D nonlinear finite element model was developed using ABAQUS/Explicit commercial program for modelling and analyzing damage mechanisms of silicon carbide reinforced aluminum matrix composites. The predicted results have shown a good agreement with the experimental data in the forms of true stress–strain curves and failure shape. Unlike the existing models, the influence of the volume fraction and size of SiC particles on the deformation, damage mechanism, failure consequences, and stress–strain curve of A359/SiC particulate composites is investigated accounting for the different possible modes of failure simultaneously.     

A Computational Geometry Generation Method for Creating 3D Printed Composites and Porous Structures

A computational method for generating porous materials and composite structures was developed and implemented. The method is based on using 3D Voronoi cells to partition a defined space into segments. The topology of the segments can be controlled by controlling the Voronoi cell set. The geometries can be realized by additive manufacturing methods, and materials can be assigned to each segment. The geometries are generated and processed virtually. The macroscopic mechanical properties of the resulting structures can be tuned by controlling microstructural features. The method is implemented in generating porous and composite structures using polymer filaments i.e., polylactic acid (PLA), thermoplastic polyurethane (TPU) and nylon. The geometries are realized using commercially available double nozzle fusion deposition modelling (FDM) equipment. The compressive properties of the generated porous and composite configurations are tested quasi statically. The structures are either porous of a single material or composites of two materials that are geometrically intertwined. The method is used to produce and explore promising material combinations that could otherwise be difficult to mix. It is potentially applicable with a variety of additive manufacturing methods, size scales, and materials for a range of potential applications.     

An MMF3 Criterion Based Multi-Scale Strategy for the Failure Analysis of Plain-Woven Fabric Composites and Its Validation in the Open-Hole Compression Tests

A modified micromechanics failure criterion (MMF3) based multi-scale analysis strategy was proposed in this article to analyze the failure behaviors of the plain-woven fabric composites. The finite-element (FE) representative unit cell (RUC) models of different scales were first established, and the RUC based stress transformation methods were developed. The micro-scale strengths of the constituents in the unidirectional laminate were achieved based on the tested macro-scale strengths. Under the micro-scale strength invariance hypothesis, the meso-scale strengths of the fiber tows from the plain-woven fabric composites were back-calculated first and were then validated and corrected with the assistance of tested strengths of the fabric laminates. With the micro-scale RUC and the calculated meso-scale strengths of the fiber tows, the micro-scale strengths of the constituents suitable for the plain-woven fabric composites were determined. The multi-scale analysis procedure for the plain-woven fabric composites was then established in providing a more direct failure observation at the constituent level. Open-hole compression specimens were tested according to the ASTM standard D6484, and the failure of the open-hole fabric laminate was simulated with the proposed multi-scale strategy. The numerical predictions were in good agreement with the experimental results, and the feasibility of the multi-scale strategy was validated.