Gholam Hossein Rahimi

1. In mechanical structural analysis, both the configuration and the choice of materials play a fundamental role in determining performance. Specific geometric configurations can enhance the struct...

This study presents an advanced, non-destructive method for evaluating the buckling behavior of sandwich beams by integrating the vibration correlation technique (VCT) with machine learning (ML) and genetic algorithms (GAs). The investigation focuses on a sandwich beam featuring an innovative honeycomb core, designed from a unit cell characterized by five adjustable parameters. Three ML models — [ ]-Nearest Neighbor, Single-Layer Perceptron, and Multi-layer Perceptron — are developed to accurately relate these design parameters to two key performance metrics: the critical buckling load-to-weight ratio and the first natural frequency-to-weight ratio. The continuous predictive model derived from these algorithms is incorporated into a GA framework to simultaneously maximize both performance indices, thereby identifying an optimal set of design parameters. The optimized configuration is then realized through 3D printing of the sandwich beam. To determine the equivalent mechanical properties of the honeycomb core, a homogenization approach combined with finite element analysis is employed. Additionally, the buckling and free vibration responses of the structure are analyzed using equations derived from Timoshenko beam theory and solved via a meshless numerical method. Comparison of the VCT-based predictions with experimental measurements confirms the accuracy and robustness of the integrated approach. The results demonstrate that the synergy of ML, evolutionary optimization, and advanced structural modeling provides a powerful tool for the design and analysis of complex structural systems.

In this paper, the delamination behavior between the face sheet and core of sandwich panels under a mixed mode of failure during TPSB (Three-Point Sandwich Beam) testing is investigated. The experiments were conducted on two sets of specimens: one reinforced with 0.25% carbon nanofiber weight and the other consisting of samples without carbon nanofibers. The core of the sandwich structures was made of PVC foam, and the corrugated cores and face sheets were made of glass-epoxy fibers. The investigated specimens had corrugated cores with different geometries, such as trapezoidal, square, and triangular. The Force-Displacement and R-CURVE diagrams of sandwich panels reinforced with and without carbon nanofibers were extracted. The strain energy release rate increased with increasing crack length, and this trend in the mixed mode of failure was generally upward. The strain energy release rate in the sandwich panel reinforced with carbon nanofibers with trapezoidal, square, and triangular corrugated cores increased by 20.7%, 39.8%, and 44.6%, respectively. These findings underscore the potential of carbon nanofiber-reinforced sandwich structures for diverse applications requiring enhanced fracture resistance under varying loading conditions. These structures have many applications in wind turbine blades and aerospace industry.

Sandwich structures are often prone to catastrophic failure due to premature separation between the core and skin layers. Geometric parameters, such as the thickness of the skin and core, along with the materials used in their manufacturing, directly influence the resistance to separation between the skin and core. This study explores how variations in skin thickness affect both failure modes and maximum load capacity in sandwich structures, utilizing both experimental testing and numerical simulations. Specimens were categorized as intact and pre-debonded samples. Each specimen featured four different skin thicknesses (3, 6, 8, and 10 layers of composite laminated skins, with corresponding thicknesses of 0.68, 1.33, 1.73, and 2.1 mm respectively). The specimens incorporated a foam-filled square corrugated core and underwent 3-point bending tests. Results revealed a significant shift in the failure mode: initially observed as upper skin fracture (V-shaped failure), it transitioned to separation between skins and cores with increased skin thickness, particularly in the presence of pre-debonding. Notably, the predominant failure mode did not involve separation between the skin and core in specimens without a pre-existing crack. Furthermore, numerical simulations effectively demonstrated the accurate capture of failure modes and loads using the Hashin and cohesive zone model (CZM).

In the present study, nonlinear numerical and experimental static bending analyses of hyperelastic beams are surveyed. Timoshenko's beam theory is used to obtain the Cauchy-Green deformation tensor. The governing equations are extracted using the neo-Hookean strain energy and Euler–Lagrange relation. Afterward, the nonlinear governing equations are solved numerically using the meshless collocation method (MCM). To validate the present investigation, a comparison study is accomplished between the acquired results from MCM and the experimental test of digital image correlation (DIC) and finite element method (FEM). The hyperelastic beam is made of natural rubber. Also, loading and boundary conditions are considered as concentrated force and clamped–clamped, respectively. In addition, the displacement contour is extracted by the DIC. The uniaxial tensile test based on the ASTM D412 standard is performed to extract the mechanical properties of natural rubber. The hyperelastic theory and neo-Hookean strain energy function are utilized to model the nonlinear behavior of rubber in the FEM. The error for the maximum displacement of system obtained from both numerical methods is less than 12% compared to the DIC. Additionally, it can be observed that in large deformations, unlike the FEM, MCM is able to predict the system's response with high accuracy.

The present study experimentally evaluates delamination of hybrid corrugated-core sandwich structures in the mode II of failure under flexural loading. Three sandwich structures with different cores, including core with no reinforcement, trapezoidal reinforcement, and triangular reinforcement, were fabricated. The strain energy release rate (G) of each specimen was then measured in the mode II of crack growth. The R-curves of the specimens were also extracted using experimental data, and by photographing the crack growth. Furthermore, the effects of the initial crack length on the R-curve were measured. In the last phase of this study, the specimens were numerically modeled using the finite element method (FEM). The delamination of the core and face sheets was modeled using the cohesive zone method (CZM), and the load–displacement curves of the specimens were extracted and compared to those of experimental data.