Browsing by Author "Arhore, Edore G."

Now showing 1 - 5 of 5
  • Thumbnail Image
    ItemOpen Access
    Comparison of GA and topology optimization of adherend for adhesively bonded metal composite joints
    (Elsevier, 2021-09) Arhore, Edore G.; Yasaee, Mehdi; Dayyani, Iman
    This paper investigates the effect of the outer adherend geometry on the strength of an adhesively bonded joint. The investigation was carried out by optimizing the joint geometry using two different numerical optimization methods. In genetic algorithm (GA) optimization with high fidelity explicit finite element analysis (FEA) and topology optimization (TOP). Both procedures were utilized on a simplified pseudo-2D model as well as a full-scale 3D model. The results showed that the outer adherend geometry directly affects the strength of a joint subjected to tensile load. For joints subjected to bending load, the geometry had little to no effect on the strength of the joint. The GA optimization process produced identical geometry for both 2D and 3D models. However, the TOP process produced different optimum geometries. The optimum joints produced by the TOP process offered the highest strength overall, while the optimum GA joint produced the best strength to weight ratio. The reasons for these results and other features of the optimized designs, including interface stress, failure mechanisms and computational efficiency are discussed in detail.
  • Thumbnail Image
    ItemOpen Access
    Experimental and numerical investigation of multi-layered honeycomb sandwich composites for impact mechanics applications
    (Elsevier, 2024-02-01) Al Ali, A.; Arhore, Edore G.; Ghasemnejad, Hessam; Yasaee, Mehdi
    This project aims to investigate the design of a multi-layered sandwich composite and its performance under impact loading conditions. An experimental and numerical assessment was performed to conclude the effect of increasing the layers of sandwich panels. Three specimens of four different sandwich panel configurations were manufactured to be tested. The skin of the sandwich panels comprises a twill carbon-reinforced epoxy resin, whereas the core consists of a 2D Nomex honeycomb core. The panels are then subjected to transverse impact loading to investigate their impact behaviour. These experimental results are then used to verify numerical models constructed in LS-Dyna. The models of the honeycomb-reinforced sandwich panels are investigated using MAT-054 and MAT-142 material cards in LS-Dyna to find the most economical computational approach. Finally, the energy absorption characteristics calculated during the experimental and numerical work are used to conclude the multi-layered sandwich composite's performance and provide design recommendations. The findings suggest that by increasing the core and shell numbers through the thickness of the panel, the specific energy absorption capability will increase.
  • Thumbnail Image
    ItemOpen Access
    Lay-up optimisation of fibre–metal laminates panels for maximum impact absorption
    (Sage, 2020-06-24) Arhore, Edore G.; Yasaee, Mehdi
    This paper introduces a methodology utilising a ply-ply damage Finite Element models with Genetic algorithm optimisation procedure to investigate the effect of lay-up configuration on the impact absorption properties of fibre metal laminates (FMLs). The methodology was carried out in two steps. In the first step, a pseudo-2D model was used to explore the vast design space to identify potential optimised layup-configurations. In the second step, the optimised configurations were studied in full 3 D, with high fidelity simulations, verifying the results obtained from the optimisation process. The design variables used include thickness and material (including fibre orientation) of each ply. The results produced an optimised configuration consisting of a metallic ply on the impacted side followed by a cross-ply composite lay-up. The results also suggest that the first composite ply (second ply of the FML) should be about 3 times thicker than the other plies
  • Thumbnail Image
    ItemOpen Access
    Neural network assisted Ga optimization of adhesively bonded composite joints
    (Unknown, 2022) Arhore, Edore G.; Yasaee, Mehdi; Dayyani, Iman
  • Thumbnail Image
    ItemOpen Access
    Optimisation of convolutional neural network architecture using genetic algorithm for the prediction of adhesively bonded joint strength
    (Springer, 2022-09-02) Arhore, Edore G.; Yasaee, Mehdi; Dayyani, Iman
    The classical method of optimising structures for strength is computationally expensive due to the requirement of performing complex non-linear finite element analysis (FEA). This study aims to optimise an artificial neural network (ANN) architecture to perform the task of predicting the strength of adhesively bonded joints in place of non-linear FEA. A manual multi-objective optimisation was performed to find a suitable ANN architecture design space. Then a genetic algorithm optimisation of the reduced design space was conducted to find an optimum ANN architecture. The generated optimum ANN architecture predicts efficiently the strength of adhesively bonded joints to a high degree of accuracy in comparison with the legacy method using FEA with a 93% savings in computational cost.