Perpustakaan Digital ITB

Recent development of global conflicts on the battlefield, including the Ukraine War, have recorded 4,710 casualties from mines and explosive remnants of war. This highlights the urgent need to enhance the capabilities of the Armored Vehicles (AV) to withstand TNT impact from 8 kg mine blast, as specified by STANAG 4569. To enhance the protective capacity of the Armored Vehicles, an additional structure is proposed to be utilized, serving as an underbelly layer. The suggested solution involves constructing a sandwich-like structure, employing 2D lightweight metastructures as the core. The research methodology encompasses subjecting 2D lightweight metastructure sandwich configurations to blast load simulations with CONWEP blast function using the LS-DYNA software and conduct optimization using Design for Six Sigma (DFSS) and machine learning methods. The ultimate goal is to achieve an optimal design for 2D Lightweight Sandwich Metastructures from various candidates of 2D Lightweight Metastructures such as 2D Double-U Hierarchical (DUH), 2D Re-entrant star structure, 2D origami structure, and 2D AuxHex structure. Optimization was performed in two steps scenario, first by using DFSS Taguchi DOE method to obtain the optimum selection based on compression load and optimum Specific Energy Absorption (SEA) criteria. By optimizing the core first, computational burden and the number of sample data points needed can be minimized, ensuring the core achieves the highest Specific Energy Absorption (SEA). The second round then fine-tunes the entire panel, optimizing the interaction between core and face sheets, leading to better-informed decisions and improved accuracy and reliability of results. The result from first optimization was then further optimized using Artificial Neural Network (ANN) and Non-dominated Sorting Genetic Algorithm-II (NSGA-II) methods to obtain overall structural efficiency against blast load, with optimum SEA and minimum Occupant Side Plate Displacement as the criteria. The first optimization selects AuxHex core made from 20MnCr5 material with 10x4 unit cell and 1.5 mm thickness as the optimum core candidate, which after second optimization gives the configuration of AuxHex core at 10% relative density, with the Occupant Side Plate made of 1300T material at 6.44 mm thick, and the Struck Side Plate using SS316L at 5.96 mm thickness as the optimum choice. Each layer of the structure is bonded with an adhesive strength of 150 MPa. The highest SEA attained by the optimized sandwich metastructure design is 181.48 J/kg, with an Occupant Side Plate Displacement of 39.54 mm.