Session: 01-01-01: General Topics of Aerospace Structures 1

Paper Number: 159576

159576 - Optimizing Impact Resistance and Energy Absorption Capabilities of Cellular Solids Using Neural Network Response Surfaces 

This study computationally optimized bending and axially dominated cellular structures using five material combinations and three impact scenarios. The primary focus was on analyzing the acceleration transmitted to the structure's base to minimize head injury risk, based on acceleration-based injury criteria. Prior to optimization, it was found that Type I structures, with the same unit cell strut/ring thickness, were lighter and more effective at reducing acceleration compared to Type II structures. A design of experiments approach was employed, running fifteen simulations for each of the thirty cases. A neural network model was trained, adjusting weights and biases until the mean squared error reached an acceptable level. The relationship between varying input thicknesses and the resulting accelerations was used to optimize the structure, reducing the acceleration transmitted to the base. The optimization process resulted in thicker unit cells at the base of the structure and thinner, more deformable unit cells at the impact surface. The analysis revealed that higher Shore hardness values reduced material usage but transmitted higher vibrations, while lower Shore values significantly reduced vibrations but added weight. When the impact kinetic energy was increased by a factor of 25, the optimized structures saw a 10% increase in material volume, emphasizing the importance of accurately predicting the impact kinetic energy a helmet wearer may experience to ensure adequate protection without excessive weight. The study also explored the tradeoff between weight and protection, comparing acceleration-time history data from simulations against head injury criteria and injury probability thresholds. Overall, the study demonstrates that viscoelastic digital materials can significantly attenuate accelerations, thereby lowering the risk of brain injury. Additionally, it underscores the importance of designing helmet structures based on current brain injury criteria to ensure effective protection.

Presenting Author: Ashfaq Adnan University of Texas at Arlington

Presenting Author Biography: Dr. Ashfaq Adnan is currently a distinguished university professor in the Mechanical & Aerospace Engineering Department and director of the Multiscale Mechanics and Physics Lab (MMPL) at UTA. In 2008, he earned his PhD in Aeronautics & Astronautics Engineering from Purdue University. Prior to joining UTA in the Fall of 2010, he was a postdoctoral research associate at Northwestern University. His current research covers cellular level brain injury mechanisms, smart sensing of brain trauma and additive manufacturing of protective materials. His research is funded by multiple Office of Naval Research (ONR) grants and Defense University Research Instrumentation Program (DURIP) grants. He is the recipient of the Department of Navy’s inaugural 2021 Distinguished Fellow Program Award. He is an ASME fellow, the recipient of 2020 Lockheed Martin Teaching Award, 2022 University Award for Outstanding Research Achievement and 2023 UTA Academy of Distinguished Scholars. He is currently appointed by the US National Science Foundation as an expert to serve the mechanics of materials and structure program within the ENG-CMMI division. He is also a co-founder and CTO of a startup “Cortex Dynamix LLC”.