NEURAL NETWORK-BASED SIMULATION OF THE INFLUENCE OF WASTE TIRE RUBBER ON POLYESTER-FIBERGLASS COMPOSITES | ||||
| Journal of Al-Azhar University Engineering Sector | ||||
| Volume 20, Issue 74, January 2025, Page 182-192 PDF (919.6 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/auej.2024.316695.1706 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Ahmed AboHassan   1; Tamer Samir Mahmoud1; Ibrahim Moussa1; Ahmed Omar Mosleh2
| ||||
| 1Mechanical Engineering, Faculty of Engineering at Shoubra, Benha University, Cairo, Egypt | ||||
| 2Mechanical Engineering Department, Faculty of engineering at Shoubra, Benha University, Shoubra | ||||
| Abstract | ||||
| This research presents a comprehensive neural network-based simulation to delve into the intricate relationship between waste tire rubber and polyester fiberglass composites. By meticulously investigating the effects of varying mesh sizes and volume percentages of rubber particles, the fabrication process utilizes hand lay-up and vacuum degassing to ensure optimal composite quality. The study aims to accurately predict the mechanical and dynamic properties of these composites. These properties, including ultimate tensile strength (UTS), strain, impact resistance, natural frequency, and damping factor, are critical determinants of the composite's performance in various applications. A neural network model was meticulously crafted and trained using the backpropagation algorithm, with a mean squared error of 10-8. This exceptional accuracy underscores the model's ability to effectively capture the complex interactions between the composite components. The model demonstrated remarkable proficiency in predicting UTS, impact resistance, natural frequency, and damping factors, achieving regression coefficients of R= 96.30%, 96.50%, 97.80%, and 97.40%, respectively. Moreover, the strain prediction accuracy was commendable, with a regression coefficient of R= 94.20%. These findings collectively underscore the immense potential of neural networks in optimizing the design of composite materials incorporating waste tire rubber. By leveraging the predictive capabilities of these models, researchers and engineers can develop sustainable materials that not only exhibit superior performance but also contribute to a more environmentally responsible future. The integration of waste tire rubber into composite materials offers a promising avenue for reducing waste and promoting circular economy principles. | ||||
| Keywords | ||||
| Neural Networks; Recycling; Polyester-Fiberglass Composites; Waste Tire Rubber Particles; Mechanical and Dynamic Properties | ||||
|
Statistics Article View: 152 PDF Download: 85 |
||||
