Application of box-behnken design and neural computation for tribo-mechanical performance analysis of iron-mud-filled glass-fiber/epoxy composite and parametric optimization using PSO

This study investigates the possible utilization of iron mine waste for developing a new class of hybrid polymer composites. The composites were fabricated using hand-layup process by reinforcing woven glass fibers in the epoxy polymer filled with different weight proportions of iron-mud. Abrasion wear experiments were conducted according to Box-Behnken design approach under controlled laboratory conditions using a dry abrasion tester. It was found that hardness, tensile modulus, impact energy and abrasion resistance of the fabricated composites improved with filler addition. Also, a prediction tool based on artificial neural network was implemented to investigate the tribological properties of the composites and the results were compared with the experimental ones. A metaheuristic approach like particle swarm optimization was also used to reveal the minimum wear (in volume) at optimal parametric combination. The results showed increase in both wear (in volume) and specific wear rate with respect to increase in the loads as well as sliding velocity. It also exhibited an increase in the wear with decrease in the specific wear rate with respect to the abrading distance. Finally, the morphology of the abraded surfaces was examined by using scanning electron microscopy and the possible abrasion mechanisms were critically examined and presented. POLYM. COMPOS., 40:1433-1449, 2019. (c) 2018 Society of Plastics Engineers