Mechanical and wear performance of Al-,mica-,SiO 2 -filled glass fiber-reinforced composites and prediction of wear properties with artificial neural networks

Abstract

This study investigates the effects of different categories of microparticles on the mechanical and tribological properties of glass fiber-reinforced composites. The tensile and wear properties of composites produced by hand lay-up technique, with Al, mica, and SiO₂ added to epoxy matrix at 2, 4, and 6%, were determined. The average particle size ranges from 20 to 50 μm for Al, <15 μm for SiO₂, and from 30 to 50 μm for mica. SiO₂ particles have flaky shapes, and Al and mica particles have irregular shapes. The highest and lowest tensile strengths were obtained in the composites with 2% Al and 6% mica fillers, respectively. Increasing the filler content in all filled composites resulted in a decrease in mechanical properties. The best mechanical properties were observed at the 2% filler content. It was determined that SiO₂, mica, and Al powders were the most effective filler types in reducing coefficient of friction of unfilled composites, in that order. An increase in filler content in the composites filled with Al and mica increased the wear rate. The lowest wear loss was observed in SiO₂-filled composites. The damage mechanisms, matrix deformations, and fiber fractures of the composites after wear testing were identified through scanning electron microscopy imaging. An artificial neural network model was generated to estimate the wear rates of the filled composites, and the artificial neural network model was successful in validating the experimental results.