Effect of fly-ash as filler material on mechanical and thermal properties of jute-wool-epoxy hybrid composite: prediction using artificial neural network model
摘要
The growing preference for natural-based composites is driven by their biocompatibility, cost-effectiveness, and availability. This study explores the effect of fly ash as a sustainable filler in jute-wool-epoxy hybrid composites, aiming to enhance mechanical, thermal, and flame-retardant properties. Composites were fabricated using the hand layup method with fixed jute and wool fiber contents (25 wt% each) and varying fly ash content (5, 10, 15, and 20 wt%). Mechanical tests, including tensile, flexural, interlaminar shear strength (ILSS), fracture toughness, and impact resistance, were conducted alongside thermogravimetric analysis (TGA) and flammability tests. The composite with 15 wt% fly ash (JCW15) showed superior mechanical performance, with tensile, flexural, ILSS, fracture toughness, and impact strength improvements ranging from 4 to 26% over other configurations. JCW20 exhibited the best thermal stability with a 28.64% char residue at 800 °C and the highest flame resistance, achieving a burning rate of 16.98 mm/min. SEM analysis confirmed improved fiber–matrix adhesion at optimal filler content (15 wt%). Furthermore, an Artificial Neural Network (ANN) model was developed using experimental data to predict tensile, flexural, and impact strengths. The model achieved high prediction accuracies of 85.35%, 84.44%, and 86.42%, respectively. These findings demonstrate the effectiveness of fly ash in improving the performance of natural fiber-based epoxy composites and highlights the potential of machine learning for reliable property prediction. The study supports the development of cost-effective, eco-friendly composites suitable for structural and thermal applications.