Tribological properties of CNT-filled epoxy-carbon fabric composites: Optimization and modelling by machine learning

Abstract

Polymer matrix composites reinforced with fibers/fillers are extensively used in several tribological components of automotive and boating applications. The mechanical performance of polymer composites improves by incorporating nanofillers as secondary reinforcement. The present research work fabricated carbon fabric-reinforced epoxy composites using the hand layup. The carbon fabric-reinforced polymer composites were fabricated with 0.1 wt%, 0.2 wt%, and 0.5 wt% of carbon nanotubes (CNT) fillers as secondary reinforcement. Tribological properties of carbon fabric-reinforced epoxy composites filled with CNT have been carried out using a pin‐on‐disc method. Adding fillers significantly improves the tribological behaviour of the carbon fabric-reinforced epoxy composites by reducing wear rate and coefficient of friction. The large surface area of interaction due to the higher aspect ratio of CNT shows improved adhesion between epoxy matrix and carbon fabrics. It improves the various mechanical and tribological characteristics of composites—also, an analysis of worn surfaces is carried out to analyze the wear mechanisms using scanning electronic microscopy. The research employs a combination of experimental analyses and machine learning (ML) techniques to explore the wear resistance, hardness, and predictive modeling of volume loss in the composites. The hyperparameter fine-tuning of ML algorithms, including Random Forest (RF), k-Nearest Neighbors (KNN), and XGBoost, demonstrates superior predictive capabilities, particularly with RF. The study bridges material science, ML, and practical applications, contributing valuable insights for developing advanced composite materials.

Keywords

• Epoxy
• Carbon fabric
• Carbon nanotubes
• Wear
• Machine learning