Tribological aspects of AA6082/graphene/TiC hybrid composite using Taguchi and Machine Learning

In the present research, hybrid composite consists of AA6082 aluminum base with TiC and n-graphene ceramics reinforcements ranging from 1 to 6 wt.%. High-energy ball milling created homogeneous mixtures between the two reinforcing components that together occupied 50 wt.% of total content. The hybrid aluminum matrix composite was fabricated through stir casting before being examined for wear resistance against the counter surface of EN32 steel by means of a pin-on-disc tribometer. The optimization process for pin-on-disc testing parameters merged Taguchi method with machine learning (ML) approaches. Load stands as the primary factor determining the dry sliding wear rate (WR) of materials according to analysis of variance (ANOVA) results while reinforcement content and speed play additional roles. The slope of both two-factor interaction effects demonstrated meaningful change. The Taguchi and ML determined that the optimized parameters would lead to a WR of 6.94×10 -4 mg/s as the minimum value. Microstructural examination using scanning electron microscopy (SEM) finds that very small grooves are seen at optimal settings, whereas severe ploughing is shown at other settings. The wear mechanism transitions from adhesive to abrasive when the speed between the pin and disc increases.