Predicting the wear performance of graphene and silicon nitride reinforced aluminium hybrid nanocomposites using artificial intelligence approach

Hybrid nanocomposites of aluminium (NHAMMCs) made from AA5052 are fabricated via stir casting route by reinforcing 12 wt% Si3N4 and 0.5 wt% of graphene for usage in aeronautical and automotive applications due to the lower density and higher strength to weight proportion. The wear characteristics of the NHAMMCs are evaluated for different axial load, rotational speed, sliding distance and sliding time based on Box-Behnken design (BBD) of response surface methodology (RSM). Orowan strengthening mechanism is identified from optical image which improves the strength of the composite. Outcomes show that with higher axial load and rotational speed, there is substantial increase in wear loss whereas with increased sliding distance and sliding time there is no considerable increase in wear loss due to the lubricating nature of the reinforced graphene particles since it has higher surface area to volume ratio. Besides, artificial intelligence approach of neuro-fuzzy (ANFIS) model is developed to predict the output responses and the results are compared with the regression model predictions. Prediction from ANFIS outplays the regression model prediction.