Study of self lubrication property of Al/SiC/Graphite hybrid composite during Machining by using artificial neural networks (ANN)

Abstract

Particulate-Metal Matrix Composites have captivated huge attention for a wide range of engineering applications because of their superior mechanical properties such as high strength, corrosion resistance and wear resistance. However, they lag behind high machinability (due to the presence of hard ceramic reinforcements) and involve expensive processing techniques that restrict their widespread application. The present work is aimed at improving machinability of Al alloy/SiC/graphite hybrid composite developed by a low-cost stir casting technique. Microstructure studies of the hybrid composite were investigated by using SEM and optical microscope. The obtained results were then correlated with mechanical properties to reveal significant enhancement when compared to AA2024 alloy without reinforcements. The machining forces of hybrid composite during drilling were studied by custom-designed strain gauge-based dynamometer. Machinability was improved with graphite incorporation, which has a self-lubricant texture, without much compromising the hardness and tensile strength of the hybrid composite. Further, neural network analysis was adopted by training the model to accurately predict the cutting forces to realize efficient composite with suitable machining parameters.