Microstructural Analysis & Tribology Characteristics of Magnesium Alloy AZ31B and Its Composites Using Machine Learning Modeling

Abstract Examining the effects of wear factors and the wear rate (WR) of magnesium (AZ31) composites is the main objective of this work. Silicon carbide (SiC) and graphene are used as reinforcing materials in the stir casting technique used to create the composite materials. In the current study, three tribological factors sliding distance, velocity, and load as well as one material factor, material type were chosen to investigate their effects on wear rate. The Taguchi technique is used to design the tests, and it has been found that load (L), followed by MT, D, and V, has the greatest impact on WR. The following are the ideal values for the influencing parameters for WR: MT = T3, L = 10 N, V = 3m/s and D = 500 m. SEM micrographs of the wear pin's surface and its by-products were used to study the wear mechanisms under the highest and lowest WR conditions. According to the SEM study, the worn surface displayed signs of oxidation, adhesion, delamination, and abrasion mechanisms. Decision trees (DT) is examples of machine learning (ML) model that were used to create an efficient prediction model that accurately predicted the output responses to the subsequent input variables. Confirmation tests were run under ideal circumstances, and the same was checked against the outcomes of DT.