Wear Rate Analysis of Metal Matrix Composite Using Machine Learning Algorithms

There is a lot of interest in a compound with improved mechanical power, toughness, wear resilience, and increased electric and thermal conductance. This work examined the tribological conduct and fabrication of the titanium metal matrix composite (TiMMC) augmented with graphene (Gr) and tungsten carbide (WC) fragments. The TiMMC, which had 8 percent mass percentages of WC and Gr, was created using stir casting. Taguchi's L27 orthogonal grid method was used for designing the tribological investigations, which were then conducted as wear experiments with a pinon-disc gadget. ANOVA and Taguchi's study show that loading and range have the most effect on wear percentage, accompanied by speed.By examining the worn areas of the composite samples using scanning electron microscopy, the wear mechanism was determined. The wear rate statistics were correctly categorized by machine learning classifying techniques such as random forest, support vector machines, and XG-Boost methods, which provided accuracy values of 72%, 66%, and 56.3%, respectively. Notwithstanding the encouraging outcomes, the research acknowledges that the system's efficiency may differ depending on certain properties of the composite component and operating circumstances. Therefore, it motivates further research to validate and extend these novel discoveries over a larger range of components and circumstances.