Predicting wear behavior of AZ31/TiC composites produced via ultrasonic vibration assisted friction stir processing using machine learning models.

This work aims to produce AZ31 Mg alloy reinforced with 15 vol.% TiC utilizing a unique Friction Stir mechanical vibration Processing (FSVP) technology, emphasizing the improvement in tribological properties of the produced composite. The dry sliding wear characteristics of the composite generated using conventional Friction Stir Processing (FSP) and the suggested FSVP process were assessed under different applied loads utilizing a pin-on-disc tribometer. The experimental findings indicate that FSVP markedly decreases the coefficient of friction and enhances wear resistance, especially under elevated load circumstances. The incorporation of 15 vol.% TiC into the AZ31 matrix reduced the wear rate by approximately 25% at moderate loads and up to 50% at elevated loads. Analysis of the wear mechanisms indicated a progression from mild oxidative and abrasive wear at lower loads to delamination, adhesion and severe plastic deformation under higher load regimes. The superior wear performance of the FSVP-processed AZ31/15 vol.% TiC composites is attributed to increased surface hardness and enhanced work-hardening behavior. Furthermore, a Gradient Boosting machine learning model exhibited excellent predictive accuracy achieving R