Predictive modeling of abrasive wear in in-situ TiC reinforced ZA37 alloy: A machine learning approach

Abstract

Abrasive wear rates are of significant interest in various industrial applications where materials are subjected to severe wear conditions. The study analyzed high-stress abrasive wear rates on the novel in-situ TiC reinforced ZA37 composites. The inclusion of in-situ TiC reinforcement in ZA37 alloy has shown potential for enhancing wear resistance. The impact of tribological test parameters on the abrasive wear response was analyzed using response surface methodology (RSM). Using tribological data, various machine-learning algorithms were trained to predict the wear behaviours of the developed composites. The performance measurements show that the machine learning models accurately predicted the abrasive wear response of test samples. Our findings suggest that machine learning can revolutionize tribology, paving the way for tribo-informatics.