Predictive Modelling of Wear and Friction in Surface Textured TiAlN Coated Ti6Al4V Alloy Using Artificial Neural Networks

Friction and wear rate are crucial properties of the Ti6Al4V alloy in bioimplants, which relate to the study of friction, wear, and lubrication of interacting surfaces in relative motion. These properties are significant factors that determine the longevity and performance of the implant. To enhance these properties, surface modifications have been carried out, such as applying a TiAlN coating on the textured surface of the alloy. In this study, the effect of high bonding coatings of TiAlN on textured Ti6Al4V was investigated using a tribometer wear tester by varying three independent parameters: sliding velocity, sliding distance, and applied load. The aim was to optimize the wear and friction properties of surface textured TiAlN coated Ti6Al4V alloy by changing values of the sliding velocities (1.25 m/s, 2.1 m/s and 3.2 m/s), sliding distance and applied load (80 N, 100 N and 120 N). To save time and costs, an artificial neural network (ANN) model using feed-forward backpropagation was developed. The model was trained and evaluated using experimental data to predict the specific wear rate and friction coefficient of the Ti6Al4V alloy. The study demonstrates that the created ANN model can accurately predict the specific wear rate and friction coefficient with an error percentage of less than 2.5% and 3.5%, respectively. Therefore, using the ANN model to predict the wear rate and coefficient of friction of Ti6Al4V alloy during the experimental process can significantly save time, effort, and money.