Investigation of the Wear Behavior of AA6063/Zirconium Oxide Nanocomposites Using Hybrid Machine Learning Algorithms

This research created hot-pressed composites of the AA6063 matrix with varying concentrations ZrO2 (0.25, 0.5, and 1 wt %). At sliding speeds 80, 120, 150 mm/s, wear performance specimen was studied at loads 10 N, 15 20 25 N. The authors analyzed counter-face material, debris, worn surfaces to learn about mechanisms. Developing these three machine learning (ML) algorithms evaluate ability predict behavior using same small dataset collected test processes. A thorough examination each model hyperparameter tuning phase performed. predictive several statistical tools. most effective decision-making for this data collection were those based on trees. Predictions made by decision tree algorithm validation measurements have an accuracy 86% 99.7%, respectively. best picked out results predictions.