Utilization of Improved Machine Learning Method Based on Artificial Hummingbird Algorithm to Predict the Tribological Behavior of Cu-Al2O3 Nanocomposites Synthesized by In Situ Method

This paper presents a machine learning model to predict the effect of Al2O3 nanoparticles content on wear rates in Cu-Al2O3 nanocomposite prepared using situ chemical technique. The developed is modification random vector functional link (RVFL) algorithm artificial hummingbird (AHA). objective AHA used find optimal configuration RVFL enhance prediction nanoparticles. preparation composite was done aluminum nitrate that added solution containing scattered copper nitrate. After that, powders CuO and were obtained, leftover liquid removed thermal treatment at 850 °C for 1 h. consolidated compaction sintering processes. microhardness with 12.5% 2.03-fold times larger than pure copper, while rate same reduced, reaching 55% lower copper. These improved properties are attributed presence their homogenized distributions inside matrix. RVFl-AHA able all composites different load speed, very good accuracy, nearly 100% 99.5% training testing, respectively, terms coefficient determination R2.