Effect of REOs on tribological behavior of aluminum hybrid composites using ANN

This chapter explores the effect of cerium oxide as rare earth oxides (REOs) on the tribological properties of aluminum hybrid composites with different compositions of reinforcement like SiC, Al₂O₃, and CeO₂. For this motive composites had been synthesized by varying SiC/Al₂O₃ from 2.5–7.5 wt.% with equal proportion and CeO₂ from 0.5–2.5 wt.% in an Al-6061 matrix. The addition of cerium oxide with contents of 0.5–2.5 wt.% to the aluminum composites leads to the formation of an intermetallic phase (Al₄Ce₃), resulting in an improved wear rate up to 87.28%. To predict the effect of incorporating REO reinforcements on the tribological behavior of hybrid composites, experimental data of wear tests are used to create 3D models named Levenberg–Marquardt Algorithm (LMA) neural networks. The consequences show that the LMA neural networks models have a high level of accuracy in the prediction of tribological properties for aluminum hybrid composites reinforced by rare earth oxides (REOs).