Enhancing wear resistance and thermal stability of laser-induced zirconia-aluminum composites

The research investigates laser processing of zirconia-aluminum composites for enhanced wear resistance and thermal stability. A structured methodology is developed, encompassing material selection, in-situ temperature measurement, theoretical modeling of failure mechanisms, and artificial intelligence-based optimization. The influence of significant laser parameters such as laser output power and cutting speed on material behavior is examined. Experimental comparisons between four groups of samples under different processing conditions for the assessment of thermal and mechanical performance are conducted in the research. Laser parameter optimization is carried out using artificial intelligence modeling using Tangent bundle neural network and magnificent frigatebird optimization. The Tangent bundle neural network effectively predicted wear rates, closely aligning with experimental trends. Its predictions ranged from 0.2 × 10⁻⁴ mm³ N⁻¹ m⁻¹ to 2.5 × 10⁻⁴ mm³ N⁻¹ m⁻¹, demonstrating superior accuracy compared to other models such as Gradient Boosted Decision Trees, Random Forest, and Adaptive Boosting. Gradient boost decision trees had slight underestimations at elevated wear rates greater than 2.0 × 10⁻⁴ mm³ N⁻¹ m⁻¹, whereas random forest had mismatches in low wear rate areas. AdaBoost indicated consistent predictions but with small discrepancies at middle values. Thermal analysis revealed linear correspondence between energy density and peak temperature under different stress conditions. With increasing energy density from 0 to 60 J/cm², maximum temperature increased in all stress levels. At the highest energy density, maximum temperature merged at about 250 °C, showing thermal saturation. The results show the optimized laser processing conditions for enhancing zirconia-aluminum composites for better wear resistance and thermal stability.