Response surface methodology and machine learning optimisations comparisons of recycled AA6061-B4C–ZrO2 hybrid metal matrix composites via hot forging forming process

Abstract

The optimal conditions of applied factors to reuse Aluminium AA6061 scraps are (450, 500, and 550) ⁰C preheating temperature, (1–15) % Boron Carbide (B4C), and Zirconium (ZrO2) hybrid reinforced particles at 120 min forging time via Hot Forging (HF) process. The response surface methodology (RSM) and machine learning (ML) were established for the optimisations and comparisons towards materials strength structure. The Ultimate Tensile Strength (UTS) strength and Microhardness (MH) were significantly increased by increasing the processed temperature and reinforced particles because of the material dispersion strengthening. The high melting point of particles caused impedance movements of aluminium ceramics dislocations which need higher plastic deformation force and hence increased the material's mechanical and physical properties. But, beyond Al/10 % B4C + 10 % ZrO2 the strength and hardness were decreased due to more particle agglomeration distribution. The optimisation tools of both RSM and ML show high agreement between the reported results of applied parameters towards the materials' strength characterisation. The microstructure analysis of Field Emission Scanning Electron Microscopy (FE-SEM) and Atomic Force Microscope (AFM) provides insights mapping behavioural characterisation supports related to strength and hardness properties. The distribution of different volumes of ceramic particle proportion was highlighted. The environmental impacts were also analysed by employing a life cycle assessment (LCA) to identify energy savings because of its fewer processing steps and produce excellent hybrid materials properties.

Keywords

• Solid-state recycling (SSR)
• Hot forging (HF)
• Hybrid materials (HM)
• Life cycle assessment (LCA)
• Response surface methodology (RSM)
• Machine learning (ML)