Integrating Genetic Algorithms, Particle Swarm, and Neural Networks for Wear Optimization of AA7178 Matrix with Nano-SiC Particles

• Published in: JOM (1989) Vol. 76; no. 6; pp. 2772 - 2785
• Main Authors: Bharat, Nikhil, Bose, P. S. C.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.06.2024
• Subjects: Alloy Microstructural Design for Wear Resistance; Chemistry/Food Science; Earth Sciences; Engineering; Environment; Physics
• ISSN: 1047-4838; 1543-1851
• DOI: 10.1007/s11837-023-06194-7

The most common method for producing composites at the nanoscale is the stir-casting process, and this study looks at the usage of silicon carbide nanoparticles as a reinforcing component in an AA7178 alloy matrix. Taguchi's design of experiments was used to examine data gathered from a pin-on-disc tribometer used to assess the composites' wear resistance. The worn surfaces were analyzed using SEM to identify the wear processes present. To accomplish the aims of minimizing wear on the composite samples and optimizing the wear test parameters, multiple linear regression analysis, particle swarm optimization, and the genetic algorithm were applied. To predict the composite material's wear rate and coefficient of friction over a range of test conditions, an artificial neural network wear model was also constructed. The sample with 3% SiC showed enhanced tribological capabilities, suggesting its potential use in a wide range of industrial settings, as shown by the results.