Optimizing Aluminum Metal Matrix Composites with SiC Nanoparticles using Taguchi-ANN Approach for Enhanced Mechanical Performance

Abstract

The current research explores the optimization of Silicon Carbide particle-reinforced aluminum metal matrix composites to improve mechanical properties. An integrated method based on Taguchi Design of Experiment and Artificial Neural Network has been adopted, with the novel approach to explore the optimal combination of parameters. The obtained best set includes the minimum load of 30 N, the minimum speed of 100 rpm, and the larger composition of 9% SiC particle. The designed L9 orthogonal experimental plan was used to conduct the experiments, and the findings explicitly indicated the significant impacts on the reduction of specific wear rate and friction force . Furthermore, the Artificial Neural Network trained through the backpropagation algorithm estimated all the percentages correctly to the ideal combination, equivalent to 100% in predicting the target responses. Moreover, the confirmation experience has validated the optimal combination, as it approaches specific wear rate of 0.0019, and friction force was 10.5. These results highlight the role of the integrated research approach for assessing the optimal parameters of aluminum MMCs to the required mechanical properties. Consequently, the current study highlights the importance of experimental plan integration and predictive modeling for optimizing materials, and it applies to various engineering fields where wear resistance and friction performance are critical.