Multi objective optimization of novel Al-Si-Mg nanocomposites: A Taguchi-ANN-NSGA-II Approach

Environmental pollution is a big problem all over the world. One of the strategies to reduce the menace environmental pollution is the development of the concept of circular economy whose major thrust is the concept of conversion of waste to wealth. The aim of this research is to optimize the production of a new nanocomposite material developed from Al-Si-Mg Alloy derived from waste beverage cans, carbon nanotubes (CNTs) derived from rice husks and nanoparticles from periwinkle shells. The multi-objective optimization of the porosity, hardness and compressive strength of the novel composite was done using the Taguchi method, Artificial Neural Network (ANN), and Nondominated Sorting Genetic Algorithm-II (NSGA-II). The ANN utilized the following input parameters: Weight percentage (wt%) of CNTs, Weight percentage (wt%) of PWSnp, stirring speed (rpm) and Stirring time (minutes). The outputs predicted by the ANN were: Porosity, Hardness, Compressive strength. To achieve optimal porosity, hardness, and compressive strength, the Taguchi-grey relational methodology was employed to simultaneously optimize the production parameters of the composite. The optimal values determined were as follows: 1.5 wt% of CNTs, 1.0 wt% of PWSnp, 100 rpm stirring speed, and 6 min of stirring time with porosity, hardness, and compressive strength values of 0.3250, 108.2738 and 410.6436 respectively. The ANN demonstrated excellent predictive capability, exhibiting correlation coefficients of 0.9617, 0.9536, and 0.9725 for the porosity, hardness, and compressive strength of the composite material, respectively. Next, the ANN was utilized as a fitness function within NSGA-II to perform multi-objective optimization for the porosity, hardness, and compressive strength of the novel material. The resulting Pareto optimal solutions and the optimum production parameters serve as a valuable guide for engineers involved in the design of optimal brake discs and other machine components, utilizing the advantageous properties of the new material.