Hardness Optimization for Al6061-MWCNT Nanocomposite Prepared by Mechanical Alloying Using Artificial Neural Networks and Genetic Algorithm

Abstract

Among artificial intelligence approaches, artificial neural networks (ANNs) and genetic algorithm (GA) are widely applied for modification of materials property in engineering science in large scale modeling. In this work artificial neural network (ANN) and genetic algorithm (GA) were applied to find the optimal conditions for achieving the maximum hardness of Al6061 reinforced by multiwall carbon nanotubes (MWCNTs) through modeling of nanocomposite characteristics. After examination the different ANN architectures an optimal structure of the model, i.e. 6-18-1, is obtained with 1.52% mean absolute error and R2 = 0.987. The proposed structure was used as fitting function for genetic algorithm. The results of GA simulation predicted that the combination sintering temperature 346 °C, sintering time 0.33 h, compact pressure 284.82 MPa, milling time 19.66 h and vial speed 310.5 rpm give the optimum hardness, (i.e., 87.5 micro Vickers) in the composite with 0.53 wt% CNT. Also, sensitivity analysis shows that the sintering time, milling time, compact pressure, vial speed and amount of MWCNT are the significant parameter and sintering time is the most important parameter. Comparison of the predicted values with the experimental data revealed that the GA–ANN model is a powerful method to find the optimal conditions for preparing of Al6061-MWCNT.