Prediction of effect of reinforcement content, flake size and flake time on the density and hardness of flake AA2024-SiC nanocomposites using neural networks


VAROL T., ÇANAKÇI A., ÖZŞAHİN Ş.

JOURNAL OF ALLOYS AND COMPOUNDS, vol.739, pp.1005-1014, 2018 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 739
  • Publication Date: 2018
  • Doi Number: 10.1016/j.jallcom.2017.12.256
  • Journal Name: JOURNAL OF ALLOYS AND COMPOUNDS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.1005-1014
  • Keywords: Artificial neural network, Flake composites, Mechanical milling, AA2024, SiC, METAL-MATRIX COMPOSITES, POWDER-METALLURGY, MECHANICAL-PROPERTIES, WEAR BEHAVIOR, INFILTRATION TECHNIQUE, ALUMINUM-ALLOY, MICROSTRUCTURE, AL, COMPRESSIBILITY, FABRICATION
  • Karadeniz Technical University Affiliated: Yes

Abstract

It is known that the size and morphology of matrix powders and reinforcement content affects the physical and mechanical properties of ceramic nanoparticle reinforced metal matrix composites. Therefore, an artificial neural network model has been developed for the prediction of density and hardness of AA2024-SiC nanocomposites fabricated by flake powder metallurgy. The weight percentage of SiC nanoparticles, AA2024 matrix size and ball milling time were selected as the inputs and the sintered density and hardness of the flake AA2024-SiC nanocomposites as the output of the model. Prediction model ofAA2024-SiC nanocompositeswas able to predict the density and hardness with a mean absolute percentage error (MAPE) of 0.18% and 0.99%, respectively. The root mean score error (RMSE) of ANN model developed for AA2024-SiC nanocomposites were 0.06% and 0.835 for the density and hardness, respectively. The results of present study shows that the density and hardness of SiC nanoparticle reinforced AA2024 matrix flake nanocomposites can be predicted with high accuracy using neural network model. (C) 2017 Elsevier B.V. All rights reserved.