Modeling of the Prediction of Densification Behavior of Powder Metallurgy Al-Cu-Mg/B4C Composites Using Artificial Neural Networks


Creative Commons License

VAROL T. , ÇANAKÇI A. , ÖZŞAHİN Ş.

ACTA METALLURGICA SINICA-ENGLISH LETTERS, vol.28, no.2, pp.182-195, 2015 (Journal Indexed in SCI) identifier identifier

  • Publication Type: Article / Article
  • Volume: 28 Issue: 2
  • Publication Date: 2015
  • Doi Number: 10.1007/s40195-014-0184-6
  • Title of Journal : ACTA METALLURGICA SINICA-ENGLISH LETTERS
  • Page Numbers: pp.182-195
  • Keywords: Al alloys, Composite, Mechanical milling, Metal matrix composite, Artificial neural network, MECHANICAL-PROPERTIES, COMPRESSIBILITY, MICROSTRUCTURE, PARAMETERS, STRESS

Abstract

Al-CuMg/B4Cp metal matrix composites with reinforcement of up to 20 wt% were produced using the powder metallurgy technique. The effects of reinforcement ratio, reinforcement size, milling time, and compact pressure on the density and porosity of the composites reinforced with 0, 5, 10, and 20 wt% B4C particles were studied. Moreover, an artificial neural network model has been developed for the prediction of the effects of the manufacturing parameters on the density and porosity of powder metallurgy Al-Cu-Mg/B4Cp composites'. This model can be used for predicting the densification behavior of Al-Cu-Mg/B4Cp composites produced under reinforcement of different sizes and amounts with various milling times and compact pressures. The mean absolute percentage error for the predicted values did not exceed 1.6%.