Prediction of Effect of Reinforcement Size and Volume Fraction on the Abrasive Wear Behavior of AA2014/B4Cp MMCs Using Artificial Neural Network

ÇANAKÇI A., ÖZŞAHİN Ş., VAROL T.

ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING, cilt.39, sa.8, ss.6351-6361, 2014 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 39 Sayı: 8
  • Basım Tarihi: 2014
  • Doi Numarası: 10.1007/s13369-014-1157-9
  • Dergi Adı: ARABIAN JOURNAL FOR SCIENCE AND ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Aerospace Database, Communication Abstracts, Metadex, Pollution Abstracts, zbMATH, Civil Engineering Abstracts
  • Sayfa Sayıları: ss.6351-6361
  • Anahtar Kelimeler: Abrasive wear, Artificial neural networks, Metal matrix composites (MMCs), Stir casting, MATRIX COMPOSITES, SLIDING WEAR, MECHANICAL-PROPERTIES, B4C, TEMPERATURE, FABRICATION, RESISTANCE
  • Karadeniz Teknik Üniversitesi Adresli: Evet

Özet

In this study, an artificial neural network (ANN) approach was used to predict the abrasive wear behavior of AA2014 aluminum alloy matrix composites reinforced with B4C particles. The abrasive wear properties of varying volume fraction of particles up to 12 % B4C particle reinforced AA2014MMC(S) produced by stir casting method were investigated using a block-on-disc wear tester. Wear tests were performed under 92 N against the abrasive suspension mixture with a novel three body abrasive. For wear behavior, the volume loss, specific wear rate and surface roughness of the composites were measured. The effect of sliding time and content of B4C particles on the abrasive wear behavior were analyzed in detail. As a result of this study, the ANN was found to be successful for predicting the volume loss, specific wear rate and surface roughness of AA2014/B4C composites. The mean absolute percentage error (MAPE) for the predicted values did not exceed 4.1 %. The results have shown that ANN is an effective technique in the prediction of the properties of MMCs, and quite useful instead of time-consuming experimental processes.