Artificial Neural Networks for Predicting the Tribological Behaviour of Al7075-SiC Metal Matrix Composites

G B, VEERESH KUMAR and PRAMOD, R (2014) Artificial Neural Networks for Predicting the Tribological Behaviour of Al7075-SiC Metal Matrix Composites. In: International Conference on Advances In Engineering And Technology - ICAET 2014, 24 - 25 May, 2014, RIT, Roorkee, India.

[img]
Preview
Text
20140726_105014.pdf - Published Version

Download (971kB) | Preview
Official URL: https://www.seekdl.org/conferences/paper/details/3...

Abstract

Wear is a complex phenomenon and the most important reason for the damage and consequent failure of machine parts. The dry sliding wear performance of particulate reinforced aluminum metal matrix composites (Al-MMCs) are being investigated by numerous researchers during the last three decades. A lot of experiments have to be conducted in order to study the wear behavior. Artificial Neural Networks help in reducing the cost of experimentation when implemented with care and enough data. In this work, wear behavior of Al7075 aluminium matrix composites reinforced with particulates has been investigated. The Al7075-SiC metal matrix composites were fabricated through liquid metallurgy process containing 2-6 wt% SiC. The dry sliding wear tests were conducted using computerized pin on disc wear testing machine. The wear properties of the composites containing SiC were better than that of the matrix material and further, the composite containing 6 wt% SiC content exhibited superior wear resistance. An ANN model was developed to predict the tribological properties of the Al7075-SiC composites. The predicted values of tribological properties using a well trained ANN were found in good agreement with measured values.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Metal Matrix Composites, Sliding Wear, Artificial Neural Networks.
Depositing User: Mr. John Steve
Date Deposited: 25 May 2019 12:28
Last Modified: 25 May 2019 12:28
URI: http://publications.theired.org/id/eprint/2684

Actions (login required)

View Item View Item