Estimation of Tribological Behavior of Al2024-TiB₂ In Situ Composite Using GMDH and ANN

Abstract:

This paper reports on prediction and comparison of tribological behavior of Al2024-TiB₂ in-situ metal matrix composites using Group Method Data Handling Technique (GMDH) and Artificial Neural Network (ANN). Experiments were carried out using Pin-On-Disc type testing machine as per ASTM standards by varying the loads and sliding velocities. Two responses namely coefficient of friction and wear rates have been considered for each experiment. It was found that, ANN is the most reliable and accurate technique for prediction compared to GMDH.

• Periodical: Applied Mechanics and Materials (Volumes 592-594)
• Pages: 1310-1314
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.592-594.1310
• Online since: July 2014
• Authors: R. Keshavamurthy*, G. Ugrasen, R. Manasa, Narasimhe Gowda
• Keywords: Aluminum Composites, Artificial Neural Network (ANN), Friction, GMDH, Wear

References

[1] S. V. Prasad and R. Asthana, Aluminium metal-matrix composite for automotive applications: tribological considerations, Tribology Letters, Vol. 17, No. 3, (2004).

DOI: 10.1023/b:tril.0000044492.91991.f3

[2] Allison, J.E., and Cole, G.S., 1993, Metal matrix composite in the automotive industry: opportunities and challenges, JOM, January, p.19–24, (1993).

DOI: 10.1007/bf03223361

[3] Shouvik Ghosh, Prasanta Sahoo, and Goutam Sutradhar Friction Performance of Al-10%SiCp Reinforced Metal Matrix Composites Using Taguchi Method. Vol. 2013, Pp. 1-9. (2013).

DOI: 10.5402/2013/386861

[4] N. Radhika, R. Subramanian, S. Venkat Prasat, Tribological Behaviour of Aluminium/Alumina/Graphite Hybrid Metal Matrix Composite Using Taguchi's Techniques Journal of Minerals & Materials Characterization & Engineering, Vol. 10, pp.427-443, (2011).

DOI: 10.4236/jmmce.2011.105032

[5] J.T. Lin.D. Battacharya V. Kecman Multiple regression and neural networks analysis in composites machining, Composite science and technology 63, pp.539-548, (2003).

DOI: 10.1016/s0266-3538(02)00232-4

[6] Genel. K, Kurnaz S. C & Durman M. Modeling of tribological properties of alumina fiber reinforced zinc-aluminum composites using ANN. Materials Science and Engineering: A Vol. 363, 1–2, 20 (2003).

DOI: 10.1016/s0921-5093(03)00623-3

[7] K. Ravikumar, K.M. Mohansundaram, G. Aramai kkannu, & R Subramanian, Artificial neural networks based prediction of wear and frictional behaviour of aluminium (A380)–fly ash composites, Tribology, Vol. 6, No 1 pp.15-19, (2012).

DOI: 10.1179/1751584x11y.0000000025

[8] Sree Harsha Nandam, S. Sankaran and B.S. Murty, Precipitation kinetics in Al-Si-Mg/TiB2 in-situ composites Transactions of The Indian Institute of Metals Vol. 64, pp.123-126, (2011).

DOI: 10.1007/s12666-011-0024-6

[9] Stepashko V.S.: Theoretical aspects of GMDH as a method of inductive modelling. - Proceedings of the II International Conference on Inductive Modelling ICIM, (2008).

[10] S. Hayken, Neural Networks: A Comprehensive Foundation, Prentice Hall, New Jersey, (1999).

[11] Laurene V. Fausett, Fundamentals of Neural Networks: Architecture, Algorithms and Applications, Prentice Hall, pp.1-449, (1993).

[12] Jatinderpal Sharma, Shikha Chawla, Sunil Dalhotra A Research Agenda On Artificial Neural Network Topologies and Data Mining In Neural Network, International Journal Of Data & Network Security, Vol. 3 No. 1, p.41, (2013).