Adaptive Neural Network for Estimation of Sliding Wear Behaviour of Al6061-Carbon Fiber Composites

Abstract:

In recent years, Al6061-carbon fiber composites are gaining wide spread popularity as they find scope in certain high-tech applications such as automobile, aerospace, transport, andprocessing industries. Thesecomposites possess high strength to weight ratio, excellent wear resistance in addition to superior mechanical properties. The experimental method of determining wear phenomenon of the developed composites is an expensive as well as a tedious process.As such engineers and scientists are focusing their attention towards developing mathematical modelsfor determining wear phenomenon. The use of mathematical modeling for prediction of wear phenomenon is an evolving research area. Hence, meager information is available as regards the mathematical model to determine wear rate of composites. Mathematical modeling is slowly gaining impetus in industries in order to assess the life of sliding components and establishing the economic loss incurred due to the wear phenomenon. In the light of the above, Al6061 carbon fiber composites were prepared by liquid metallurgical route and then machined to a standard size of pin.On the pins, sliding wear test was conducted on a pin-on-disc apparatus using C-45 steel disc as per ASTM Standard. Data generated was then used in developing AdaptiveNeuro Fuzzy Inference System (ANFIS). The ANFIS logic was created using the fuzzy logic tool box of Matlab 7.10 Version. For simulating, actual working conditions used to establish the sliding wear behaviour of Al6061-xwt%Carbon fiber composites (x=5, 10) including variable parameters such as Varying load (from 10-60N in step of 10N), Sliding distance, Weight fraction (5-10%) keeping other parameters constant such as track diameter 20 mm, Speed 500rpm and Pin diameter 8 mm were used.The adopted fuzzy model employs hybrid learning techniques for updating the premise and consequent parameter. The predicted values of sliding wear rate of Al6061-xwt% carbon fiber are in close agreement with the experimental results.

• Periodical: Applied Mechanics and Materials (Volumes 592-594)
• Pages: 1267-1271
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.592-594.1267
• Online since: July 2014
• Authors: Chinnakurli Suryanarayana Ramesh, Nirupama Mohan*, Vikram K.S. Jain, Harsha R. Gudi
• Keywords: ANFIS, Hybrid Composite, MATLAB, Sliding Wear

References

[1] M.A. Hassan, M.A., M. A, Elsharief, A. Aboul Kasem, S Ramesh, Purbolaksono J, A fuzzy model for evaluation and prediction of slurry erosion of 5127 steels, Materials & Design, 39 (2012) pp.186-191.

DOI: 10.1016/j.matdes.2012.02.012

[2] N. Ghosh,Y. BRavi,A. Patra, S. Mukhopadhyay, S. Paul.A. RMohanty, A.B. Chattopadhyay, Estimation of tool wear during CNC milling using neural network-based sensorfusion. Mech. Syst. Signal Process,. 21, 466–479.

DOI: 10.1016/j.ymssp.2005.10.010

[3] P. Bhattacharyya, D. Sengupta, , S. Mukhopadhyay, Cuttingforce-based real-time estimation of tool wear in face millingusing a combination of signal processing techniques, Mech. Syst. Signal Process, 21, 2007, 2665–2683.

DOI: 10.1016/j.ymssp.2007.01.004

[4] C. S. Ramesh, V. K. S. Jain, R. Keshavamurthy, Z. A. Khan& M. Hadfield, Prediction of slurry erosive wear behaviour of Al6061 alloy using a fuzzy logic approach, 78, 109-119, (2013).

DOI: 10.2495/secm130091

[5] Y. Blake, Sliding wear prediction of ductile materials, Journal of Materials Processing Technology, 118, (2001), p.205–211.

DOI: 10.1016/s0924-0136(01)00918-9

[6] C. S Ramesh, Abrar Ahmed, Prediction of wear co-efficient of Al6063-TiB2 Insitu composites, Proceeding of National Tribology Conference-2011, Dec 8-10, (2011).

[7] Jyh-Shing Roger Jang, Adaptive-Network-Based Fuzzy inference System, IEEE Transactions of Systems, Man and Cybernetics, Vol. 23, No. 3 (1993), pp.665-685.

DOI: 10.1109/21.256541

[8] Information on http: /www. mathworks. com.