Statistical models for predicting wear and friction coefficient of palm kernel activated carbon-epoxy composite using the ANOVA
Abstract
Purpose
The purpose of this study was to propose statistical models for predicting wear and friction coefficient of the palm kernel activated carbon-epoxy composite using the analysis of variance (ANOVA).
Design/methodology/approach
All the specimens were formed into 10-mm diameter pins of 30-mm length each. The tribological test was conducted using a pin-on-disc tribometer. The results of the coefficient of friction (COF) and the wear rate were then analysed using the ANOVA. Regression analysis was used to derive the predictive equations for both friction coefficient and wear rate.
Findings
It was found that the most significant parameter that affects the COF is the weight composition, while for the wear rate, it is the operating temperature. The proposed statistical models have 90-94 per cent reliability.
Research limitations/implications
The equation models are only limited within the tested parameters and ranges in the plastic deformation region.
Originality/value
These models can be very useful for material design engineers in avoiding the component failures occurring prematurely.
Keywords
Acknowledgements
The authors are grateful for the contributions by members of the Green Tribology and Engine Performance (G-TriboE) research group. This research was funded by a grant from the Ministry of Higher Education Malaysia (grant no. ERGS/2013/FKM/TK01/UTEM/02/04/E00016).
Citation
Mat Tahir, N.A., Abdollah, M.F.B., Hasan, R., Amiruddin, H. and Abdullah, M.I.H.C. (2017), "Statistical models for predicting wear and friction coefficient of palm kernel activated carbon-epoxy composite using the ANOVA", Industrial Lubrication and Tribology, Vol. 69 No. 5, pp. 761-767. https://doi.org/10.1108/ILT-02-2016-0031
Publisher
:Emerald Publishing Limited
Copyright © 2017, Emerald Publishing Limited