Universiti Teknologi Malaysia Institutional Repository

Fuzzy rule-based model to estimate surface roughness and wear in hard coatings

Jaya, A.S.M. and Hashim, S.Z.M. and Haron, H. and Muhamad, M.R. and Rahman, M.N.A. (2012) Fuzzy rule-based model to estimate surface roughness and wear in hard coatings. Conference Proceedings - IEEE International Conference on Systems, Man and Cybernetics . ISSN 1062-922X

Full text not available from this repository.

Official URL: http://dx.doi.org/10.1109/ICSMC.2012.6377873

Abstract

n this paper, a new approach in predicting the surface roughness and flank wear of hard coatings using fuzzy rule-based model is implemented. Hard coatings is important for cutting tool due to its excellent performances in 800°C temperature during high speed machining. The coating process were run using Physical Vapor Deposition (PVD) magnetron sputtering process. An experiment matrix called Response Surface Methodology (RSM) was used to collect data based on optimized data point. Sputtering power, substrate bias voltage and substrate temperature were used as the variables, and coating roughness and flank wear as the output responses of the coating process. The collected experimental data were used to develop fuzzy rules. Five triangular membership functions (MFs) for input variables and nine MFs for output responses were used in constructing the models. The results of fuzzy rule-based models were compared against the experimental result based on the percentage error, co-efficient determination (R2) and model accuracy. The rule-based model for coating roughness showed an excellent result with respective smallest percentage error, R2 and model accuracy were 0.85%, 0.953 and 89.20% respectively. Meanwhile, the fuzzy flank wear model indicated 6.38%, 0.91 and 81.79% for smallest percentage error, R2 and model accuracy. Thus, fuzzy logic can be a good alternative in predicting coating roughness and flank wear in hard coatings.

Item Type:Article
Uncontrolled Keywords:Cybernetics
Subjects:Q Science
Divisions:Others
ID Code:47017
Deposited By: Haliza Zainal
Deposited On:22 Jun 2015 13:56
Last Modified:06 Aug 2017 11:47

Repository Staff Only: item control page