Journals / Uludağ Üniversitesi Mühendislik Fakültesi Dergisi / 2020 / Cilt: 25 Sayı: 3

Prediction And Modelling Wear Resistance of Epoxy Matrix Composite Using Artificial Neural Network and Response Surface Design

TEPKİ YÜZEYİ TASARIMI VE YAPAY SİNİR AĞLARI YAKLAŞIMI UYGULANARAK EPOKSİ MATRİSLİ KOMPOZİT MALZEMENİN AŞINMA DAYANIMININ TAHMİNİ VE MODELLENMESİ

Pages
1325–1344
Publication Date
2020-12-31
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Mean square error (MSE) and R 2 statistics are also examined in order to explain variability in response variable and it is concluded that RSM yields better results which are 1.317 and %81.1, respectively.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Epoxy resin is a widely used material in various of industries especially construction, aviation and automative.
📊
Key Findings & Data:Factors that affect epoxy-based composite’s wear rate have been investigated and process optimization has been conducted in this paper.
💡
Conclusion & Impact:Mean square error (MSE) and R 2 statistics are also examined in order to explain variability in response variable and it is concluded that RSM yields better results which are 1.317 and %81.1, respectively.
🔬Design:Modeling & Simulation
📍Scope:54 sample
⚙️Variables:Tepki yüzeyi tasarımı · aşınma · epoksi matrisli kompozit malzeme

Abstract

Epoxy resin is a widely used material in various of industries especially construction, aviation and automative. Factors that affect epoxy-based composite’s wear rate have been investigated and process optimization has been conducted in this paper. In order to predict the effect of glass and ferrochromium reinforcement in wear resistance of epoxy, total number of 54 sample has been produced where design points are determined by Central Composite Design (CCD). After samples have been tested via wear test machine, results are compared with Artificial Neural Network (ANN) and Response Surface Methodology (RSM) wear predictions. Mean absolute percentage error (MAPE) shows that ANN (8.18%) outperforms RSM (9.42%) in terms of wear prediction accuracy. Mean square error (MSE) and R 2 statistics are also examined in order to explain variability in response variable and it is concluded that RSM yields better results which are 1.317 and %81.1, respectively. Besides, it is found that glass reinforcement results in decrease in wear rate. Minimum wear rate for small sized particle is obtained at level where glass and ferrochromium reinforcement rates are 17.07% and 2.93%, respectively. For large sized particles, minimum wear rate is obtained where both reinforcements are at rate 17.07%.

Özet

Yapılan çalışmada; inşaat, otomotiv ve havacılık gibi birçok sektörde geniş bir kullanım alanına sahip olan epoksi matrisli kompozit malzemenin aşınma davranışına etki eden faktörler incelenmiş olup, süreç optimizasyonu gerçekleştirilmiştir. Cam ve ferrokrom (karbür) katkı maddelerinin epoksi matrisli kompozit malzemenin aşınma dayanımına etkisini tahmin etmek için, Merkezi Birleşik Tasarım (MBT) uygulanarak toplam 18 deney noktasında 54 adet deney numunesi üretilmiştir. Üretilen numunelerin aşınma tepki değerleri ölçülerek Tepki Yüzeyleri Tasarımı (TYT) ve Yapay Sinir Ağları (YSA) aşınma tahmin modelleri oluşturulmuş ve bu modellerin tahmin performansı değerleri karşılaştırılmıştır. YSA yaklaşımının, sınama setinin aşınma oranı tahmininde ortalama yüzde hata değeri (MAPE) %8,18 olarak hesaplanmış olup, TYT yaklaşımının MAPE değeri %9,42 olarak bulunmuştur. Tepki değişkenindeki değişkenliğin açıklanmasında ve epoksi matrisli kompozit malzemenin aşınma davranışının tahmin edilmesinde R 2 ve ortalama kare hata (MSE) istatistikleri de incelenmiş olup, bu istatistiklerde MSE için 1,317 ve R 2 için %81,1 değerleri ile TYT yaklaşımının YSA yaklaşımına göre daha başarılı olduğu sonucuna ulaşılmıştır. Ayrıca, cam katkı oranının artması ile aşınma oranının büyük ölçüde azaldığı görülmüştür. Minimum aşınma oranı; küçük parçacıklarda cam ve ferrokrom katkı oranının sırasıyla %17,07 ve %2,93 olduğu, büyük parçacıklarda iki katkı oranının da %17,07 olduğu durumda elde edilmiştir.

🔖

Cite This Article

Karakurt, N. F., Sağbaş, A. (2020). Prediction And Modelling Wear Resistance of Epoxy Matrix Composite Using Artificial Neural Network and Response Surface Design. Uludağ Üniversitesi Mühendislik Fakültesi Dergisi, Cilt: 25 Sayı: 3, 1325–1344. https://doi.org/10.17482/uumfd.711221

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (25)

  1. Basavarajappa, S., Yadav, S.M., Kumar, S., Arun, K.V. ve Narendranath, S. (2011) Abrasive Wear Behavior of Granite-Filled Glass-Epoxy Composites by SiC Particles Using Statistical Analysis, Polymer-Plastics Technology and Engineering, 50, 516-524. doi: 10.1080/03602559.2010.543734
  2. Bonner, W.H. (1962) Aromatic polyketones and preparation thereof, U.S. Patent Specification.
  3. Briscoe, B., Pogosian, A. ve Tabor, D. (1974) The friction and wear of high density polythene: The action of lead oxide and copper oxide fillers, Wear, 27, 19-34. doi: 10.1016/0043-1648(74)90081-7
  4. Dadrasi, A., Fooladpanjeh, S. ve Gharahbagh, A.A. (2019) Interactions between HA/GO/epoxy resin nanocomposites: optimization, modeling and mechanical performance using central composite design and genetic algorithm, Journal of the Brazilian Society of Mechanical Sciences and Engineering, 41(2). doi: 10.1007/s40430-019-1564-7
  5. Diler, E. A. ve Ipek, R. (2013) Main and interaction effects of matrix particle size, reinforcement particle size and volume fraction on wear characteristics of Al–SiCp composites using central composite design, Composites Part B: Engineering, 50, 371–380. doi: 10.1016/j.compositesb.2013.02.001
  6. Genel, K., Kurnaz, S., & Durman, M. (2003) Modeling of tribological properties of alumina fiber reinforced zinc–aluminum composites using artificial neural network, Materials Science and Engineering: A, 363(1-2), 203–210. doi: 10.1016/s0921-5093(03)00623-3
  7. Ghasemi, A. ve Zahediasl, S. (2012) Normality Tests for Statistical Analysis: A Guide for Non-Statisticians, Int J Endocrinol Metab, 10, 486-489. doi: 10.5812/ijem.3505
  8. Jiang, Z., Zhang, Z. ve Friedrich, K. (2007) Prediction on wear properties of polymer composites with artificial neural networks, Composites Science and Technology, 67(2), 168–176. doi: 10.1016/j.compscitech.2006.07.026
  9. Khuri, A.I. ve Mukhopadhyay, S. (2010) Response surface methodology, Wiley Interdisciplinary Reviews: Computational Statistics, 2, 128-149. doi: 10.1002/wics.73
  10. Kranthi, G. ve Satapathy, A., (2010) Evaluation and prediction of wear response of pine wood dust filled epoxy composites using neural computation, Computational Materials Science, 49, 609–614. doi: 10.1016/j.commatsci.2010.06.001
  11. Kumar, D.S. ve Rajmohan, M. (2018) Optimizing Wear Behavior of Epoxy Composites Using Response Surface Methodology and Artificial Neural Networks, Polymer Composites, 40(7), 2812–2818. doi: 10.1002/pc.25089
  12. Mohan, N., Natarajan, S. ve Kumareshbabu, S. (2011) Abrasive wear behaviour of hard powders filled glass fabric–epoxy hybrid composites, Materials & Design, 32, 1704-1709. doi: 10.1016/j.matdes.2010.08.050
  13. Montgomery, D.C. (2001) Design and Analysis of Experiment. John Wiley and Sons Inc, 12-13, ABD.
  14. Öztemel, E. (2012) Yapay sinir ağları, Papatya yayıncılık, İstanbul.
  15. Padhi, P.K. ve Satapathy, A. (2014) Solid Particle Erosion Behavior of BFS-Filled Epoxy–SGF Composites Using Taguchis Experimental Design and ANN. Tribology Transactions, 57, 396-407. doi: 10.1080/10402004.2013.877178
  16. Pati, P. R. ve Satapathy, A. (2014) Prediction and simulation of wear response of Linz-Donawitz (LD) slag filled glass-epoxy composites using neural computation, Polymers for Advanced Technologies, 26(2), 121–127. doi: 10.1002/pat.3421
  17. Rashed, F. ve Mahmoud, T. (2009) Prediction of wear behaviour of A356/SiCp MMCs using neural networks, Tribology International, 42(5), 642–648. doi: 10.1016/j.triboint.2008.08.010
  18. Rothon, R.N. (1997) Mineral Fillers in Thermoplastics: Filler Manufacture, Journal of Adhesion, 64, 87–109. doi:10.1007/3-540-69220-7_2
  19. Rout, A. ve Satapathy, A. (2012) Analysis of Dry Sliding Wear Behaviour of Rice Husk Filled Epoxy Composites Using Design of Experiment and ANN, Procedia Engineering, 38, 1218-1232. doi: 10.1016/j.proeng.2012.06.153
  20. Siddhartha, S., Patnaik, A. ve Bhatt, A.D. (2011) Mechanical and dry sliding wear characterization of epoxy–TiO2 particulate filled functionally graded composites materials using Taguchi design of experiment, Materials & Design, 32, 615-627. doi:10.1016/j.matdes.2010.08.011
  21. Suresha, B., Chandramohan, G., Prakash, J.N., Balusamy, V. ve Sankaranarayanasamy, K. (2006) The Role of Fillers on Friction and Slide Wear Characteristics in Glass-Epoxy Composite Systems, Journal of Minerals and Materials Characterization and Engineering, 5, 87-101. doi: 10.4236/jmmce.2006.51006
  22. Varol, T., Canakci, A. ve Ozsahin, S. (2013) Artificial neural network modeling to effect of reinforcement properties on the physical and mechanical properties of Al2024–B4C composites produced by powder metallurgy, Composites Part B: Engineering, 54, 224–233. doi: 10.1016/j.compositesb.2013.05.015
  23. Witt, S.F. ve Witt, C.A. (1992) Modeling and forecasting demand in tourism, Academic, Londra.
  24. Zhang, Z., Breidt, C., Chang, L., Haupert, F. ve Friedrich, K. (2004), Enhancement of the wear resistance of epoxy, Composites: Part A, 35, 1385-1392. doi: 10.1016/j.compositesa.2004.05.005
  25. Zhang, Z., Friedrich, K. ve Velten, K. (2002) Prediction on tribological properties of short fibre composites using artificial neural networks, Wear, 252(7-8), 668–675. doi: 10.1016/s0043-1648(02)00023-6