Prediction of Tribological Properties of Micro-Particle-Filled Composites Using Artificial Neural Network

Year 2025, Volume: 6 Issue: 1, 68 - 82, 19.06.2025

Hüseyin Gürbüz , Mehmet Emin Demir , Şehmus Baday , İbrahim Halil Akcan

https://doi.org/10.55546/jmm.1581662

Abstract

Artificial neural network (ANN) is one of the effective techniques used to solve complex engineering problems such as tribological performance of polymer composites. The aim of this study is to model the wear performance of boron carbide (B4C) and graphite (Gr) filled epoxy matrix composite material with ANN. Composites with 5 wt.%, 10 wt. % and 15 wt. % B4C and Gr fillers in glass fiber reinforced epoxy resin were prepared by simple hand lay-up technique. Wear tests were performed on these composites under dry sliding conditions according to Taguchi's orthogonal array design. Using experimental data, an ANN model was trained and tested to predict the effect of various control factors on wear behavior. In the created ANN model, the network structure is feed forward and back propagation, Levenberg–Marquardt training algorithm, tansig transfer function is used to estimate experimental results. The values of the training, validation, testing and general regression coefficients for coefficient of friction (COF) were 0.9936 – 0.99996 – 0.99807 and 0.9911, respectively, while for the wear rate, they were 0.9968 – 0.99891 – 0.83971 and 0.93886. According to the regression coefficient values obtained from the ANN model created for the wear rate and COF values, the experimental results were consistent with high accuracy rates.

Keywords

Micro Particle , Glass Fiber , Composite , Wear , Artificial Neural Networks

Project Number

BTÜBAP-2021-YL-020

Yapay Sinir Ağı ile Mikro Parçacık Dolgulu Kompozitlerin Tribolojik Özelliklerinin Tahmini

Abstract

Yapay sinir ağı (YSA), polimer kompozitlerin tribolojik performansı gibi karmaşık mühendislik problemlerini çözmek için kullanılan etkili tekniklerden biridir. Bu çalışmanın amacı bor karbür (B4C) ve grafit (Gr) dolgulu epoksi matrisli kompozit malzemenin aşınma performansını YSA ile modellemektir. Cam elyaf takviyeli epoksi reçine içerisinde ağırlıkça %5, %10 ve %15 oranında B4C ve Gr dolgulu kompozitler basit elle yatırma tekniği ile hazırlanmıştır. Bu kompozitlere, Taguchi'nin ortogonal dizi tasarımına göre kuru kayma koşullarında aşınma deneyleri gerçekleştirilmiştir. Deneysel veriler kullanılarak, aşınma davranışı üzerinde çeşitli kontrol faktörlerinin etkisini tahmin etmek amacıyla bir YSA modeli eğitilmiş ve test edilmiştir. Oluşturulan YSA modelinde, ağ yapısı ileri beslemeli ve geri yayılımlı, eğitim algoritması Levenberg–Marquardt, transfer fonksiyonu tansig kullanarak deneysel sonuçlar tahmin edilmiştir. Sürtünme katsayısı (COF) için eğitim, doğrulama, test ve genel regresyon katsayı değerleri sırasıyla 0,9936 – 0,99996 – 0,99807 ve 0,9911 iken, aşınma oranı için ise 0,9968 – 0,99891 – 0,83971 ve 0,93886 olarak elde edilmiştir. Aşınma oranı ve COF değerleri için oluşturulan YSA modelinden elde edilen regresyon katsayısı değerlerine göre, deneysel sonuçların yüksek doğruluk oranlarıyla tutarlı olduğu görülmüştür.

Keywords

Mikro Parçacık , Cam Elyaf , Kompozit , Aşınma , Yapay Sinir Ağları

Supporting Institution

Bu çalışma, Batman Üniversitesi Bilimsel Araştırma Projeleri Koordinasyon Birimi tarafından “BTÜBAP-2021-YL-020” kodlu proje ile desteklenmiştir.

Project Number

BTÜBAP-2021-YL-020

Thanks

Bu çalışma, Batman Üniversitesi Bilimsel Araştırma Projeleri Koordinasyon Birimi tarafından “BTÜBAP-2021-YL-020” kodlu proje ile desteklenmiştir.

References

• Abenojar J., Martínez M. A., Velasco F., Pascual-Sánchez V., Martín-Martínez J. M., Effect of Boron Carbide Filler on the Curing and Mechanical Properties of an Epoxy Resin. The Journal of Adhesion 85, 216-238, 2009.
• Agrawal S., Singh K. K., Sarkar P. K., A comparative study of wear and friction characteristics of glass fibre reinforced epoxy resin, sliding under dry, oil-lubricated and inert gas environments. Tribology International 96, 217-224, 2016.
• Arun A., Kumar Singh K., Friction and Wear Behaviour of Glass Fibre Reinforced Polymer Composite (GFRP) under Dry and Oil Lubricated Environmental Conditions. Materials Today: Proceedings 4, 7285-7292, 2017.
• Chowdhury M. A., Helali M., The effect of amplitude of vibration on the coefficient of friction for different materials. Tribology International 41, 307-314, 2008.
• Chowdhury M. A., Shuvho B. A., Debnath U. K., Nuruzzaman D. M., Prediction and Optimization of Erosion Rate of Carbon Fiber–Reinforced Ebonite Using Fuzzy Logic. Journal of Testing and Evaluation 47, 1244-1258, 2019.
• Ciurana J., Quintana G., Garcia-Romeu M. L., Estimating the cost of vertical high-speed machining centres, a comparison between multiple regression analysis and the neural networks approach. International Journal of Production Economics 115, 171-178, 2008.
• Çetkin E., Demir M. E., Ergün R. K., The effect of different fillers, loads, and sliding distance on adhesive wear in woven e-glass fabric composites. Proceedings of the Institution of Mechanical Engineers, Part E: Journal of Process Mechanical Engineering:095440892211368, 2022.
• Demir M. E., Cetkin E., Ergün R. K., Denizhan O., Tribological and mechanical properties of nanofilled glass fiber reinforced composites and analyzing the tribological behavior using artificial neural networks. Polymer Composites 45, 4233-4249, 2024.
• Demir M. E., Çelik Y. H., Kilickap E., Effect of matrix material and orientation angle on tensile and tribological behavior of jute reinforced composites. Materials Testing, 61(8), 806-812, 2019.
• Demir M. E., Investigation of The Abrasive Wear Behavior of GFRC And CFRC with Different Parameters Using Taguchi And Artificial Neural Networks Method. Politeknik Dergisi 28(1), 215-228, 2024.
• Divya G. S., Keshavamurthy R., Siddaraju C., Murthy K. V. S., Investigation on Sliding Wear Properties of Nano Metallic Particle Reinforced Hybrid Composites Through Design of Experiments and ANN. Journal of The Institution of Engineers (India): Series D 105(3), 1551-1562, 2023.
• Findik F., Yilmaz R., Köksal T., Investigation of mechanical and physical properties of several industrial rubbers. Materials & Design 25, 269-276, 2004.
• Friedrich K., Polymer composites for tribological applications. Advanced Industrial and Engineering Polymer Research 1, 3-39, 2018.
• Gürbüz H., Akcan İ. H., Baday Ş., Demir M. E., Investigation of Drilling Performances, Tribological and Mechanical Behaviors of GFRC Filled with B4C and Gr. Arabian Journal for Science and Engineering 1-18, 2024.
• Gyurova L. A., Friedrich K., Artificial neural networks for predicting sliding friction and wear properties of polyphenylene sulfide composites. Tribology International 44, 603-609, 2011.
• Jayan J. S., Appukuttan S., Wilson R., Joseph K., George G., Oksman K., An introduction to fiber reinforced composite materials. Woodhead Publishing Series in Composites Science and Engineering. (ed by K Joseph, K Oksman, G George, R Wilson & SBT-FRC Appukuttan) Woodhead Publishing, pp 1–24, 2021.
• Jiang Z., Zhang Z., Friedrich K., Prediction on wear properties of polymer composites with artificial neural networks. Composites Science and Technology 67, 168-176, 2007.
• Kim S. S., Shin M. W., Jang H., Tribological properties of short glass fiber reinforced polyamide 12 sliding on medium carbon steel. Wear 274, 34-42, 2012.
• Kose H., Bayar I., Ergün R. K., Experimental optimization of CuO and MgO hybrid nanoparticle reinforcement ratios to enhance fatigue life of GFRP composites. Polymer Composites 45, 11125-11137, 2024.
• Kranthi G., Satapathy A., Evaluation and prediction of wear response of pine wood dust filled epoxy composites using neural computation. Computational Materials Science 49, 609-614, 2010.
• Kumar S., Priyadarshan, Ghosh S. K., Statistical and artificial neural network technique for prediction of performance in AlSi10Mg-MWCNT based composite materials. Materials Chemistry and Physics 273, 125136, 2021a.
• Kumar S., Priyadarshan, Ghosh S. K., Statistical and computational analysis of an environment-friendly MWCNT/NiSO4 composite materials. Journal of Manufacturing Processes 66, 11-26, 2021b.
• Kumar S., Singh K. K., Tribological behaviour of fibre-reinforced thermoset polymer composites: A review. Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials: Design and Applications 234, 1439-1449, 2020.
• Li Y., Zhang H., Porwal H., Huang Z., Bilotti E., Peijs T., Mechanical, electrical and thermal properties of in-situ exfoliated graphene/epoxy nanocomposites. Composites Part A: Applied Science and Manufacturing 95, 229-236, 2017.
• Negi A. S., Katiyar J. K., Kumar S., Kumar N., Patel V. K., Physicomechanical and abrasive wear properties of hemp/Kevlar/carbon reinforced hybrid epoxy composites. Materials Research Express 6, 115304, 2019.
• Ouyang J. H., Li Y. F., Zhang Y. Z., Wang Y. M., Wang Y. J., High-Temperature Solid Lubricants and Self-Lubricating Composites: A Critical Review. Lubricants 10(8), 177, 2022.
• Padhi P. K., Satapathy A., Analysis of Sliding Wear Characteristics of BFS Filled Composites Using an Experimental Design Approach Integrated with ANN. Tribology Transactions 56, 789-796, 2013.
• Pati P. R., Prediction and wear performance of red brick dust filled glass–epoxy composites using neural networks. International Journal of Plastics Technology 23, 253-260, 2019.
• Ray S., Parametric Optimization and Prediction of Abrasion Wear Behavior of Marble-Particle-Filled Glass–Epoxy Composites Using Taguchi Design Integrated with Neural Network. JOM 73, 2050-2059, 2021.
• Rodrigues D. D., Broughton J. G., Silane surface modification of boron carbide in epoxy composites. International Journal of Adhesion and Adhesives 46, 62-73, 2013.
• Shtub A., Versano R., Estimating the cost of steel pipe bending, a comparison between neural networks and regression analysis. International Journal of Production Economics 62, 201-207, 1999.
• Singh K. K, Ansari M. T. A., Azam M. S., Fatigue life and damage evolution in woven GFRP angle ply laminates. International Journal of Fatigue 142, 105964, 2021.
• Subbaya K. M., Suresha B., Rajendra N., Varadarajan Y. S., Taguchi approach for characterization of three-body abrasive wear of carbon-epoxy composite with and without SiC filler. Composite Interfaces 19, 297-311, 2012.
• Sudheer M., Hemanth K., Raju K., Bhat T., Enhanced Mechanical and Wear Performance of Epoxy/glass Composites with PTW/Graphite Hybrid Fillers. Procedia Materials Science 6, 975-987, 2014.
• Suresha B., Chandramohan G., Sadananda Rao P. R., Sampathkumaran P., Seetharamu S., Influence of SiC filler on mechanical and tribological behavior of glass fabric reinforced epoxy composite systems. Journal of Reinforced Plastics and Composites 26(6), 565-578, 2007.
• Teli G., Mahakur V. K., Paul R. Bhowmik S., Investigation of Dry Sliding Tribological Behaviour of Epoxy Composites Filled with Hemp Particulates Using Artificial Neural Networks. Arabian Journal for Science and Engineering 48, 3989-4001, 2023.
• Thakur R. K. Singh K. K., Abrasive waterjet machining of fiber-reinforced composites: a state-of-the-art review. Journal of the Brazilian Society of Mechanical Sciences and Engineering 42, 381, 2020.
• Turaka S., Reddy K. V. K., Sahu R. K., Katıyar J. K., Mechanical properties of MWCNTs and graphene nanoparticles modified glass fibre-reinforced polymer nanocomposite. Bulletin of Materials Science 44, 194, 2021.
• Zhang Z., Friedrich K. Velten K., Prediction on tribological properties of short fibre composites using artificial neural networks. Wear 252, 668-675, 2002.
• Zhao F., Li G., Österle W., Häusler I., Zhang G., Wang T., Wang Q., Tribology International Tribological investigations of glass fi ber reinforced epoxy composites under oil lubrication conditions. Tribiology International 103, 208-217, 2016.