Aramid Elyaf Takviyeli Polimer Matris Kompozitlerin Mekanik Deney Sonuçlarının Yapay Sinir Ağlarıyla Tahminleri ve İstatistiksel Analizleri

Kompozit malzemelerde polimer matris en çok tercih edilen malzeme türlerinden biri haline gelmesi personel koruma ya da diğer zırh malzemesi türlerinde de kullanımını artmakla birlikte gelişmeye devam etmektedir. Mekanik deney sonuçlarını birçok yöntemle matematiksel modellemesi yapılırken en fazla tercih edilenlerden bir tanesi de yapay sinir ağları olmaktadır. Kompozit malzemelerin birden fazla parçacık ya da malzemelerin birleşimiyle oluşmasından dolayı birleşime dahil olan parça ya da malzemenin mekanik sonuçlara etkisinin anlamlı veya anlamsız olduğunu söylemenin en güzel yollarından biri de istatistiksel analizlerdir. İstatistiksel analizlerle bilgi ve yorumlanmasıyla birlikte diğer oran ya da türevlerinin kıyaslama ya da karşılama yapılmasına da olanak sağlamaktadır. Bu çalışmada; Numuneler 8 katlı Aramid elyaf takviyeli ve dolgu malzemesi olarak ağırlıkça %0, %1, %2 ve %4 oranında TiB2 ilaveli olup ayrıca 450 ve 900 oryantasyona sahip olarak üretilmiştir. Yapay sinir ağları ve İstatistiksel analizler ağırlıkça %1 TiB2 ilaveli ve 900 oryantasyona sahip kompozitler diğer oranlara göre balistik amaca en anlamlı sonucu vermiştir.

Predictions and Statistical Analysis of Mechanical Experiment Results of Aramid Fiber Reinforced Polymer Matrix Composites with Artificial Neural Networks

In composite materials, the polymer matrix continues to develop as one of the most preferred material types, and its use in personnel protection or other armor material types is increasing. While mathematical modeling of mechanical test results with many methods is done, artificial neural networks are one of the most preferred ones. Statistical analysis is one of the best ways to say that the effect of the part or material included in the combination on the mechanical results is significant or insignificant, since composite materials are formed by more than one particle or combination of materials. It also allows comparison or compensation of other ratios or derivatives, along with information and interpretation through statistical analysis. In this study; The samples were produced with 8-layer Aramid fiber reinforced and 0%, 1%, 2% and 4% TiB2 additions as filling material and also with 450 and 900 orientations. Artificial neural networks and Statistical analysis, composites with 1% TiB2 addition and 900 orientation gave the most significant result for ballistic purpose compared to other ratios.

___

  • Akşehirli Ö., Cangür Ş, Ankaralı H., Sungur M.A. (2012). 24 Faktöriyel Tasarımların Sağlık Alanında Kullanımı. Düzce Üniversitesi Sağlık Bilimleri Enstitüsü Dergisi, 2(3): 1-6.
  • Balaji R., Nadarajan M., Selokar A., Kumar S.S., Sivakumar S. (2019). Modelling And Analysis of Disk Brake Under Tribological Behaviour of Al-Al2O3 Ceramic Matrix Composites/Kevlar® 119 Composite/C/Sic-Carbon Matrix Composite/Cr-Ni-Mo-V Steel. Proceedings 18, 3415–3427.
  • Balaji N. S., Jayabal S., Kalyana S. (2016). A. Neural Network Based Prediction Modeling for Machinability Characteristics of Zea Fiber-Polyester Composites. Trans Indian Inst Met (2016) 69(4):881–889.
  • Chen CT ve Gu GX (2019). Machine Learning For Composite Materials. MRS Communications Volume 9, Issue 2, 556–566.
  • Eyecioğlu Ö. (2021). Bazalt/PANI Kompozitlerinin Dielektrik Özelliklerinin Tahmini İçin Makine Öğrenmesi Modellerinin Karşılaştırılması. Avrupa Bilim ve Teknoloji Dergisi, Sayı 23, S. 817-826.
  • Fazilat H., Ghatarband M., Mazinani S., Asadi Z.A, Shiri M.E., Kalaee M.R. (2012). Predicting The Mechanical Properties Of Glass Fiber Reinforced Polymers Via Artificial Neural Network And Adaptive Neuro-Fuzzy İnference System. Computational Materials Science 58 (2012) 31–37.
  • Gedik İ. (2010). İç-içe Tasarımlarda Dayanıklı Analiz ve Uygulamaları. Yüksek Lisans Tezi Ankara Üniversitesi Fen Bilimleri Enstitüsü İstatistik Anabilim Dalı, Ankara, 93 Sayfa.
  • James R. Brown, Daryl K.C. Hodgeman (1982). An E.s.r. Study of the Thermal Degradation of Kevlar 49 Aramid., Volume 23, Issue 3, March 1982, Pages 365-368.
  • Kubat C., Kiraz A., Atakan Ü. (2017). Matlab Yapay Zekâ ve Mühendislik Uygulamaları Kitabı, İstanbul, Abaküs Yayınları.
  • Kumar S.S., Priyadarshan , Kumar G. S. (2021). Statistical and Artificial Neural Network Technique for Prediction of Performance İn AlSi10Mg-MWCNT Based Composite Materials. Department of Mechanical Engineering, Indian Institute of Technology (ISM) Dhanbad, 826004, Volume 273, 125136.
  • Kadi H.E. (2006). Modeling the mechanical behavior of fiber-reinforced polymeric composite materials using artificial neural networks—A review. Composite Structures Volume 73, Issue 1, May 2006, Pages 1-23.
  • Kuwada M. (1993). Analysis of Variance of Balanced Fractional Factorial Designs. Discrete Mathematics Volume 116, Issues 1–3, 1 June 1993, Pages 335-366.
  • Madara S.R., Pillai S.R., Selvan M. C.P., Heirle J.V. (2021). Modelling of Surface Roughness in Abrasive Waterjet Cutting of Kevlar 49 Composite Using Artificial Neural Network. 2214-7853/ Elsevier, Volume 46, Part 1, 2021, Pages 1-8.
  • Mays D.P. and Myers R.H. (1997). Design and Analysis for a Two-Level Factorial Experiment in The Presence of Variance Heterogeneity. Computational Statistics & Data Analysis Volume 26, Issue 2, 4 December 1997, Pages 219-233.
  • Morrison D.A. (2002). Further Difficulties With Multifactorial Analysis of Variance: Random and Nested Factors and İndependence of Data. Infection, Genetics and Evolution Volume 2, Issue 2, December 2002, Pages 149-152.
  • Nguyen H.T., Nguyen K.T.Q, Tu C. Le, Soufeiani L., Mouritz A.P (2021). Predicting Heat Release Properties of Flammable Fiber-Polymer Laminates Using Artificial Neural Networks. Composites Science and Technology 215, 109007.
  • Okkan U., Serbeş Z.A., Gedik N. (2018). MATLAB ile Levenberg-Marquardt Algoritması Tabanlı YSA Uygulaması. DÜMF Mühendislik Dergisi 9:1 (2018): 351-362.
  • Ramaiah G.B., Chennaiaha R.Y., Satyanarayanarao G.K. (2021). Investigation And Modeling On Protective Textiles Using Artificial Neural Networks For Defense Applications. Materials Science And Engineering B 168 (2010) 100–105.
  • Saraç M.F, Buran D., Koru M. (2018). Investigation of Thermal and Mechanical Properties of Aramid Fiber Reinforced Thermoplastic Polyurethane Elastomer Composites. Süleyman Demirel University Journal of Natural and Applied Sciences V.22, Issue 2, 477-481.
  • Stanimirova L., Kazura M. Vd. (2013). High-Dimensional Nested Analysis of Variance to Assess the Effect of Production Season, Quality Grade and Steam Pasteurization on The Phenolic Composition of Fermented Rooibos Herbal Tea. Talanta, Volume 115, 15 October 2013, Pages 590-599.
  • Suresh N., Balamurugan L., Geethan K.A.V., Kumar M.S. (2021). Statistical Analysis of Mechanical Properties of Al-SiC-WC and Al-SiC-Al2O3 Hybrid Composites. Materials Today: Proceedings, Volume 42, Part 2, 2021, Pages 312-318.
  • Wanga F., Huanga G., Cheng G., Li Y. (2021). Multi-Level Factorial Analysis for Ensemble Data-Driven Hydrological Prediction. Advances in Water Resources Volume 153, July 2021, 103948.
  • Zhang Z., Friedrich K. (2003). Artificial Neural Networks Applied to Polymer Composites. Composites Science and Technology 63, 2029–2044.
Uluslararası Mühendislik Araştırma ve Geliştirme Dergisi-Cover
  • Başlangıç: 2009
  • Yayıncı: -