Çeşitli FRP Kumaşlarla Güçlendirilmiş Ahşap Malzemenin Eğilme Özelliklerinin Yapay Sinir Ağları ile Tahmini

Abstract

Son zamanlarda çelik ve betonarme yapıların güçlendirilmesinde olduğu gibi ahşap yapıların güçlendirilmesinde de fiber takviyeli polimer (FRP) uygulamaları kullanılmaya başlanmıştır. Dış etkenlerin ve depremin neden olduğu hasarların zamanla giderilmesi, restorasyon ile yapının taşıma kapasitesinin artırılması, erken yorulma ve yapılan hatalar sonucu oluşabilecek kırılmaların önlenmesi gibi nedenlerle ahşap yapıların güçlendirilmesi gerekmektedir. Zamanla deprem sonucu hasar gören yapıların onarım ve güçlendirme yöntemlerinin iyileştirilmesi gerekliliği ortaya çıkmaktadır. Bu çalışmada, karbon, cam, aramid ve bazalt olmak üzere 4 farklı FRP kumaş ile güçlendirilmiş İroko ve Dişbudak ağaç türlerinin maksimum yükü, yer değiştirmesi, eğilme dayanımı ve elastisite modülü eğilme testi ile belirlenmiştir. Deneysel çalışma sonucunda, güçlendirilmiş numunelerin maksimum yük, yer değiştirme, eğilme dayanımı ve elastisite modülü değerleri yapay sinir ağları (YSA) ile tahmin edilmiştir. Sonuç olarak, YSA kullanılarak FRP ile güçlendirilmiş bir ahşap malzemenin eğilme özelliklerinin tahmin edilebileceği belirlenmiştir.

Keywords

• Ahşap Malzeme
• FRP
• Güçlendirme
• Yapay Sinir Ağları
• Iroko
• Dişbudak

Prediction of Flexural Properties of Wood Material Reinforced with Various FRP Fabrics by Artificial Neural Networks

Abstract

Recently, fiber reinforced polymer (FRP) applications have started to be used in the reinforcement of wooden structures, such as in the reinforcement of steel and reinforced concrete structures. It is necessary to strengthen the wooden structures for reasons such as removing the damages caused by external factors and earthquakes in time, increasing the load-bearing capacity of the structure by restoration, preventing early fatigue and breakages that may occur as a result of mistakes made in the design. The necessity to improve the repair and strengthening methods of the structures damaged as a result of the earthquake over time arises. In this study, the maximum load, displacement, flexural strength and modulus of elasticity of the wood material of Iroko and Ash tree species reinforced with 4 different FRP fabrics, namely carbon, glass, aramid and basalt, were determined by bending test. As a result of the experimental study, the maximum load, displacement, flexure strength and elasticity modulus values of the reinforced samples were estimated by artificial neural network (ANN). As a result, it was determined that the flexural properties of a wood material strengthened with FRP by using ANN can be predicted.

Keywords

• Wood Materials
• FRP
• Reinforcement
• Artificial Neural Network
• Iroko
• Ash

References

1. [1] I. Usta, ‘‘The current state of the wood material impregnation industry in Turkey and suggestions for its development,’’ M.S. thesis, Dep. Woodworking Industrial Engineering, Hacettepe University, Ankara, Turkey, 1993.
2. [2] B. Uysal, ‘‘Wood material lecture notes,’’ Karabuk University, Faculty of Technical Education, Karabük, Turkey, 2005. [3] J. E. Winandy, ‘‘Wood properties,’’ Encyclopedia of Agricultural Science, vol. 4, pp. 549-561, 1994.
3. [4] P. O. Kettunen, ‘‘Wood: Structure and properties,’’ Uetikon-Zuerich: Trans Tech Publication, 2006.
4. [5] S. Kilincarslan and Y. Şimşek Türker ‘’Experimental investigation of the rotational behaviour of glulam column-beam joints reinforced with fiber reinforced polymer composites,’’ Composite Structures, vol. 262, 2021.
5. [6] H. T. Sahin, M. B. Arslan, S. Korkut and C. Sahin. ‘’Colour Changes of Heat‐Treated Woods of Red‐Bud Maple, European Hophornbeam And Oak’’. Color Research & Application, vol. 36, no. 6, 462-466, 2011.
6. [7] C. K. Sahin and B. Onay. ‘’Alternatıve wood wpecies for playgrounds wood from fruit trees’’. Wood Research, vol. 65, no. 1, pp. 149- 160, 2020.
7. [8] C. Sahin, M. Topay and A.A. Var. ‘’A study on some wood species for landscape applications: surface color, hardness and roughness changes at outdoor conditions’’. Wood Research, vol. 65, no. 3, pp. 395-404, 2020.
8. [9] R. Günay, Traditional Wooden Structures Problems and Solutions, Istanbul: Birsen Publishing, pp. 43-64, 2002.

9. [10] B. R. Öztürk, ‘‘Mechanical properties of laminated wooden beams from Turkish yellow pine,’’ Journal of Istanbul Technical University, vol. 5, no. 2, pp. 25-36, 2006.
10. [11] S. Kilincarslan and Y. S. Turker, ‘‘Investigation of wooden beam behaviors reinforced with fiber rein-forced polymers,’’ Organic Polymer Material Research, vol. 2, no. 1, pp. 1-7, 2020.
11. [12] Y. Şahin, Introduction to Composite Materials, Ankara: Gazi Bookstore, pp. 2-33, 2000.
12. [13] Ş. Kilincarslan, Y. Şimşek Türker and M. İnce, ‘‘Prediction of heat-treated spruce wood surface roughness with artificial neural network and random forest algorithm,’’ The International Conference on Artificial Intelligence and Applied Mathematics in Engineering, Cham, 2020.
13. [14] S. Bolat and Ö. Kalenderli, “Electrode shape optimization with artificial neural network using levenberg-marquardt algorithm,” International XII. Turkish Symposium on Artificial Intelligence and Neural Networks – TAINN, 2003, pp. 256-261.
14. [15] S. Subaşı, “Prediction of mechanical properties of cement containing class C fly ash by using artificial neural network and regression technique,” Scientific Research and Essay, vol. 4, no. 4, pp. 289-297, 2009. [16] S. S. Dorvlo, Jervase, J.A. and Al-Lawati, A., “Solar radiation estimation using artificial neural network,” Applied Energy, vol. 71, pp. 307–319, 2002. [17] D. L. Shrestha and D. P. Solomatine, ‘‘Machine learning approaches for estimation of prediction ınterval for the model output,’’ Neural Networks, vol. 19, no. 2, pp. 225-235, 2006.
15. [18] T. Sasakawa, J. Hu and K. Hirasawa,‘‘A brainlike learning system with supervised, unsupervised, and reinforcement learning,’’ Electrical Engineering in Japan, vol. 162, no. 1, pp. 32-39, 2008.
16. [19] S. Ghosh-Dastidar and H. Adeli, ‘‘A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection,’’ Neural Networks, vol. 22, no. 10, pp. 1419-1431, 2009.
17. [20] T. D. Sanger, ‘‘Optimal unsupervised learning in a single-layer linear feedforward neural network,’’ Neural Networks, vol. 2, no. 6, pp. 459-473, 1989.
18. [21] S. Singh, T. Jaakkola, M. L. Littman and C. Szepesvári, ‘‘Convergence results for single step on policy reinforcement learning algorithms,’’ Machine Learning, vol. 38, no. 3, pp. 287-308, 2000.
19. [22] V. Rodriguez-Galiano, M. Sanchez-Castillo, Chica-Olmo, M. and M. J. O. G. R. Chica-Rivas, ‘‘Machine learning predictive models for mineral prospectivity: an evaluation of neural networks, random forest, regression trees and support vector machines,’’ Ore Geology Reviews, vol. 71, pp. 804-818, 2015.
20. [23] M. Van Gerven and S. Bohte, ‘‘Artificial neural networks as models of neural ınformation processing,’’ Frontiers in Computational Neuroscience, no.11, pp.114, 2017.
21. [24] G. Zhang, B. E. Patuwo, M. Y. Hu, ‘‘Forecasting with artificial neural networks: the state of the art,’’ International Journal of Forecasting, vol. 14, no. 1, pp. 35-62, 1998.
22. [25] A. Recchia, ‘‘R-Squared measures for two-level hierarchical linear models using SAS,’’ Journal of Statistical Software, vol. 32, no. 2, pp. 1-9, 2010.