Prediction of Elasticity for Turkish Red Pine (Pinus Brutia Ten.) Lumber Using Linear Modeling and Artificial Neural Networks (ANN)

Öz

In this study, elasticity of Turkish Red Pine (Pinus brutia Ten.) lumbers was predicted using lineer modeling and artificial neural networks (ANN). The lumber samples represent 30-80 years old red pine trees harvested from a south west site in Turkey. Natural frequency values of lumbers in 38 mm x 89 mm in cross section and 3 meters in length were measured by stress wave device. Linear modeling and ANN were evaluated by employing several optimization techniques using some physical measurements from the logs and lumbers. Static elasticity values of the lumbers were determined using three point bending tests. Coefficients of determination between measured and predicted MOE's for linear modeling and ANN were 0.87 and 0.91, respectively. Among the ANN models studied the model which uses visual classes, density, width, annual ring width, moisture content, and natural frequency as inputs gave the highest coefficient of determination of 0.91. The results show that linear modeling and ANN can provide accurate elasticity prediction for Turkish Red Pine lumber coming from different logs. 

Anahtar Kelimeler

• Red Pine lumber
• Elasticity
• Linear modeling
• Artificial neural networks

Kızılçam (Pinus brutia Ten.) Kerestesinde Elastikiyet Modülünün Doğrusal Modelleme ve Yapay Sinir Ağları İle Tahmini

Öz

Bu çalışmada Kızılçam kerestesinde elastikiyet modülü (EM) doğrusal modelleme ve yapay sinir ağları (YSA) metotları kullanılarak tahmin edilmiştir. Kereste örnekleri Türkiye'nin güneyinden 30-80 yaşlarında kesilen kızılçam ağaçlarından elde edilmiştir. 3 metre boyunda 38 x 89 mm enine kesitteki kerestelerin doğal frekans değerleri stres dalga aygıtıyla belirlenmiştir. Doğrusal modelleme ve YSA kereste ve tomruklardan elde edilen bazı fiziksel ölçümler ile doğal frekans kullanılarak farklı optimizasyon teknikleri ile değerlendirilmiştir. Kerestelerin (EM) değerleri aynı zamanda laboratuvar ortamında 3-nokta eğilme testi ile belirlenmiştir. Doğrusal modelleme ve YSA kullanılarak tahmin edilen EM ile ölçülen EM değerleri arasında bulunan regresyon katsayıları sırasıyla 0.87 ve 0.91'dir. Geliştirilen YSA modelleri arasında görsel sınıf, yoğunluk, kereste genişliği, yıllık halka genişliği, rutubet miktarı ve doğal frekansı kullanan model en yüksek regresyon katsayısını vermiştir. Çalışma sonuçlarına göre farklı tomruklardan elde edilen kerestelerde EM doğrusal modelleme ve YSA kullanılarak yüksek hassasiyette tahmin edebilir.

Anahtar Kelimeler

• Kızılçam kereste
• Elastikiyet modülü
• Doğrusal modelleme
• Yapay sinir ağları

Kaynakça

1. Altınok, N., 2006. Use of artificial neural network for prediction of mechanical properties of a-Al2O3 particulate-reinforced al–si10mg alloy composites prepared by using stir casting process. Journal of
2. Composiıte Materials, 40: 9, 779-796. Avramidis, S., Iliadis, L., 2005(a). Predicting wood thermal conductivity using artificial neural networks.
3. Wood and Fiber Science, 37(4), 682-690. Avramidis, S., Iliadis L., 2005(b). Wood-Water sorption isotherm prediction with artificial neural networks: A preliminary study. Holzforschung, 59 (3), 336-341.
4. Avramidis, S., Wu, H., 2006. Artificial neural network and mathematical modeling comparative analysis of nonisothermal diffusion of moisture in wood. Holz als
5. Roh- und Werkstoff, 65, 89–93. Budak, A., Can, İ., 2008. Yapay sinir ağları ile tek eksenli bileşik eğilme altındaki betonarme kolon kesitlerinin donatı hesabı. Fırat Üniversitesi Fen ve Mühendislik Bilimleri Dergisi, 20 (1), 135-143.
6. Clausen, C.A, Ross, RJ., Forsman, J.W, Balachowski, J.D., 2001. Condition Assessment of Roof Trusses of
7. Quincy Mine Blacksmith Shop In Keweenaw National Historical Park. FPL-RN-0281, Forest Products Laboratory, Madison.
8. Demir, F., 2008. Prediction of elastic modulus of normal and high strength concrete by artificial neural networks. Construction and Building Materials, 22, 1428–1435.

9. Dıvós, F., Tanaka, T., 2005. Relation between static and dynamic modulus of elasticity of wood. Acta Silv. Lign. Hung., 1, 105-110.
10. Esteban, L.G., Fernandez, F.G., de Palacios, P., 2009.
11. MOE Prediction in Abies pinsapo boiss. Timber: application of an artificial neural network using non- destructive testing. Computers and Structures, 87, 1360–1365.
12. Fernandez, G.F., Esteban, L.G., de Palacios, P., Navarro, N., Conde, M., 2008. Prediction of standard particleboard mechanical properties utilizing an artificial neural network and subsequent comparison with a multivariate regression model. Invest Agrar Sist Recur For., 17(2), 178–87.
13. Han, G., Wu, Q., Wang, X., 2005. Stress-Wave velocity of wood-based panels: effect of moisture, product type, and material direction. Forest Products Journal, 56(1), 28-33.
14. Hola, J., Schabowicz, K., 2005. Appication of ANN to determine concrete compressive strength based on non-destructive tests. Journal of Civil Engineering and Management, 11 (1), 23-32.
15. Lee, S.C., 2003. Prediction of Concrete Strength Using
16. Artificial Neural Networks. Engineering Structures, 849–857. Mansfield, S.D., Iliadis, L., Avramidis, S., 2007. Neural network prediction of bending strength and stiffness in Western Hemlock (Tsuga heterophylla Raf.). Holzforschung, 61(6), 707–16.
17. Mansfield, S.D., Kyu-Young, K., Lazaros, I., Stavros, T., Avramidis, S., 2011. Predicting the strength of populus spp. clones using artificial neural networks and E-regression support vector machines (ε-rSVM). Holzforschung, 65(6), 855–863.
18. Noorzaei, J., Hakim, S.J.S., Jaafar, M.S., Thanoon, W.A.M., 2007. Development of artificial neural networks for predicting concrete compressive strength. International Journal of Engineering and Technology, 4(2), 141-153.
19. Özşahin, S., 2012. The use of the artificial neural network for modeling the moisture absorption and thickness swelling of oriented strand board. BioResources, 7(1), 1053-1067.
20. Pham, D.T., Soroka, A.J., Ghanbarzadeh, A., Koc, E., Otri, S., Packianather, M., 2006. Optimising Neural
21. Networks for Identification of Wood Defects Using the Bees Algorithm. IEEE International Conference on Industrial Informatics, 1346-1351.
22. Ross, R.J., Brashaw, B.K., 1998. Pellerin, R.F.,
23. Nondestructive evaluation of wood. Forest Products Journal, 48(1), 14–19. Rozema, P., 2007. Timber Grader MTG-Brookhuis
24. Micro-Eletronics BV. the Netherlands. Samarasinghe, S., Kulasiri, D., Jamieson, T., 2007.
25. Neural networks for predicting fracture toughness of ındividual wood samples. Silva Fennica, 41(1), 105– Sağıroğlu, Ş., Beşdok, E., Erler, M., 2003.
26. Mühendislikte Yapay Zekâ Uygulamaları I: Yapay Sinir Ağları, Ufuk Kitabevi. Kayseri. Seyhan, A.T., Gokmen, T., Murat, K., Metin, T., 2005.
27. Artificial neural network (ANN) prediction of compressive strength of VARTM processed polymer composites. Computational Materials Science, 34, 99–105. Tho, K.K., Swaddiwudhipong, S., Liu, Z.S., Hua, J., 2004.
28. Artificial neural network model for material characterization by ındentation, modelling simul. Mater. Sci. Eng., 12, 1055–1062.
29. Topçu, İ.B., Sarıdemir, M., 2008. Prediction of
30. Compressive Strength of Concrete Containing Fly Ash Using Artificial Neural Networks and Fuzzy Logic. Computational Materials Science, 41,305–311. TS EN 1611-1, 2002. Biçilmiş yapacak odun
31. (kereste)- İğne yapraklı (yumuşak) odunların görünüşlerine göre sınıflandırılması, Bölüm 1: Avrupa ladinleri, göknarları, çamları ve duglas göknarları. TS EN 1927-2, 2009. İğne yapraklı (yumuşak) ağaç yuvarlak yapacak odunlarının kalite sınıflandırması - Bölüm 2: Çamlar.
32. Vijayabaskar, V., Gupta, R., Chakrabarti, P.P., Bhowmick, A.K., 2006. Prediction of properties of rubber by using artificial neural networks. Journal of
33. Applied Polymer Science, 100, 2227–2237.
34. Wang, X., Ross, R.J., Mattson, J.A., Erickson, J.R., Forsman, J.W., Geske, E.A., Wehr, M.A., 2001. Several
35. Nondestructive Evaluation Techniques for Assessing Stiffness and MOE of Small- Diameter Logs. FPL-RP- 600, Forest Products Laboratory Research Paper. Wang, X., Divos, F., Pilon, C., Brashaw, B.K., Ross, R.J., Pellerin, R.F., 2004. Assessment of Decay in Standing
36. Timber Using Stress Wave Timing Nondestructive Evaluation Tools. FPL−GTR−147, USDA Forest Products Laboratory, Madison. Yang, T.H., Wang, S.Y., Lin, C.J., Tsai, M.J., 2008.
37. Evaluation of the mechanical properties of douglas-fir and japanese cedar lumber and ıts structural glulam by non-destructive techniques. Construction and Building Materials, 22, 487-493. Yurtoğlu, H., 2006. Yapay sinir ağlan metodolojisi ile öngörü modellemesi. Uzmanlık Tezi, Yıldız Teknik
38. Üniversitesi, Sosyal Bilimler Enstitüsü, İstanbul.