Çınar akçaağaç (Acer platanoides L.) odununun değişen nem seviyelerinde mekanik özelliklerinin makine öğrenimi tabanli tahmini

Öz

Bu çalışmanın amacı %0, %12, %27 ve %50 denge rutubetlerinde koşullandırılmış akçaağaç (Acer platanoides L.) odununun mekanik tepkisini sistematik olarak incelemektir. Hazırlanan örnekler üzerinde eğilme direnci (ED) ve elastikiyet modülü (EM) liflere paralel basınç direnci (LPBD) testleri gerçekleştirilmiştir. Daha sonra Rastgele Orman (RO) regresyonu, Polinom Regresyonu (PR) ve Destek Vektör Regresyonu (DVR) gibi makine öğrenimi teknikleri, ahşabın mekanik tepkilerini tahmin etmek için kullanılmıştır. Nem içeriği %0'dan %50'ye çıktıkça eğilme mukavemeti %65,15; LPBD ise %60 azalmıştır. Bu sonuçlar, nem içeriğinin akçaağaç odununun mekanik mukavemeti üzerinde önemli bir etkiye sahip olduğunu açıkça göstermektedir. RF modeli, tüm nem koşullarında tutarlı bir şekilde yüksek belirleme katsayısı (R² ≈ 0,90) korunarak güçlü bir genelleme performansı sergilemiştir. PR, çeşitli nem içeriği seviyelerinde önemli ölçüde doğru sapma tahminleri (R² 0,780 ile 0,999 arasında) ve elastikiyet tahminlerinde dikkate değer varyasyonlar sağlamıştır. SVR modeli, sapma için önemli ölçüde doğru tahminler sağlamış (R² = 0,806–0,986), ancak elastikiyet modülü tahminlerinde (R² = 0,585–0,971), özellikle lif doygunluk noktasına yakın bölgelerde önemli farklılıklar gözlenmiştir.

Anahtar Kelimeler

• Akçaağaç
• Nem içeriği
• Mekanik özellikler
• Makine öğrenimi
• Rastgele orman
• Polinom regresyonu
• Destek vektör regresyonu

Proje Numarası

1919B012302777

Machine learning-based prediction of some mechanical properties of sycamore maple (Acer platanoides L.) wood at varying moisture levels

Öz

This study systematically investigates the mechanical response of Norway maple (Acer platanoides L.) wood conditioned at equilibrium moisture contents of 0%, 12%, 27%, and 50%. Modulus of rupture (MOR) and modulus of elasticity (MOE) tests, as well as compressive strength parallel to the grain (CS//) tests, were conducted on the prepared samples. Subsequently, machine learning techniques such as Random Forest (RF) regression, Polynomial Regression (PR), and Support Vector Regression (SVR) were used to predict the mechanical response of the wood. As moisture content increased from 0% to 50%, bending strength decreased by 65.15%, and LPBD decreased by 60%. These results clearly demonstrate that moisture content has a significant effect on the mechanical strength of white ash wood. The RF model demonstrated strong generalization performance, maintaining a consistently high coefficient of determination (R² ≈ 0.90) across all moisture conditions. PR provided significantly accurate deflection predictions (R² between 0.780 and 0.999) at various moisture content levels and notable variations in elasticity predictions. The SVR model provided significantly accurate predictions for deviation (R² = 0.806–0.986), but notable differences were observed in elasticity predictions (R² = 0.585–0.971), particularly in regions close to the fiber saturation point.

Anahtar Kelimeler

• Sycamore maple wood
• Moisture content
• Mechanical properties
• Machine learning
• Random forest
• Polynomial regression
• Support vector regression

Destekleyen Kurum

TÜBİTAK

Proje Numarası

1919B012302777

Teşekkür

The authors gratefully acknowledge the financial support provided by the Scientific and Technological Research Council of Turkey (TÜBİTAK) under the 2209-A University Student Research Projects Support Program (Project No: 1919B012302777).

Kaynakça

1. Ciupak, A., Dziwulska-Hunek, A., Gładyszewska, B., Kwaśniewska, A., 2019. The relationship between physiological and mechanical properties of Acer platanoides L. and Tilia cordata Mill. leaves and their seasonal senescence. Scientific Reports, 9(1): 4287.
2. Deomano, E.C., Zink-Sharp, A., 2004. Bending properties of wood flakes of three southern species. Wood Fiber Science, 36(4): 493–499.
3. Dietsch, P., Franke, S., Franke, B., Gamper, A., Winter, S., 2015. Methods to determine wood moisture content and their applicability in monitoring concepts. Journal of Civil Structural Health Monitoring, 5: 115-127.
4. DIN 52186:1978-06, Testing of wood; bending test, Deutsches Institut für Normung, Berlin, Germany, 1978.
5. Duchesne, I., Vincent, M., Wang, X.A., Ung, C.H., Swift, D.E., 2016. Wood mechanical properties and discoloured heartwood proportion in sugar maple and yellow birch grown in New Brunswick. BioResources, 11(1): 2007-2019.
6. Erten, P., Sözen, M.R. 1994. Determination of Some Physical and Mechanical Properties of Stone Pine (Pinus pinea), Pine (Pinus nigra Arnold) and Sycamore Maple (Acer platonodies) Wood. Eastern Black Sea Forestry Research Institute Technical Bulletin Series. 266:, 1-37.
7. Green, D.W., Evans J.W., Logan J.D., Nelson, W.J., 1999. Adjusting modulus of elasticity of lumber for changes in temperature. Madison, WI: Forest Products Society. Forest Products Journal, 49(10): 82–94.
8. Green, D.W., Evans, J.W., 2008. The immediate effect of temperature on the modulus of elasticity of green and dry lumber. Wood Fiber Science, 40(3): 374–383.

9. Haftkhani, A.R., Abdoli, F., Sepehr, A., Mohebby, B., 2021. Regression and ANN models for predicting MOR and MOE of heat-treated fir wood. Journal of Building Engineering, 42:, 102788.
10. ISO 13061-2:2014. Physical and Mechanical Properties of Wood—Test Methods for Small Clear Wood Specimens—Part 2: Determination of Density for Physical and Mechanical Tests; International Organization for Standardization, Geneva, Switzerland, 2014.
11. Kollmann, F., 1951. Technologie des Holzes und der Holzwerkstoffe, vol 1, 2nd edn. Springer, Berlin.
12. Kollmann, F.F.P., Coté, W.A., 1968: Principles of wood science and technology. Volume I: Solid wood. Springer-Verlag. Berlin, Heidelberg, New York, 592 pp.
13. Nasir, V., Kooshkbaghi, M., & Cool, J., (2020). Sensor fusion and random forest modeling for identifying frozen and green wood during lumber manufacturing. Manufacturing Letters, 26:, 53–58. https://doi.org/10.1016/j.mfglet.2020.09.011.
14. Öge, M., Ak, Ş., Özdemir Öge, T., Küçük, Y., Günen, A., Gök, M.S., 2024. Estimation of wear performance of Ni-based superalloys with machine learning models. Proceedings of the 2nd International Palestra Scientific Research Congress, 246–262.
15. Özdemir Öge, T. (2025). Enhancement and machine learning-based prediction of tribological properties of PC/PBT/GNPs nanocomposites. ACS Omega, 10(22): 23639−23662.
16. Özdemir Öge, T., Öge M., 2025. Prediction of tribological properties of PC‐PBT/GNP‐MWCNT nanocomposites using machine learning models. Journal of Applied Polymer Science, 142: e56834.
17. Schubert, M., Luković, M., Christen, H., 2020. Prediction of mechanical properties of wood fiber insulation boards as a function of machine and process parameters by random forest. Wood Science and Technology, 54(3): 703-713. Sedlar, T., Šefc, B., Stojnić, S., Sinković, T., 2021. Wood quality characterization of sycamore maple (Acer pseudoplatanus L.) and its utilization in wood products industries. Croatian Journal of Forest Engineering: Journal for Theory and Application of Forestry Engineering, 42(3): 543-560.
18. Sell, J., 1997. Eigenschaften und Kenngrößen von Holzarten (4. Ausg.). Dietikon. Baufachverlag.
19. Sözen, E., Bardak, T., Kayahan, K., 2021. Determination of the effect of laminated veneer lumber (LVL) moisture content on pressure resistance by artificial ıntelligence. Turkish Journal of Forestry, 22(2): 157-164.
20. TS ISO 13061-1, Physical and mechanical properties of wood - Test methods for small clear wood specimens - Part 1: Determination of moisture content for physical and mechanical tests. Turkish Standards Institute, Ankara, Turkiye, 2021.
21. TS ISO 13061-17, Physical and mechanical properties of wood — Test methods for small clear wood specimens Part 17: Determination of ultimate stress in compression parallel to grain, International Organization for Standardization: Geneva, Switzerland, 2020.
22. von Wedel, K., 1964. Untersuchungen ueber Eigenschaften, Verwertung und Verwendung des Ahornholzes (Doctoral dissertation).
23. Wang, S.Y., Wang, H.L., 1999. Effects of moisture content and specific gravity on static bending properties and hardness of six wood species. Journal of wood science, 45: 127-133.
24. Wood Handbook 72 (1987) Wood as an engineering material. Forest Products Laboratory, Forest Service, USDA, P4-36-39.