MAKİNE ÖĞRENMESİ ALGORİTMALARI KULLANILARAK BETON BASINÇ DAYANIMININ ÖZDİRENÇ YÖNTEMİNDEN TAHMİNİ

Abstract

Betonun tek eksenli basınç dayanımının doğru biçimde belirlenmesi, yapı güvenliği ve performans değerlendirmesi açısından kritik öneme sahiptir. Geleneksel dayanım belirleme yöntemleri tahribatlı deneylere dayandığından özellikle mevcut yapılarda uygulanmaları sınırlıdır. Bu nedenle doğru akım özdirenç yöntemine dayalı tahribatsız test (NDT) yöntemleri ile makine öğrenmesi algoritmalarının birlikte kullanımı önemli bir alternatif oluşturmaktadır. Bu çalışmada betonun basınç dayanımı, elektrik özdirenç, su/çimento oranı (w/c) ve donatı çapı verileri kullanılarak RF (Rastgele Orman) ve XGBoost (Aşırı Gradyan Artırma) algoritmaları ile tahmin edilmiştir. Toplam 334 beton numunesinden elde edilen deneysel veriler kullanılmıştır. Oluşturulan model ile basınç dayanımı arasında orta düzeyde pozitif ilişkiler bulunmuştur (R=0.77-0.79). RF modeli, GridSearchCV kullanılarak yapılan hiperparametre optimizasyonu ile en iyi sonucu üretmiştir. Model performansı R, RMSE ve MAE ile değerlendirilmiştir. Elde edilen sonuçlara göre RF modeli R=0.79, RMSE=12.79 ve MAE= 8.39 değeri ile orta düzeyde tahmin başarısı göstermiştir. Bulgular, elektrik özdirenç, su/çimento oranı ve donatı çapı ile beton dayanımı arasında orta düzeyde bir model kurulabileceğini göstermektedir.

Keywords

• Beton Dayanımı
• Özdirenç
• Makine Öğrenmesi
• Random Forest
• XGBoost

Ethical Statement

Çıkar Çatışması (Conflict of Interest) Yazar tarafından herhangi bir çıkar çatışması beyan edilmemiştir.

ESTIMATION OF CONCRETE COMPRESSIVE STRENGTH FROM RESISTIVITY METHOD USING MACHINE LEARNING ALGORITHMS

Abstract

Accurate determination of the uniaxial compressive strength of concrete is critical for structural safety and performance evaluation. Traditional strength determination methods rely on destructive testing, limiting their application, especially in existing structures. Therefore, the combined use of non-destructive testing (NDT) methods based on direct current resistivity method and machine learning algorithms constitutes a significant alternative. In this study, the compressive strength of concrete was estimated using RF (Random Forest) and XGBoost (Extreme Gradient Boosting) algorithms with electrical resistivity, water/cement ratio, and reinforcement diameter data. Experimental data obtained from a total of 334 concrete samples were used. A moderately positive correlation was found between the model developed and compressive strength (R=0.77-0.79). The RF model produced the best result with hyperparameter optimization using GridSearchCV. Model performance was evaluated using R, RMSE and MAE. According to the results obtained, the RF model showed a moderate level of prediction success with R=0.79, RMSE=12.79, and MAE=8.39. The findings indicate that a moderate model can be established between electrical resistivity, water/cement ratio, reinforcement diameter and concrete strength.

Keywords

• Concrete Strength
• Resistivity
• Machine Learning
• Random Forest
• XGBoost

Ethical Statement

No conflict of interest was declared by the author

References

1. ASTM C1202-12, 2012. Standard Test Method for Electrical Indication of Concrete’s Ability to Resist Chloride Ion Penetration, ASTM International, West Conshohocken, PA, 2012, 8 pp.
2. Aceña, V., Diego, I.M., Fernández, R.R., Moguerza, J.M., 2022. Minimally overfitted learners: A general framework for ensemble learning, Knowledge-Based Systems, 254,109669. https://doi.org/10.1016/j.knosys.2022.109669
3. Asteris, P.G., Apostolopoulou, M., Skentou, A.D., Karayiannis, G., Gemi, L., 2019. Application of artificial neural networks for the prediction of the compressive strength of cement-based mortars. Computers and Concrete, 24(4), 329–345.
4. Asteris, P.G., Koopialipoor, M., Armaghani, D.J., Ghanbari, M., Skentou, A.D., 2021. Prediction of cement-based mortars compressive strength using machine learning techniques. Neural Computing and Applications, 33(19), 13089–13121. https://doi.org/10.1007/s00521-021-06004-8
5. Breiman, L., 2001. Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
6. Chandrahas, N.S., Choudhary, B.S., Teja, M.V., Kumar, S., 2022. XGBoost algorithm to simultaneous prediction of rock fragmentation and induced ground vibration using unique blast data. Applied Sciences, 12(10), 5269. https://doi.org/10.3390/app12105269
7. Chen, T., Guestrin, C., 2016. XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). Association for Computing Machinery. https://doi.org/10.1145/2939672.2939785
8. Chen, H., Asteris, P. G., Armaghani, D.J., Ghanbari, M., Skentou, A.D., 2019. Assessing dynamic conditions of the retaining wall using two hybrid intelligent models. Applied Sciences, 9(6), 1042. https://doi.org/10.3390/app9061042

9. Ekin, N., Uyanık, O., 2021a. Anisotropy of reinforced concrete from geophysical methods. Science of Sintering, 53(3), 323–345.
10. Ekin, N., Uyanık, O., 2021b. Comparison of static and dynamic elastic moduli in concrete: Effects of compressive strength, curing conditions and reinforcement. Iranian Journal of Science and Technology, Transactions A: Civil Engineering, 45(4), 2327–2343.
11. Ekin, N., 2021a. Relationship between amplitude anisotropy and compressive strength of reinforced concrete depending on curing conditions. Indian Journal of Engineering and Materials Sciences, 28(3), 286–299.
12. Ekin, N., 2021b. Concrete anisotropy estimated from ultrasonic signal amplitudes. Materials Testing, 63(12), 1164–1173.
13. Hou, S., Liu, Y., Yang, Q., 2021. Real-time prediction of rock mass classification based on TBM operation big data and stacking technique of ensemble learning. Journal of Rock Mechanics and Geotechnical Engineering, 14(1), 123–143. https://doi.org/10.1016/j.jrmge.2021.05.004
14. Kardani, N., Zhou, A., Nazem, M., Zhang, L., 2021. Improved prediction of slope stability using a hybrid stacking ensemble method based on finite element analysis and field data. Journal of Rock Mechanics and Geotechnical Engineering, 13(1), 188–201. https://doi.org/10.1016/j.jrmge.2020.05.011
15. Lin, C.J., Wu, N.J., 2021. An ANN model for predicting the compressive strength of concrete. Applied Sciences, 11(9), 3798. https://doi.org/10.3390/app11093798
16. Luo, G., 2016. A review of automatic selection methods for machine learning algorithms and hyper-parameter values. Network Modeling Analysis in Health Informatics and Bioinformatics, 5(1), 1–16. https://doi.org/10.1007/s13721-016-0125-6
17. Mai, H.V.T., Nguyen, M.H., Trinh, S.H., Nguyen, H., 2023. Optimization of machine learning models for predicting the compressive strength of fiber-reinforced self-compacting concrete. Frontiers of Structural and Civil Engineering, 17(2), 284–305.
18. Mangalathu, S., Shin, H., Choi, E., Kim, J., 2021. Explainable machine learning models for punching shear strength estimation of flat slabs without transverse reinforcement. Journal of Building Engineering, 39, 102300. https://doi.org/10.1016/j.jobe.2021.102300
19. Mao, H., Deng, X., Jiang, H., Li, Z., 2021. Driving safety assessment for ride-hailing drivers. Accident Analysis & Prevention, 149, 105574. https://doi.org/10.1016/j.aap.2020.105574
20. Nasiri, H., Homafar, A., Chelgani, S.C., 2021. Prediction of uniaxial compressive strength and modulus of elasticity for travertine samples using an explainable artificial intelligence. Results in Geophysical Sciences, 8, 100034. DOI:10.1016/j.rgsci.2021.100034
21. Öncü, Z., 2025. Estimation of concrete strength by using artificial intelligence (AI) methods, Nondestructive Testing and Evaluation, DOI: 10.1080/10589759.2025.2536660
22. Pearson, K., 1896. Mathematical contributions to the theory of evolution—III. Regression, heredity, and panmixia. Philosophical Transactions of the Royal Society of London A, 187, 253–318. https://doi.org/10.1098/rsta.1896.0007
23. Rodgers, J.L., Nicewander, W.A., 1988. Thirteen ways to look at the correlation coefficient. The American Statistician, 42(1), 59–66. https://doi.org/10.1080/00031305.1988.10475524
24. Sabbağ, N., 2016. Beton Kalitesinin Jeofizik Yöntemlerle Belirlenmesi (Doktora tezi), Süleyman Demirel Üniversitesi, 172 s.
25. Sabbağ, N., Uyanık, O., 2018a. Determination of the reinforced concrete strength by apparent resistivity depending on the curing conditions. Journal of Applied Geophysics, 155, 13–25. https://doi.org/10.1016/j.jappgeo.2018.03.008
26. Sabbağ, N., Uyanık, O., 2018b. Determination of reinforced dry concrete strength by electrical resistivity method. Pamukkale University Journal of Engineering Sciences, 24(6), 1230–1236.
27. Sabbağ, N., Uyanık, O., 2017. Prediction of reinforced concrete strength by ultrasonic velocities. Journal of Applied Geophysics, 141, 13–23. https://doi.org/10.1016/j.jappgeo.2017.04.005
28. Thornton, C., Hutter, F., Hoos, H.H., Leyton-Brown, K., 2013. Auto-WEKA: Combined selection and hyperparameter optimization of classification algorithms. In Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 847–855). Association for Computing Machinery. https://doi.org/10.1145/2487575.2487629
29. Uyanık, O., 1995. Isparta havaalanının zemin özelliklerinin jeofizik (sismik-rezistivite) yöntemlerle incelenmesi (Yüksek lisans tezi). Süleyman Demirel Üniversitesi, 139 s.
30. Uyanık, O., Sabbağ, N., Uyanık, N. A., Öncü, Z., 2019. Prediction of mechanical and physical properties of some sedimentary rocks from ultrasonic velocities. Bulletin of Engineering Geology and the Environment, 78(8), 6003–6016. https://doi.org/10.1007/s10064-019-01501-6
31. Yang, Z., Wu, Y., Zhou, Y., Tang, H., Fu, S., 2022. Assessment of machine learning models for the prediction of rate-dependent compressive strength of rocks. Minerals, 12, 731. https://doi.org/10.3390/min12111731
32. Zhou, J., Li, E., Wei, H., Zhang, J., 2019. Random forests and cubist algorithms for predicting shear strengths of rockfill materials. Applied Sciences, 9(8), 1621. https://doi.org/10.3390/app9081621