Modeling the Stress-Strain Curve of Confined Concrete Using Ensemble Machine Learning Models
DOI:
https://doi.org/10.26439/ciic2025.8896Palabras clave:
Confined concrete, ensemble models, machine learning (ML), stress-strain curveResumen
The objective of this research is to develop an efficient and broadly applicable data-driven model capable of determining the stress-strain curve of confined concrete. Therefore, experimental data of 115 specimens of reinforced concrete columns with square and circular cross-sections were collected from previous investigations in which uniaxial compression tests were performed. Using this data, Random Forest (RF), Adaptive Boosting (AdaBoost), and Extreme Gradient Boosting (XGBoost) models were evaluated to define the most accurate model. Subsequently, the final selected model (based on XGBoost) was optimized, achieving R2 values of 0.97 for the peak stress of confined concrete (ƒcc), 0.93 for the axial strain at peak confined stress (ɛcc), and 0.81 and 0.73 for the axial strains at which stress drops to 85 % (ɛ85) and 50 % (ɛ50), respectively, when evaluated with the testing data. In addition, the SHapley Additive exPlanations (SHAP) technique was used to explain and determine the importance of different parameters in the outcome of the predictive model. Based on the predictions of the XGBoost model, a proposed stress-strain curve was formulated. Finally, a comparison of ƒcc, ɛcc, and the stress strain curve was performed taking into account the experimental results, the previous models, and the proposed model. The comparison results indicate that the proposed model shows a closer agreement with the experimental data.
Descargas
Referencias
[1] A. Fafitis and S. P. Shah, “Lateral Reinforcement for High-Strength Concrete Columns,” in ACI Special Publication SP-87, High-Strength Concrete, American Concrete Institute, Farmington Hills, MI, USA, Sept. 1985, pp. 213–232.
[2] J. B. Mander, M. J. N. Priestley, and R. Park, “Theoretical Stress-Strain Model for Confined Concrete,” Journal of Structural Engineering, vol. 114, no. 8, pp. 1804–1826, Sep. 1988, doi: https://doi.org/10.1061/(asce)0733-9445(1988)114:8(1804)
[3] J. Hoshikuma, K. Kawashima, K. Nagaya, and A. Taylor, “Stress-Strain Model for Confined Reinforced Concrete in Bridge Piers,” Journal of Structural Engineering, vol. 123, no. 5, pp. 624–633, May 1997, doi: https://doi.org/10.1061/(asce)07339445(1997)123:5(624)
[4] R. Park and T. Paulay, Estructuras de Concreto Reforzado, 2nd ed., Mexico City, Mexico: Limusa, 1983.
[5] A. Géron, Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 2nd ed., Sebastopol, CA, USA: O’Reilly Media, 2019.
[6] L. Breiman, “Random Forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, Oct. 2001, doi: https://doi.org/10.1023/A:1010933404324
[7] Y. Freund and R. E. Schapire, “A Decision Theoretic Generalization of On-Line Learning and an Application to Boosting,” Journal of Computer and System Sciences, vol. 55, no. 1, pp. 119–139, Aug. 1997, doi: https://doi.org/10.1006/jcss.1997.1504
[8] T. Chen and C. Guestrin, “XGBoost: A Scalable Tree Boosting System,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’16, vol. 1, no. 1, pp. 785–794, Aug. 2016, doi: https://doi.org/10.1145/2939672.2939785
[9] X. Guan, H. Burton, M. Shokrabadi, and Z. Yi, “Seismic Drift Demand Estimation for Steel Moment Frame Buildings: From Mechanics Based to Data-Driven Models,” Journal of Structural Engineering, vol. 147, no. 6, Jun. 2021, doi: https://doi.org/10.1061/(asce)st.1943541x.0003004
[10] C. F. Jekel, G. Venter, M. P. Venter, N. Stander, and R. T. Haftka, “Similarity measures for identifying material parameters from using inverse hysteresis analysis,” loops International Journal of Material Forming, vol. 12, no. 3, pp. 355–378, Jul. 2018, doi: https://doi.org/10.1007/s12289-018-1421-8
[11] B. Li, Strength and ductility of reinforced concrete members and frames constructed using high strength concrete, Ph.D. dissertation, Dept. of Civil Engineering, Univ. of Canterbury, Christchurch, New Zealand, 1993.
[12] T. Nagashima, S. Sugano, H. Kimura, and A. Ichikawa, “Monotonic axial compression test on ultra-high-strength concrete tied columns,” in Proc. 10th World Conf. Earthquake Engineering (10th WCEE), Madrid, Spain, 1992, pp. 2983–2988.
[13] S. Razvi and M. Saatcioglu, “Confinement Model for High-Strength Concrete,” Journal of Structural Engineering, vol. 125, no. 3, pp. 281-289, Mar. 1999, doi: https://doi.org/10.1061/(asce)0733-9445(1999)125:3(281)
[14] M. Suzuki, M. Akiyama, K.-N. Hong, I. D. Cameron, and W. L. Wang, “Stress-strain model of high-strength concrete confined by rectangular ties,” in Proc. 13th World Conf. Earthquake Engineering (13th WCEE), Vancouver, BC, Canada, Aug. 2004, Paper No. 3330, pp. 74–76.
[15] S. M. Lundberg, G. G. Erion, and S.-I. Lee, “Consistent individualized feature attribution for tree ensembles,” arXiv preprint arXiv:1802.03888, 2019. [Online]. Available: http://arxiv.org/abs/1802.03888
