تجزیه و تحلیل مدل‌‌های ریاضی پیش‌‌بینی مقاومت‌‌های مکانیکی بتن‌‌های مسلح به الیاف فولادی به روش تجربی و مبتنی بر یادگیری ماشین

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه عمران، دانشکده فنی و مهندسی، دانشگاه قم، قم، ایران

2 دانشکده مهندسی راه آهن، دانشگاه علم و صنعت، تهران، ایران

چکیده

هدف از مقاله‏ حاضر، ارائه مدل‏ ریاضی بهینه برای پیش‌بینی مقاومت‌های مکانیکی بتن مسلح به الیاف فولادی است. برای این منظور ضمن مطالعه و بررسی آزمایشگاهی روابط محققین قبلی برای پیش‌بینی مقاومت‌های فشاری، کششی و خمشی بتن الیافی مسلح شده با الیاف فولادی، روابط ریاضی بهینه حاکم بر مسئله به روش یادگیری ماشین بررسی شده است. تمرکز تحقیق حاضر بر روی آن دسته از بتن‌های مسلح به الیاف فولادی است که در ساخت آنها از الیاف به اشکال صاف، موجدار و دوسرقلاب در مقیاس ماکرو استفاده شده است. طی این تحقیق روابط ریاضی مستخرج از روش یادگیری ماشین که با استفاده از مدلسازی پیش‏بینی مقاومت‌های فشاری، کششی و خمشی بتن‌های مسلح به الیاف فولادی و با به کارگیری روش رگرسیون نمادی بر مبنای یک پایگاه داده‌ی مشتمل بر 2283 داده‏ی بین المللی ارائه شده بررسی می‌گردد. کارایی مدل‏های به کار گرفته شده در این تحقیق با استفاده از سنجه‏های آماری آنالیز خطا نظیر RMSE و MAPE سنجیده شده است. نتایج نشان می‌دهد که پارامترهای سایز بزرگترین سنگدانه، مدول الاستیسیته، مقاومت فشاری بتن شاهد، نسبت آب به سیمان، درصد حجمی و طول الیاف، نسبت ابعادی الیاف و مقاومت کششی یا خمشی مرتبط با نوع خروجی مورد بررسی برای بتن الیافی بیشترین اثر را بر روی پیش‌بینی مقاومت دارند. بررسی نتایج نشان می‌دهد که فرمول‏های ارائه شده برای پیش‌بینی مقاومت بتن مسلح برای الیاف مورد بررسی نسبت به روابط ریاضی گذشته از دقت قابل ملاحظه‌ای برخوردار است.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Analysis of Mathematical Models for Predicting the Mechanical Resistance of Concrete Reinforced with Steel Fibers Using Experimental and Machine Learning Methods

نویسندگان [English]

  • Mohmmad Hossein Taghavi Parsa 1
  • Morteza Esmaeili 2
  • mohammadreza adlparvar 1
1 University of Qom
2 Iran University of Science and Technology
چکیده [English]

The purpose of this article is to present an optimal mathematical model for predicting the mechanical resistance of concrete reinforced with steel fibers. For this purpose, while studying and investigating the relationships of previous researchers for predicting the compressive, tensile and bending strengths of fiber concrete reinforced with steel fibers, the optimal mathematical relationships governing the problem have been investigated by machine learning method. The focus of the current research is on those concretes reinforced with steel fibers, which are made of smooth, wavy and double-crossed fibers on a macro scale. During this research, the mathematical relationships extracted from the machine learning method, which is used to model the prediction of compressive, tensile and bending strengths of concrete reinforced with steel fibers and by applying the symbolic regression method based on a database consisting of 2283 The provided international data is checked. The efficiency of the models used in this research has been measured using error analysis statistics such as RMSE and MAPE. The results show that the parameters of the size of the largest aggregate, modulus of elasticity, compressive strength of control concrete, water-cement ratio, volume percentage and length of fibers, dimensional ratio of fibers and tensile or bending strength related to the type of output investigated for fiber concrete have the greatest effect. They are resistant to prediction. Examining the results shows that the formulas presented for predicting the strength of reinforced concrete for the examined fibers have considerable accuracy compared to the previous mathematical relationships.

کلیدواژه‌ها [English]

  • Fiber Concrete
  • Machine Learning
  • Experimental Formulas
  • Steel Fibers
  • Modeling Algorithms

مراجع

[1] D. Mostofi Nejad, Technology and Mix Design of Concrete, Arkan Publications, 44th ed., 2016. (In Persian)
[2] A. Khaloo, N. Kim, Mechanical Properties of Normal to High-Strength Steel Fiber-Reinforced Concrete, ACI Cement, Concrete, and Aggregate, 18(2) (1996) 92–97.
[3] P.S. Song, S. Hwang, Mechanical properties of high-strength steel fiber-reinforced concrete, Construction and Building Materials, 18 (2004) 669–673.
[4] A.S. Ezeldin, P.N. Balaguru, Normal and high-strength fiber-reinforced concrete under compression, Journal of Materials in Civil Engineering, 4(4) (1992) 415–429.
[5] M.C. Nataraja, N. Dhang, A.P. Gupta, Stress–strain curves for steel-fiber reinforced concrete in compression, Cement and Concrete Composites, 21(5–6) (1999) 383–390.
[6] S. Yazıcı, G. Inan, V. Tabak, Effect of aspect ratio and steel fiber volume fraction on the mechanical properties of SFRC, Construction and Building Materials, 21(6) (2007) 1250–1253.
[7] J. Thomas, A. Ramaswamy, Mechanical Properties of Steel Fiber-Reinforced Concrete, Journal of Materials in Civil Engineering, 19(5) (2007) 385–392.
[8] Y. Ou, M. Tsai, K. Liu, K. Chang, Compressive behavior of steel-fiber-reinforced concrete with a high reinforcing index, Journal of Materials in Civil Engineering, 24(2) (2012) 207–215.
[9] H. Salehi, R. Burgueño, Emerging artificial intelligence methods in structural engineering, Engineering Structures, 171 (2018) 170–189.
[10] M. Açikgenç, M. Ulaş, K.E. Alyamaç, Using an artificial neural network to predict mix compositions of steel fiber-reinforced concrete, Construction and Building Materials, (2014).
[11] M. Moradi, A.R. Bagherieh, M.R. Esfahani, Relationship of Tensile Strength of Steel Fiber Reinforced Concrete Based on Genetic Programming, International Journal of Optimization in Civil Engineering, 6(3) (2016) 349–363.
[12] S. Lu et al., A Novel Feature Selection Approach Based on Tree Models for Evaluating the Punching Shear Capacity of Steel Fiber-Reinforced Concrete Flat Slabs, Materials, 13 (2020) 3902.
[13] H.B. Ly et al., Computational hybrid machine learning based prediction of shear capacity for steel fiber reinforced concrete beams, Sustainability, 12(7) (2020) 2709.
[14] J. Rahman, K.S. Ahmed, N.I. Khan, K. Islam, S. Mangalathu, Data-driven shear strength prediction of steel fiber reinforced concrete beams using machine learning approach, Engineering Structures, 233 (2021) 111743.
[15] E. Alotaibi, O. Mostafa, N. Nassif, M. Omar, M.G. Arab, Prediction of Punching Shear Capacity for Fiber-Reinforced Concrete Slabs Using Neuro-Nomographs Constructed by Machine Learning, Journal of Structural Engineering, 147(6) (2021) 04021075.
[16] O.B. Olalusi, P.O. Awoyera, Shear capacity prediction of slender reinforced concrete structures with steel fibers using machine learning, Engineering Structures, 227 (2021) 111470.
[17] M. Kang, D.Y. Yoo, R. Gupta, Machine learning-based prediction for compressive and flexural strengths of steel fiber-reinforced concrete, Construction and Building Materials, 266 (2021) 121117.
[18] Y. Li et al., Compressive Strength of Steel Fiber-Reinforced Concrete Employing Supervised Machine Learning Techniques, Materials, 15 (2022).
[19] D. Zheng et al., Flexural Strength Prediction of Steel Fiber-Reinforced Concrete Using Artificial Intelligence, Materials, 15 (2022).
[20] A. Shatnawi et al., Shear Strength Prediction of Slender Steel Fiber Reinforced Concrete Beams, Buildings, (2022).
[21] M. Ali et al., Genetic Programming-based Algorithms for Modeling SFRC at High Temperatures, Composites Part C, (2024).
[22] C. Cheng et al., Accurate Prediction of Punching Shear Strength of Steel Fiber-Reinforced Concrete Slabs, Buildings, (2024).
[23] M. Ali, L. Chen, Q. Qureshi, D. Alsekait, A. Khan, K. Arif, M. Luqman, D. Elminaam, A. Hamza, M. Khan, Genetic Programming-based Algorithms Application in Modeling the Compressive Strength of Steel Fiber-Reinforced Concrete Exposed to Elevated Temperatures, Composites Part C: Open Access, (2024).
[24] H. Chen, J. Yang, X. Chen, A convolution-based deep learning approach for estimating compressive strength of fiber reinforced concrete at elevated temperatures, Construction and Building Materials, (2021).
[25] E. Sadrossadat, H. Basarir, A. Karrech, M. Elchalakani, Multi-objective mixture design and optimisation of steel fiber reinforced UHPC using machine learning algorithms and metaheuristics, Engineering with Computers, 38 (2021) 2569–2582.
[26] J. Li, K. Cheng, S. Wang, F. Morstatter, R.P. Trevino, J. Tang, H. Liu, Feature selection: A data perspective, ACM Computing Surveys, 50(6) (2017) 1–45.
[27] E. Romero, J.M. Sopena, G. Navarrete, R. Alquézar, Feature selection forcing overtraining may help to improve performance, Proceedings of the International Joint Conference on Neural Networks, 3 (2003) 2181–2186.
[28] M.H. Taghavi Parsa, M.R. Adlparvar, M. Esmaeili, Machine Learning-Based Empirical Formulations for Strength Properties of Steel Fiber Reinforced Concrete, Journal of Rehabilitation in Civil Engineering, 13(1) (2024) 27–44.
[29] G.C. McDonald, Ridge regression, Wiley Interdisciplinary Reviews: Computational Statistics, 1(1) (2009) 93–100.
[30] ASTM, ASTM C150, Standard Test Method for Flexural Strength of Concrete (Using Simple Beam with Third-Point Loading), Philadelphia, 2009.
[31] ASTM, ASTM C1602, Mixing Water Used in the Production of Hydraulic Cement Concrete, Philadelphia, 2006.
[32] ASTM, ASTM A820/A820M, Standard Specification for Steel Fibers for Fiber-Reinforced Concrete, Philadelphia, 2009.
[33] ASTM, ASTM C29/C29M, Standard Test Method for Bulk Density and Voids in Aggregate, ASTM International, 2020.
[34] ASTM, ASTM C127, Standard Test Method for Relative Density (Specific Gravity) and Absorption of Coarse Aggregate, ASTM International, 2015.
[35] ASTM, ASTM C136/C136M, Standard Test Method for Sieve Analysis of Fine and Coarse Aggregates, ASTM International, 2014.
[36] ACI 544-1R, State-of-the-Art Report on Fiber Reinforced Concrete, ACI Journal Proceedings, 70(11) (1973).
[37] ASTM, ASTM C1017, Chemical Admixtures for Use in Producing Flowing Concrete, Philadelphia, 2007.
[38] F.F. Wafa, S.A. Ashour, Mechanical properties of high strength fiber reinforced concrete, ACI Materials Journal, 89(5) (1992) 449–455.
نشریه مهندسی عمران امیرکبیر
دوره 57، شماره 10
دی 1404
صفحه 1807-1838

فایل ها

هم رسانی

ارجاع به این مقاله

آمار