Document Type : Original Article
Authors
1
Assistant Professor, Department of Civil Engineering, Khorasgan branch, Islamic Azad University, Khorasgan, Isfahan, Iran
2
Department of Civil Engineering, University of Science and Technology of Mazandaran, Behshahr, Iran
3
Department of Civil Engineering, Khorasgan branch, Islamic Azad University, Khorasgan, Isfahan, Iran
Abstract
In this study, polypropylene fibers (0, 1, 3, and 5%) and bagasse ash (0, 10, 20, 30, 40, and 50%) were used as partial cement replacements, along with a superplasticizer at 2.5% of the cementitious materials. The mechanical strength and durability of the modified concrete were evaluated at 28 and 90 days. Additionally, the study explored the application of machine learning models, including CatBoost, LightGBM, and an Ensemble model, for predicting the compressive and flexural strength of fiber- and bagasse-reinforced concrete. The results indicated that the CP3B20 mix—containing 3% polypropylene fibers and 20% bagasse ash—demonstrated the best performance among the tested mixtures. It achieved a compressive strength of 41.7 MPa compared to 32 MPa for the control mix, a flexural strength of 14.9 MPa versus 5 MPa, a water absorption rate of 0.9% versus 5%, and permeability of 320 coulombs compared to 3149 coulombs for the control mix. The Ensemble model exhibited high accuracy, with an R² of 0.85–0.90 for compressive strength and 0.92 for flexural strength, and RMSE values of 1.50–2.00 MPa and 0.65–1.10 MPa, respectively.
Keywords
)
Send comment about this article