Nonlinear Load-Deflection Analysis of Steel Rebar-Reinforced Concrete Beams: Experimental, Theoretical and Machine Learning Analysis

The integration of cutting-edge technologies into reinforced concrete (RC) design is reshaping the construction industry, enabling smarter and more sustainable solutions. Among these, machine learning (ML), a subset of artificial intelligence (AI), has emerged as a transformative tool, offering unprecedented accuracy in prediction and optimization. This study investigated the flexural behavior of steel rebar RC beams, focusing on varying concrete compressive strengths via theoretical, experimental and ML analysis. Nine steel rebar RC beams with low (SC20), moderate (SC30) and high (SC40) concrete compressive strength, measuring 150 × 200 × 1100 mm, were produced and subjected to three-point bending tests. An average error of less than 5% was obtained between the theoretical calculations and the experiments of the ultimate load-carrying capacity of reinforced concrete beams. By combining three-point bending experiments with ML-powered prediction models, this research bridges the gap between experimental insights and advanced analytical techniques. A groundbreaking aspect of this work is the deployment of 18 ML regression models using Python’s PyCaret library to predict deflection values with an impressive average accuracy of 95%. Notably, the K Neighbors Regressor and Gradient Boosting Regressor models demonstrated exceptional performance, providing fast, consistent and highly accurate predictions, making them an invaluable tool for structural engineers. The results revealed distinct failure mechanisms: SC30 and SC40 RC beams exhibited ductile flexural cracking, while SC20 RC beams showed brittle shear cracking and failure with sudden collapse.

Cette oeuvre a été publiée sous la license Creative Commons Attribution 4.0 (CC-BY 4.0). Il est autorisé de partager et adapter l'oeuvre tant que l'auteur est crédité et la license est indiquée (avec le lien ci-dessus). Vous devez aussi indiquer si des changements on été fait vis-à-vis de l'original.