Assessment of hybrid fiber reinforced graphene nano-engineered concrete composites: From experimental testing to explainable machine learning modeling

The exceptional characteristics of graphene, relating to its chemical, mechanical, and thermal properties, make it a multifunctional material suitable for a variety of applications in construction materials. This research comprehensively examined the fresh, mechanical, and microstructural properties of graphene nano-engineered hybrid fiber reinforced concrete (GNFRC) incorporating 0-0.06% of graphene and 0-0.5% of hybrid fibers (steel, glass, and polypropylene). The experimental results demonstrate that the optimal concrete mix containing 0.06% graphene, 0.5 vol.% polypropylene, and 0.5 vol.% steel fibers exhibited a maximum of 22%, 36%, and 14% increase in compressive strength, tensile strength, and elastic modulus relative to the control at 90 days. Additionally, the scanning electron microscopy coupled with energy-dispersive x-ray spectroscopy images of the concrete specimens indicated the development of amorphous intermediate hydration products is the main reason for the growth of long-term mechanical properties of the concrete mixes. This study also investigates three machine learning techniques (k-Nearest Neighbors, polynomial regression and XGBoost) to predict the compressive strength of GNFRC utilizing datasets comprising 438 different mix proportions of GNFRC. Among them, the XGBoost model showed highest accuracy with an R 2 of 0.994. The use of SHapley Additive exPlanations (SHAP) and Partial Dependency Plot (PDP) analyses identified the most influential mix elements as cement and curing age on compressive strength. The findings ensure the potential of graphene and hybrid fibers to develop durable, high-performance concrete composites suitable for demanding structural applications.