Enhancing M30 concrete performance using tamarind seed polysaccharide and GGBFS with experimental validation, COMSOL Multiphysics simulation and machine learning prediction
摘要
Concrete plays an important role in construction; however, conventional mixtures often face challenges related to strength and durability. This study investigated the impact of partially substituting ground granulated blast furnace slag (GGBFS) with varying amounts of tamarind seed polysaccharide (TSP), a natural biopolymer, on the performance of M30 grade concrete. Experiments were conducted to assess the compressive strength, elastic modulus, and flexural strength at 7, 14, and 28 days intervals. The results showed an increase in compressive strength over time, with Sample 3 reaching 26.9 MPa at 7 days, 33.2 MPa at 14 days, and 35.2 MPa at 28 days. The simulated outcomes were comparable, forecasting 37.1 MPa at 28 days for the same sample. The beam deflections under load were nearly identical in both the experimental and simulated scenarios, differing by less than 0.002 mm, thus validating the accuracy of the simulations. Machine learning models, such as Random Forest, XGBoost, and SVM, were trained on the data to predict the mechanical properties, with Random Forest demonstrating superior performance, achieving an R2 of 0.99 and an MAE as low as 0.25 in strength prediction. This holistic approach, which combines experimental, computational, and AI techniques, highlights the potential of TSP and GGBFS for developing sustainable concrete mixes with enhanced mechanical properties, thus supporting more environmentally friendly construction practices.