Multivariate analysis of variance in nano-silica in concrete evolution: modelling strength and sustainability

This comprehensive research traces the evolution of concrete technology, focusing on nanotechnology, specifically nano-silica, as a highly promising avenue for enhancing concrete properties. The study systematically compares traditional concrete with nano-silica-reinforced concrete, shedding light on the pivotal roles of superplasticizers and nano-silica in determining compressive strength over a curing period ranging from 1 to 365 days. The analysis encompasses key factors such as the water-to-cement ratio, cement content (C), and content (S), gravel content (G), superplasticizer (SP), and Nano silica (NS), totaling 820 meticulously collected, analyzed, and modeled datasets.This research employs extensive datasets and diverse modeling techniques to predict compressive strength accurately. Key findings underscore the influence of the water-cement ratio and superplasticizers in traditional concrete, while nano-silica consistently interacts with other factors, except for curing time. The study presents numerical models for compressive strength estimation and contributes to sustainable construction practices. Utilizing statistical modeling, the research establishes optimal models with minimal root mean square error (RMSE). Correlation analysis reveals nuanced connections between traditional and nano-silica-containing concrete, with a marginal strength difference not exceeding 5 MPa. Various models, including nonlinear regression, full quadratic models, and an artificial neural network (ANN), are employed to predict compressive strength. Significantly, the study finds that the Artificial Neural Network (ANN) model consistently outperforms other models in predicting the compressive strength of conventional concrete, while the Full Quadratic (FQ) model exhibits remarkable consistency, especially in forecasting the strength of traditional concrete. Sensitivity analysis underscores the pivotal roles of factors such as water-cement ratio, cement content, and superplasticizer in influencing model accuracy. Notably, nano-silica, identified through sensitivity analysis, significantly contributes to predictive accuracy, highlighting its unique and influential role in shaping concrete strength. This research deepens our understanding of the multifaceted factors influencing nano-silica-infused concrete strength, emphasizing the necessity to consider multiple variables for precise predictions.