Yikun Li

Nano-silica modified concrete (NSC) has been widely applied in engineering practice. However, conventional manual mix proportion design is both time-consuming and costly. In this study, four machine learning models—XGBoost, CatBoost, Random Forest, and AdaBoost—were trained to predict the compressive strength of NSC. Based on the best-performing model, the NSGA-II algorithm was employed to develop a multi-objective optimization framework, considering compressive strength, cost, and carbon emissions as objectives. The results indicated that XGBoost achieved the highest accuracy, with R² = 0.99 and RMSE = 1.80 MPa. Feature importance analysis further revealed that nano-silica content was strongly correlated with strength (0.82) and cost (0.85). Using NSGA-II, a set of Pareto-optimal solutions was generated. The NSGA-II algorithm produced Pareto-optimal solutions, highlighting the trade-offs among the three objectives. This integrated approach effectively reduces experimental workload and provides a valuable reference for sustainable NSC mix proportion design.

Nano-modified concrete has attracted wide attention due to its improved mechanical properties. Among them, compressive strength is the most critical indicator. However, testing nano-concrete is costly and complex because it requires control over many factors, such as nanoparticle content and dispersion. Machine learning offers a data-driven way to predict compressive strength more efficiently. It reduces trial-and-error efforts and supports mix design optimization. Currently, machine learning is more adept at handling complicated datasets than experimental and traditional statistical models. In this article, the development of machine learning research in predicting the strength of concrete enhanced by nanoparticles is reviewed. First, we systematically outline a three-phase ML framework encompassing data curation, model development, and validation protocols; next, popular algorithms and their uses in predicting the strength of nano-modified concrete are evaluated, such as Artificial Neural Networks, K-Nearest Neighbor, Random Forest, etc. Ultimately, the article offers a forward-looking perspective on how future machine learning advancements can foster and accelerate the development of nano-modified concrete.