EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

AI-driven sustainable materials engineering for predicting mechanical properties and CO 2 emissions of basalt fiber-reinforced concrete

Tao Luo, Li Li, Xin Zhao, Bin Sun, Ali H. AlAteah, Mamoun Alqedra, Sahar A. Mostafa

REVIEWS ON ADVANCED MATERIALS SCIENCE · 23.01.2026 · 65 (1)

DOI: 10.1515/rams-2025-0325 ↗

Kurz gefasst

Fachartikel in der Zeitschrift REVIEWS ON ADVANCED MATERIALS SCIENCE, online erschienen am 11.11.2026. Verfasst von Tao Luo und 6 weiteren Autor:innen. Die Arbeit nennt 108 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Basalt fiber-reinforced concrete (BFRC) has emerged as a promising advanced construction material that enhances mechanical performance and durability through fiber reinforcement while contributing to sustainable development. However, accurately predicting its key properties remains challenging due to the complex and nonlinear interactions among mixture constituents and environmental factors. This study develops hybrid machine learning models to predict the compressive strength (CS), splitting tensile strength (STS), and CO 2 emissions of BFRC using a comprehensive dataset. Five hybrid models were implemented, including DNN–HHO, KNN–GA, GB–WOA, SVR–PSO, and RF–GWO. Among the individual hybrid models, the DNN–HHO model achieved the best predictive performance for both compressive strength (CS) and splitting tensile strength (STS), with high accuracy in testing ( R 2 = 0.9720 and 0.9544) and low error values (RMSE = 3.26 MPa and 0.417 MPa, respectively), while the ensemble prediction provided the highest overall predictive accuracy. The model exhibited minimal bias and strong stability, as reflected by near-zero mean residuals (−0.003) and low dispersion (2 σ = ±1.459 MPa). Furthermore, error analysis indicated weak dependency between prediction errors and output values ( R 2 = 0.2004), confirming controlled heteroscedasticity and consistent performance. Reliability assessment showed low variance in both training and testing phases, highlighting robust generalization capability. In addition, the DNN–HHO successfully captured the nonlinear relationship between compressive and tensile strength, outperforming conventional models such as ACI 318-19 and Eurocode 2, while showing a slight decrease in relative tensile capacity with increasing compressive strength. For cradle-to-gate CO 2 emissions prediction, the model achieved outstanding accuracy ( R 2 = 0.997) with most errors within ±5 kg/m 3 , confirming its high predictive accuracy and reliability. Finally, a user-friendly graphical user interface (GUI) was developed to deploy the optimized DNN–HHO model, enabling rapid and accurate prediction of BFRC properties and facilitating practical decision-making in sustainable concrete mix design.

Zitieren

APA

Luo, T., Li, L., Zhao, X., Sun, B., AlAteah, A. H., Alqedra, M., & Mostafa, S. A. (2026). AI-driven sustainable materials engineering for predicting mechanical properties and CO 2 emissions of basalt fiber-reinforced concrete. REVIEWS ON ADVANCED MATERIALS SCIENCE, 65(1). https://doi.org/10.1515/rams-2025-0325

Vancouver

Luo T, Li L, Zhao X, Sun B, AlAteah AH, Alqedra M, et al. AI-driven sustainable materials engineering for predicting mechanical properties and CO 2 emissions of basalt fiber-reinforced concrete. REVIEWS ON ADVANCED MATERIALS SCIENCE. 2026;65(1). doi:10.1515/rams-2025-0325

BibTeX
@article{Luo2026,
  author = {Luo, Tao and Li, Li and Zhao, Xin and Sun, Bin and AlAteah, Ali H. and Alqedra, Mamoun and Mostafa, Sahar A.},
  title = {AI-driven sustainable materials engineering for predicting mechanical properties and CO 2 emissions of basalt fiber-reinforced concrete},
  journal = {REVIEWS ON ADVANCED MATERIALS SCIENCE},
  year = {2026},
  volume = {65},
  number = {1},
  doi = {10.1515/rams-2025-0325}
}
Kostenlos starten →