EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Machine learning-based compressive strength estimation in nanomaterial-modified lightweight concrete

Nashat S. Alghrairi, Farah N. Aziz, Suraya A. Rashid, Mohd Z. Mohamed, Amer M. Ibrahim

Open Engineering · 27.04.2024 · 14 (1)

DOI: 10.1515/eng-2022-0604 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Open Engineering, online erschienen am 27.04.2024. Verfasst von Nashat S. Alghrairi und 4 weiteren Autor:innen. Die Arbeit wurde bisher 24-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The development of nanotechnology has led to the creation of materials with unique properties, and in recent years, numerous attempts have been made to include nanoparticles in concrete in an effort to increase its performance and create concrete with improved qualities. Nanomaterials are typically added to lightweight concrete (LWC) with the goal of improving the composite’s mechanical, microstructure, freshness, and durability qualities. Compressive strength is the most crucial mechanical characteristic for all varieties of concrete composites. For this reason, it is essential to create accurate models for estimating the compressive strength (CS) of LWC to save time, energy, and money. In addition, it provides useful information for planning the construction schedule and indicates when the formwork should be removed. To predict the CS of LWC mixtures made with or without nanomaterials, nine different models were proposed in this study: the gradient-boosted trees (GBT), random forest, tree ensemble, XGBoosted (XGB), Keras, simple regression, probabilistic neural networks, multilayer perceptron, and linear relationship model. A total of 2,568 samples were gathered and examined. The most significant factors influencing CS during the modeling process were taken into account as input variables, including the amount of nanomaterials, cement, water-to-binder ratio, density, the content of lightweight aggregates, type of nano, fine and coarse aggregate content, and water. The performance of the suggested models was assessed using a variety of statistical measures, including the coefficient of determination ( R 2 ), scatter index, mean absolute error, and root-mean-squared error (RMSE). The findings showed that, in comparison to other models, the GBT model outperformed the others in predicting the compression strength of LWC mixtures enhanced with nanomaterials. The GBT model produced the best results, with the greatest value of R 2 (0.9) and the lowest value of RMSE (5.286). Furthermore, the sensitivity analysis showed that the most important factor influencing the prediction of the CS of LWC enhanced with nanoparticles is the water content.

Zitieren

APA

Alghrairi, N. S., Aziz, F. N., Rashid, S. A., Mohamed, M. Z., & Ibrahim, A. M. (2024). Machine learning-based compressive strength estimation in nanomaterial-modified lightweight concrete. Open Engineering, 14(1). https://doi.org/10.1515/eng-2022-0604

Vancouver

Alghrairi NS, Aziz FN, Rashid SA, Mohamed MZ, Ibrahim AM. Machine learning-based compressive strength estimation in nanomaterial-modified lightweight concrete. Open Engineering. 2024;14(1). doi:10.1515/eng-2022-0604

BibTeX
@article{Alghrairi2024,
  author = {Alghrairi, Nashat S. and Aziz, Farah N. and Rashid, Suraya A. and Mohamed, Mohd Z. and Ibrahim, Amer M.},
  title = {Machine learning-based compressive strength estimation in nanomaterial-modified lightweight concrete},
  journal = {Open Engineering},
  year = {2024},
  volume = {14},
  number = {1},
  doi = {10.1515/eng-2022-0604}
}
Kostenlos starten →