Machine Learning Prediction of Reinforcement Bond Strength in Fly ASH-GGBS Based Alkali-Activated Concrete
Journal for Research in Applied Sciences and Biotechnology · 25.06.2026 · 5 (3) · 63-71
DOI: 10.55544/jrasb.5.3.8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal for Research in Applied Sciences and Biotechnology, online erschienen am 25.06.2026. Verfasst von Amar Pal und 2 weiteren Autor:innen. Die Arbeit nennt 17 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
The environmental baggage of making ordinary Portland cement (OPC) is becoming a greater concern, spurring the industry to develop greener alternatives such as alkali-activated concrete (AAC). Here we examine the bond strength of AAC derived from fly ash and ground granulated blast furnace slag (GGBS) and propose machine learning (ML) as a way to forecast that performance with some exactitude. Our approach was to subject reinforced AAC specimens to pull-out tests, changing up the fly ash to GGBS ratio along with the activator modulus and curing temperature. We also used scanning electron microscopy (SEM) to get a microstructural read on the pore characteristics and the interfacial transition zone (ITZ), which are what dictate the interaction between the steel and the concrete. All the data from these experiments and the SEM were put through three ML algorithms: an Artificial Neural Network (ANN), Support Vector Regression (SVR) and Gradient Boosting Regression (GBR). We evaluated their output by the R², RMSE and MAPE figures. What we found is that a well-balanced mix of fly ash and GGBS in the AAC leads to gel densification and less porosity, and with it superior bond strength. The GBR model proved to be the most reliable of the bunch with an R² of 0.95. In the end, our results indicate you can use an ML-driven method to design and vet the structural soundness of low-carbon AAC systems without having to depend so heavily on the lab.
Zitieren
APA
Pal, A., Singh, M., & Sharma, R. (2026). Machine Learning Prediction of Reinforcement Bond Strength in Fly ASH-GGBS Based Alkali-Activated Concrete. Journal for Research in Applied Sciences and Biotechnology, 5(3), 63-71. https://doi.org/10.55544/jrasb.5.3.8
Vancouver
Pal A, Singh M, Sharma R. Machine Learning Prediction of Reinforcement Bond Strength in Fly ASH-GGBS Based Alkali-Activated Concrete. Journal for Research in Applied Sciences and Biotechnology. 2026;5(3):63-71. doi:10.55544/jrasb.5.3.8
BibTeX
@article{Pal2026,
author = {Pal, Amar and Singh, Mayankeshwar and Sharma, Ritu},
title = {Machine Learning Prediction of Reinforcement Bond Strength in Fly ASH-GGBS Based Alkali-Activated Concrete},
journal = {Journal for Research in Applied Sciences and Biotechnology},
year = {2026},
volume = {5},
number = {3},
pages = {63--71},
doi = {10.55544/jrasb.5.3.8}
}
EUVIMED