Civil Engineering and Architecture Vol. 12(5), pp. 3193 - 3206
DOI: 10.13189/cea.2024.120506
Reprint (PDF) (1651Kb)

Prediction of the Compressive and Tensile Strengths of Geopolymer Concrete Using Artificial Neural Networks

Ali A. Mahameid , Amjad A. Yasin ^*, Ahmad B. Malkawi
Department of Civil Engineering, Faculty of Engineering Technology, Al-Balqa Applied University, Jordan

ABSTRACT

Geopolymer concrete is an environmentally friendly alternative to traditional Portland cement concrete. This research investigates the use of Artificial Neural Networks (ANN) to predict the compressive and tensile strengths of such concrete. A strict materials selection was applied by assessing the use of fly ash class-F and Ground Granulated Blast Furnace Slag (GGBS) as geopolymer source materials. The ANN model performed exceptionally well with 75 different concrete mix combinations, generating an extremely low Mean Squared Error (MSE) of 2.9x10^-5, suggesting a scant 2% variation between predictions and targets. The study demonstrates a strong agreement between the ANN predictions and the experimental values across a wide range of concrete strengths (10 to 80 MPa), guaranteeing a complete dataset. Regression analysis demonstrates the model's dependability, with correlation coefficients (R) of 0.993, 0.819, and 0.956 for the training, testing, and validation datasets, respectively. A constant R-value of 0.932 across all datasets adds to the ANN model's accuracy. The model's dependability in predicting geopolymer concrete strengths was confirmed by predicting a new dataset extracted from the literature, which yielded high agreement with a maximum error of 3%.

KEYWORDS
Geopolymer, Concrete, Fly Ash, GGBS, Strength, ANN

Cite This Paper in IEEE or APA Citation Styles
(a). IEEE Format:
[1] Ali A. Mahameid , Amjad A. Yasin , Ahmad B. Malkawi , "Prediction of the Compressive and Tensile Strengths of Geopolymer Concrete Using Artificial Neural Networks," Civil Engineering and Architecture, Vol. 12, No. 5, pp. 3193 - 3206, 2024. DOI: 10.13189/cea.2024.120506.

(b). APA Format:
Ali A. Mahameid , Amjad A. Yasin , Ahmad B. Malkawi (2024). Prediction of the Compressive and Tensile Strengths of Geopolymer Concrete Using Artificial Neural Networks. Civil Engineering and Architecture, 12(5), 3193 - 3206. DOI: 10.13189/cea.2024.120506.