Predicting compressive strength of geopolymer concrete using novel DMOA-ELM model
REVIEWS ON ADVANCED MATERIALS SCIENCE · 23.01.2026 · 65 (1)
DOI: 10.1515/rams-2025-0300 ↗Kurz gefasst
Fachartikel in der Zeitschrift REVIEWS ON ADVANCED MATERIALS SCIENCE, online erschienen am 16.11.2026. Verfasst von Zhifei Lu und 2 weiteren Autor:innen. Die Arbeit nennt 87 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Cement concrete is one of the most frequently used construction materials in the world. However, its dependence on regular Portland cement leads to high energy consumption and significant greenhouse gas emissions. Therefore, alternative binders such as geopolymer materials have gained more and more interest as sustainable construction materials. In this study, hybrid machine learning models, DMOA-ELM and DMOA-MLP are used for prediction of compressive strength of geopolymer concrete (CoSGePC). The settings of Extreme Learning Machine (ELM) and Multi-Layer Perceptron (MLP) models were optimized by using the Dwarf Mongoose Optimization Algorithm (DMOA). The models were built with 295 data samples and nine input parameters including FA, GGBS, Na 2 SiO 3 , NaOH, FAg, Gravel 4/10, Gravel 10/20, WS and Molarity of NaOH. The model performances were assessed in terms of MAE, RMSE, R, VAF, NS, WI and SI. The results showed that DMOA-ELM gave the best prediction performance. In the test phase, the DMOA-ELM attained MAE, RMSE, NS, VAF, WI, R, and SI values of 1.96644, 2.3068, 0.98525, 98.52678 %, 0.8249, 0.993, and 0.0565, respectively. The external validation of 20 independent samples further proved the better generalization capability of DMOA-ELM. The MAE, RMSE, R, and SI values were 1.0224, 1.2652, 0.9886, and 0.0358, respectively. The Monte–Carlo method based uncertainty analysis also indicated that the DMOA-ELM had the lowest uncertainty value of 33.0937 % which was less than the permissible value of 35 %. Sensitivity analysis results revealed that FAg was the most influential parameter with r_ij = 0.928 while Gravel 4/10 was the least influential parameter with r_ij = 0.544. The results show that the proposed DMOA-ELM model is a reliable and robust method for forecasting CoSGePC and may be used to design sustainable geopolymer concrete mixtures.
Zitieren
APA
Lu, Z., Vatin, N. I., & Huang, J. (2026). Predicting compressive strength of geopolymer concrete using novel DMOA-ELM model. REVIEWS ON ADVANCED MATERIALS SCIENCE, 65(1). https://doi.org/10.1515/rams-2025-0300
Vancouver
Lu Z, Vatin NI, Huang J. Predicting compressive strength of geopolymer concrete using novel DMOA-ELM model. REVIEWS ON ADVANCED MATERIALS SCIENCE. 2026;65(1). doi:10.1515/rams-2025-0300
BibTeX
@article{Lu2026,
author = {Lu, Zhifei and Vatin, Nikolai Ivanovich and Huang, Jiandong},
title = {Predicting compressive strength of geopolymer concrete using novel DMOA-ELM model},
journal = {REVIEWS ON ADVANCED MATERIALS SCIENCE},
year = {2026},
volume = {65},
number = {1},
doi = {10.1515/rams-2025-0300}
}
EUVIMED