EUVIMEDEUVIMED entdecken
Crossref

Enhanced Predictive Modeling and Insights into Geopolymer Concrete Compressive Strength Prediction

Hai-Bang Ly

Journal of Testing and Evaluation · 01.07.2025 · 53 (4) · 938-959

DOI: 10.1520/jte20230808 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Testing and Evaluation, online erschienen am 14.02.2025. Verfasst von Hai-Bang Ly. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 55 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The primary objective of this investigation is to construct a machine learning (ML) model to enhance the predictive accuracy of geopolymer concrete (GPC) compressive strength (CS) while extracting valuable insights from input variables. Considering the crucial task of optimizing hyperparameters, three efficient optimization algorithms are harnessed for this purpose: the FOX optimization algorithm (FOX), the golden jackal optimization algorithm, and the honey badger algorithm. The dataset under examination incorporates 156 experimental samples comprising 15 variables. Through a systematic exploration of population size and epoch parameters, hyperparameters are fine-tuned for the 12 XGB models. Notably, model XGB_06, which employs FOX with a population size of 20, attains the highest coefficient of determination value of 0.9662 on the testing dataset, achieving a fine equilibrium between computational efficiency and superior accuracy. This model’s excellence is further proved by its outperformance of eight well-known ML algorithms during the comparative analysis step. In addition, this study presents the results of partial dependence plots, shedding light on the intricate relationships between input variables and GP’s CS. The insights gleaned here have far-reaching implications, offering avenues for refining the predictive modeling of GPC’s CS and extending its applications within materials science and engineering.

Zitieren

APA

Ly, H. (2025). Enhanced Predictive Modeling and Insights into Geopolymer Concrete Compressive Strength Prediction. Journal of Testing and Evaluation, 53(4), 938-959. https://doi.org/10.1520/jte20230808

Vancouver

Ly H. Enhanced Predictive Modeling and Insights into Geopolymer Concrete Compressive Strength Prediction. Journal of Testing and Evaluation. 2025;53(4):938-959. doi:10.1520/jte20230808

BibTeX
@article{Ly2025,
  author = {Ly, Hai-Bang},
  title = {Enhanced Predictive Modeling and Insights into Geopolymer Concrete Compressive Strength Prediction},
  journal = {Journal of Testing and Evaluation},
  year = {2025},
  volume = {53},
  number = {4},
  pages = {938--959},
  doi = {10.1520/jte20230808}
}
Kostenlos starten →