Transfer learning on stacked machine-learning model for predicting pull-out behavior of steel fibers from concrete
Published: 26 June 2025| Version 2 | DOI: 10.17632/8fv65cyhz6.2
Contributors:
Torkan Shafighfard, Neda Asgarkhani, Farzin Kazemi, Dooyeol YooDescription
A dataset was compiled with nine inputs, including fiber type (F-T), diameter (d), aspect ratio (l/d), inclination angle (θ), embedment length (Le), fiber tensile strength (σt), loading rate (V), water-to-cement ratio (W/C), and concrete compressive strength (σc). The output parameters were the average bond strength (τavg), equivalent bond strength (τeq), pullout energy (W), and peak load (Fmax) with the corresponding slip (S). A total of 472 data points were compiled through a literature review of 47 peer-reviewed academic articles (see [21, 47–92]). For more information, please see the paper linked to this dataset.
Files
Institutions
- Politechnika Gdanska
- Yonsei University
Categories
Data Science, Machine Learning, Concrete Technology, Fiber, Transfer Learning