Data-driven Kriging model for predicting concrete compressive strength and parameter correlation analysis
- Author
- Yifei Liu, Cao MaoSen and Magd Abdel Wahab (UGent)
- Organization
- Abstract
- The concrete compressive strength (CS) is an important parameter used for durability design and service life prediction of concrete structures in civil engineering projects. It usually has a high nonlinear relationship with the age and main components of concrete, which makes it difficult for traditional regression analysis methods to perform predictive modelling. This study presents a data-driven Kriging model for predicting concrete CS under standard curing period. Two popular machine learning algorithms, namely Artificial Neural Network (ANN) and Support Vector Regression (SVR), are used for comparisons to validate the predictive ability of Kriging model. In addition, a parameter correlation analysis is implemented to reveal the intrinsic association of the selected seven main components of concrete and concrete CS. This study led to the following conclusions: (1) compared with ANN and SVR, the data-driven Kriging model has the highest accuracy in predicting concrete CS, and (2) the results of the parameter correlation analysis coincide with the physical laws of concrete CS.
Downloads
-
(...).pdf
- full text (Published version)
- |
- UGent only
- |
- |
- 1.96 MB
Citation
Please use this url to cite or link to this publication: http://hdl.handle.net/1854/LU-01GSJNEX44K5CQZB4CXQ2PP1R0
- MLA
- Liu, Yifei, et al. “Data-Driven Kriging Model for Predicting Concrete Compressive Strength and Parameter Correlation Analysis.” Proceedings of the 5th International Conference on Numerical Modelling in Engineering : Volume 1 : Numerical Modelling in Civil Engineering, NME 2022, edited by Magd Abdel Wahab, vol. 311, Springer, 2023, pp. 119–28, doi:10.1007/978-981-19-8429-7_11.
- APA
- Liu, Y., MaoSen, C., & Abdel Wahab, M. (2023). Data-driven Kriging model for predicting concrete compressive strength and parameter correlation analysis. In M. Abdel Wahab (Ed.), Proceedings of the 5th International Conference on Numerical Modelling in Engineering : volume 1 : Numerical Modelling in Civil Engineering, NME 2022 (Vol. 311, pp. 119–128). https://doi.org/10.1007/978-981-19-8429-7_11
- Chicago author-date
- Liu, Yifei, Cao MaoSen, and Magd Abdel Wahab. 2023. “Data-Driven Kriging Model for Predicting Concrete Compressive Strength and Parameter Correlation Analysis.” In Proceedings of the 5th International Conference on Numerical Modelling in Engineering : Volume 1 : Numerical Modelling in Civil Engineering, NME 2022, edited by Magd Abdel Wahab, 311:119–28. Singapore: Springer. https://doi.org/10.1007/978-981-19-8429-7_11.
- Chicago author-date (all authors)
- Liu, Yifei, Cao MaoSen, and Magd Abdel Wahab. 2023. “Data-Driven Kriging Model for Predicting Concrete Compressive Strength and Parameter Correlation Analysis.” In Proceedings of the 5th International Conference on Numerical Modelling in Engineering : Volume 1 : Numerical Modelling in Civil Engineering, NME 2022, ed by. Magd Abdel Wahab, 311:119–128. Singapore: Springer. doi:10.1007/978-981-19-8429-7_11.
- Vancouver
- 1.Liu Y, MaoSen C, Abdel Wahab M. Data-driven Kriging model for predicting concrete compressive strength and parameter correlation analysis. In: Abdel Wahab M, editor. Proceedings of the 5th International Conference on Numerical Modelling in Engineering : volume 1 : Numerical Modelling in Civil Engineering, NME 2022. Singapore: Springer; 2023. p. 119–28.
- IEEE
- [1]Y. Liu, C. MaoSen, and M. Abdel Wahab, “Data-driven Kriging model for predicting concrete compressive strength and parameter correlation analysis,” in Proceedings of the 5th International Conference on Numerical Modelling in Engineering : volume 1 : Numerical Modelling in Civil Engineering, NME 2022, Ghent, Belgium, 2023, vol. 311, pp. 119–128.
@inproceedings{01GSJNEX44K5CQZB4CXQ2PP1R0,
abstract = {{The concrete compressive strength (CS) is an important parameter used for durability design and service life prediction of concrete structures in civil engineering projects. It usually has a high nonlinear relationship with the age and main components of concrete, which makes it difficult for traditional regression analysis methods to perform predictive modelling. This study presents a data-driven Kriging model for predicting concrete CS under standard curing period. Two popular machine learning algorithms, namely Artificial Neural Network (ANN) and Support Vector Regression (SVR), are used for comparisons to validate the predictive ability of Kriging model. In addition, a parameter correlation analysis is implemented to reveal the intrinsic association of the selected seven main components of concrete and concrete CS. This study led to the following conclusions: (1) compared with ANN and SVR, the data-driven Kriging model has the highest accuracy in predicting concrete CS, and (2) the results of the parameter correlation analysis coincide with the physical laws of concrete CS.}},
author = {{Liu, Yifei and MaoSen, Cao and Abdel Wahab, Magd}},
booktitle = {{Proceedings of the 5th International Conference on Numerical Modelling in Engineering : volume 1 : Numerical Modelling in Civil Engineering, NME 2022}},
editor = {{Abdel Wahab, Magd}},
isbn = {{9789811984280}},
issn = {{2366-2557}},
language = {{eng}},
location = {{Ghent, Belgium}},
pages = {{119--128}},
publisher = {{Springer}},
title = {{Data-driven Kriging model for predicting concrete compressive strength and parameter correlation analysis}},
url = {{http://doi.org/10.1007/978-981-19-8429-7_11}},
volume = {{311}},
year = {{2023}},
}
- Altmetric
- View in Altmetric