A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
K-Fold Cross-Validation Technique for Predicting Ultimate Compressive Strength of Circular CFST Columns
Concrete-Filled Steel Tube (CFST) columns are a type of composite structure that has been widely used nowadays. It is known for its ability to withstand ultimate compressive strength and its efficient responsiveness in buildings with a modern appearance that transcends large spatial spans. Many design standards of many countries have also proposed calculation formulas, but design equations meet limitations due to experimental tests. If experimental data is large, specifically 663 specimens of CFST columns, determining the critical compressive strength for each sample is reliable and convenient in the case of Machine Learning prediction. In this study, prediction models of Supervised Learning approaches in Machine Learning such as Linear Regression, Logistic Regression, Linear Support Vector Regressor, Random Forest, Decision Tree, and KNN will be performed effectively in predicting the ultimate strength of CFST columns with only experimental raw data. The results obtained by this study show different influences in using the prediction models of the supervised learning approaches. Besides, to increase the stability of the prediction models, a cross-validation technique called K-fold is used with the data set divided into two parts, including 80% for training data and 20% for test data.
K-Fold Cross-Validation Technique for Predicting Ultimate Compressive Strength of Circular CFST Columns
Concrete-Filled Steel Tube (CFST) columns are a type of composite structure that has been widely used nowadays. It is known for its ability to withstand ultimate compressive strength and its efficient responsiveness in buildings with a modern appearance that transcends large spatial spans. Many design standards of many countries have also proposed calculation formulas, but design equations meet limitations due to experimental tests. If experimental data is large, specifically 663 specimens of CFST columns, determining the critical compressive strength for each sample is reliable and convenient in the case of Machine Learning prediction. In this study, prediction models of Supervised Learning approaches in Machine Learning such as Linear Regression, Logistic Regression, Linear Support Vector Regressor, Random Forest, Decision Tree, and KNN will be performed effectively in predicting the ultimate strength of CFST columns with only experimental raw data. The results obtained by this study show different influences in using the prediction models of the supervised learning approaches. Besides, to increase the stability of the prediction models, a cross-validation technique called K-fold is used with the data set divided into two parts, including 80% for training data and 20% for test data.
K-Fold Cross-Validation Technique for Predicting Ultimate Compressive Strength of Circular CFST Columns
Lecture Notes in Civil Engineering
Reddy, J. N. (editor) / Wang, Chien Ming (editor) / Luong, Van Hai (editor) / Le, Anh Tuan (editor) / Nguyen, Tran-Trung (author) / Nguyen, Phu-Cuong (author)
2022-09-21
8 pages
Article/Chapter (Book)
Electronic Resource
English
CFST Columns , Decision Tree , KNN , Linear Regression , Machine Learning , Prediction Models , Supervised Learning Energy , Sustainable Architecture/Green Buildings , Structural Materials , Geotechnical Engineering & Applied Earth Sciences , Building Construction and Design , Construction Management , Environmental Policy , Engineering
Nonlinear ANN Modeling for Predicting Ultimate Strength of CFST Columns
| Springer Verlag | 2024
Study on Long-Term Behavior and Ultimate Strength of CFST Columns
| British Library Conference Proceedings | 2009
Compressive behavior of circular CFST columns externally reinforced using CFRp composites
| Online Contents | 2015