FraMCoS 11 2023 Bangalore, India

Machine learning regression models for peak shear strength prediction of squat shear walls

A Reinforced Concrete wall is designed to contribute to most if not the whole lateral load carrying capacity of the structure. The lateral load carrying capacity of such wall is indicated by the peak shear strength of the walls. The…

First page of: Machine learning regression models for peak shear strength prediction of squat shear walls
Year 2023
Downloads 1
File Size 12.9 MB
Download PDF (12.9 MB)

Abstract

A Reinforced Concrete wall is designed to contribute to most if not the whole lateral load carrying capacity of the structure. The lateral load carrying capacity of such wall is indicated by the peak shear strength of the walls. The strength of Slender Walls can be found fairly accurately by section analysis which has been studied quite extensively and has been reported by most of the codes.Howeverthecodesfailtoprovideanaccuratemethodfordeterminingthepeakshearstrengthof Squat Walls. Various theories based on mechanics developed to study the behavior of concrete have been applied to structural walls which give accurate results but are time-consuming and resource- heavy. This complexity in peak shear strength calculation has been attributed to various parameters affecting the behavior of such walls dominated by shear. Machine learning models Polynomial Re- gression, KNN Regression, Decision Tree Regression, Random Forest and Boosting Method have been done on a database of 594 Squat Structural Walls.The accuracy of these models has been re- ported and the importance of parameters has been found.