New Predictive Model for the Computation of Reinforced Concrete Column Shear Strength

Abstract

The computing of the fundamental period of structures during seismic design is well documented in 13 design codes but is mainly dependent on the height of the structure, which is considered to be the most influential 14 parameter. It is, however, important to consider a phenomenon called the soil-structure interaction (SSI), as this 15 has been found to have a detrimental effect, especially for buildings founded on soft soils. A pilot research project 16 foresaw the use of machine learning (ML) algorithms trained on relatively limited data sets for the development 17 of a more accurate and objective fundamental period formula. Therefore, a data set that consists of 98,308 18 fundamental period data points was created through the use of a High-Performance Computer (HPC), which is 19 the largest data set of its kind. The HPC results were then used to train, test, and validate different ML algorithms. 20 It was found that XGBoost-HYT-CV with hyperparameter tuning performed the best with a correlation of 99.99% 21 and a Mean Average Percentage Error (MAPE) of 0.5%. Furthermore, the XGBoost-HYT-CV model 22 outperformed all under-study ML models when using an additional data set that consisted of out-of-sample 23 building geometries and soil properties, with a resulting MAPE of 9%. Finally, irregular buildings were also used 24 to test the performance of the proposed predictive models.