Machine Learning-Based Approach for Modelling Soil-Structure Interaction Effects on Reinforced Concrete Structures Subjected to Earthquake Excitations

Abstract

In the chapter, the authors develop a machine learning (ML)-based model that has the potential to make rapid predictions for seismic responses with SSI effects and determine the seismic performance levels. The authors select several input parameters for training, validation, and testing of the present model. The present high speed and accurate data generation methods can be incorporated as a tool for safe seismic assessment and design of sustainable earthquake resistant structures. Finally, the authors will test the soil-pile model experimentally on a shake table (with strain gauges and accelerometers) when subjected to harmonic load with varying frequencies in the range 3Hz to 12Hz and base acceleration ranging from 0.05g to 0.3g. The present approach shall provide substantial information for design of piles and the response of piles subjected to earthquake excitations.