MACHINE LEARNING-BASED INVESTIGATION ON SEISMIC ANALYSIS OF A G+12-STORY BUILDING BY USING ETABS SOFTWARE WITH A SHEAR WALL: RESPONSE SPECTRUM ANALYSIS

Abstract

Globally, nowadays the population is growing increasingly, and engineers are focusing on constructing high-rise buildings. High-rise buildings should resist the lateral loads like earthquake and wind loads. Earthquakes are commonly occurring worldwide, releasing an immense amount of energy, which shows the impact on high-rise buildings that move laterally. To avoid these, high-rise buildings are designed by providing shear walls. In the present study is analyzing a G+12 building by providing it without and with a shear wall by using response spectrum analysis in ETABS software. A shear wall is a vertical cantilever beam that resists the lateral loads like seismic load. The main objective of this study is to examine the structural vulnerable factors such as story drift and story displacement. The analysis is carried out by considering seismic zone IV, which has a zone factor of 0.24, and the soil type is medium. IS 1893:2016 guidelines are used for response spectrum analysis. The random forest regressor outperformed other models in terms of accuracy in forecasting and generalization faults for important structural characteristics. ETABS evaluation found that story height and shear wall layout and RSA were the most relevant factors, supporting standard building engineering concepts. In the present study obtained R2=1 in two cases, then this investigation concludes that the best fitting for machine learning model.