Application of Artificial Neural Network for Efficient Earthquake Resistant Design of Building: A Study

Abstract : This research paper presents the development and implementation of an Artificial Neural Network (ANN) model for the design of efficient earthquake-resistant reinforced concrete buildings. With the increasing frequency and severity of seismic events worldwide, it is imperative to enhance the seismic performance of structures to ensure the safety of occupants and the preservation of property. Traditional design methods often rely on simplified assumptions and empirical equations, which may not fully capture the complexities of seismic behavior. By utilizing a comprehensive dataset of structural designs, EQ records and performance evaluations, the ANN model has been trained to generate efficient and EQ resilient structural designs while meeting safety standards. In this paper, for development of ANN model huge data required such as input and output data, for input data various earthquake ground motion (EQGM) data are considered here and structural design parameter collected from Finite Element Software (ETABS) as output data. Using ETABS 3-D 2 Story RC frame building is designed and Time History Analysis has been performed using different EQ ground motion data. Training of ANN model has been done using INPUT data such as PGA, PGV, PGD, and Time duration of EQGM and design parameter of RC building as OUTPUT data such as Maximum Displacement, Story drift, Base shear, Overturning moment. After getting a robust ANN model the same model can be used for different seismic zone to predict the design parameter of RC structure for new EQGM data efficiently. Prediction of design parameter from ANN model for new EQGM can be compared with the design parameter of RC building from ETABS. Therefore, efficient earthquake resistant design and safety of RC building can be ensured in different seismic zone with less computational effort. The developed model works fine and the procedure adopted here promises future application and research.