UTILIZATION OF ARTIFICIAL INTELLIGENCE IN SEISMIC ANALYSIS OF REINFORCED CONCRETE FRAMED BUILDINGS

Abstract

Earthquakes are among the most catastrophic natural disasters, causing significant losses in life and property worldwide. The primary cause of structural damage during an earthquake is the unpredictable lateral loads imposed on buildings. To mitigate such damage, the structural framework must be robust enough to withstand these loads. Advances in software such as ETABS (Extended Three-dimensional Analysis of Building Systems) have enhanced the understanding of structural behavior under seismic loads. Recently, the integration of artificial intelligence (AI) techniques, particularly artificial neural networks (ANNs), has shown promise in improving the seismic analysis of reinforced concrete framed buildings. This paper focuses on the application of ANN in predicting the base shear and deflection of an eight-story reinforced concrete building modeled using ETABS. The research involves training an ANN model using data generated from ETABS simulations, followed by validation of the ANN predictions against ETABS results. Key metrics such as Scatter Index (SI), root-mean-squared error (RMSE), and correlation coefficient (R²) are utilized to evaluate the accuracy of the ANN model. The study highlights the challenges in AI model development, including data quality and generalization to different contexts. The findings demonstrate that the ANN model achieves high prediction accuracy, with a scatter index of less than 1% for both base shear and deflection, and v a correlation coefficient close to 1. These results underscore the potential of ANN as a reliable tool for early prediction of structural response during seismic events. The study concludes by suggesting that ANN-based approaches can significantly enhance seismic analysis, offering a more efficient and accurate alternative to traditional methods, with future potential for real-time structural health monitoring and assessment.