New generalized ANN-based hybrid broadband response spectra generator using physics-based simulations

Estimation of the seismic risk associated with infrastructures requires site-specific hazard studies. Further, for nonlinear time history analysis, one broadband ground motion. In modern times, physics-based simulations (PBS) deriving motion future earthquakes have been considered. The PBS helps decrease uncertainties related to estimation compared prediction equations. methods a specific frequency threshold limit resulting from high computational demand. Hence, hybrid are required attain spectra simulated This study uses new artificial neural network (ANN)-based model generate using low-frequency spectral acceleration PBS, source, path, and site parameters as input variables. A detailed parametric performance evaluation was made identify optimal in conjunction best-suited ANN architecture. is demonstrated Iwate ( $$M_w$$ 6.9, 2008) earthquake. We found that predicted values developed agree recorded data. Furthermore, histories generated ordinate matching technique estimated spectra.