Application of wavelet neural network for dynamic analysis of moment resisting frames

Wavelet neural network (WNN), as a data-driven technique, is applied for performing the dynamic analysis of moment resisting frames. Many researchers have recommended the empirical equation of fundamental period of vibration to estimate base shear close to the actual value. Data set of 52 buildings is generated using ETABS software. The data set is used to develop three WNN models to estimate fundamental period of vibration, maximum base shear and the top floor displacement. An empirical equation for fundamental period is then obtained from the predicted period values in terms of the height of the building. The equation is then modified for the error on the basis of the experiment performed on mild steel frames. An error of 10.75% is observed in the analytical and experimental values of the period of vibration. It is seen that WNN equation of period of vibration is closer to the measured values than that obtained from the equations recommended by other researchers. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.