Identification of pulse-like ground motions using artificial neural network

Abstract

For more than 20 years, the concept of near-fault pulse-like ground motion has been a topic of great interest due to its distinct characteristics, particularly due to directivity or fling effects, which are hugely influenced by the rupture mechanism. These unexpected characteristics, along with their effective frequency, energy rate, and damage indices, create a near-fault, pulse-like ground motion capable of causing severe damage to structures. One of the most common approaches for identifying these ground motions is done by conducting wavelet decomposition of the ground motion time history to extract a pulse signal and eventually categorize an earthquake by comparing the original signal to the residual one. However, to overcome the intensive calculations required in this approach, this study proposes using artificial neural networks to identify pulse-like ground motions through classification to predict their pulse period by means of regression analysis. Furthermore, the study is intended to evaluate the reliability and accuracy of various artificial neural networks in identifying pulse-like ground motions and predicting their pulse periods. In general, the results of the study have shown that the artificial neural network can identify pulse-like earthquakes and reliably predict their pulse period.