SIMULATION OF NEAR FAULT GROUND MOTIONS USING NEURO FUZZY NETWORKS AND WAVELET ANALYSIS

The existence of recorded accelerograms to perform dynamic inelastic time history analysis is of the utmost importance especially in near-fault regions where directivity pulses, known as the most important characteristics of these ground motions, impose extreme demands on structures and cause widespread damages. But due to the lack of recorded acceleration time histories, it is common to generate proper artificial ground motions. In this study, in order to generate near-fault pulse-like ground motions, first, it is proposed to extract velocity pulses from an ensemble of near-fault pulse-like ground motions and then simulate nonpulse-type ground motion using Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and wavelet packet transform (WPT). In the next step, the pulse-like ground motion is produced by superimposing directivity pulse on the previously generated nonpulse-type motion in a way that it is compatible with an specified near-field spectrum. Particle swarm optimization (PSO) is employed to optimize both the parameters of pulse model and cluster radius in subtractive clustering and principle component analysis (PCA) is used to reduce the dimension of ANFIS input vectors. Finally, a number of interpretive examples are presented to show how the proposed method works.

Keywords:

ANFIS , Artificial Records , Near-fault , Pulse-like , Wavelet Packet Transform