Use of the artificial neural network for peak ground acceleration estimation

Abstract EN

The aim of this paper is to estimate the maximum Peak Ground Acceleration (PGA) of the three components (vertical, east-west and north-south) using the feed-forward artificial neural network method (ANN) with a conjugate gradient backpropagation rule for the training.

The inputs are the magnitude, the focal depth, the epicentral distance, the thickness of the sedimentary layers below the site down to a shear wave velocity equal to 800 m/s and the corresponding resonant frequency, while the target result is the PGA.

Data collected from the KiK-net seismic data base in Japan have been used.

1850 records at 102 sites are considered in the training phase, while 326 records are kept for the test phase.

The obtained results show that PGA computed using the ANN method are close to those recorded.

Finally, a simple example is presented in which 55 records are used to compare the ANN method with two Ground Motion Prediction Equations (GMPEs).

This example demonstrates how the ANN works and shows its potential.